DE112018001304T5 - ADAPTIVE TORQUE REDUCTION BY A MICRO-HYBRID SYSTEM - Google Patents

Abstract

Various methods and arrangements for reducing noise, vibration, and roughness (NVH) in a Skip Fire engine control system are described. In one aspect, a firing order is used to operate the engine in a dynamic firing level modulation manner. A compensating torque is determined by adaptive control, which is applied to a drive train by an energy storage / release device. The compensation torque is designed to at least partially cancel out a torque variation generated by the ignition sequence. Various methods, powertrain controls, arrangements, and computer software associated with the above operations are also described.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

The present application also claims priority from the preliminary U.S. Patent Application No. 62 / 470,487 , filed on March 13, 2017, which is hereby incorporated by reference in its entirety.

FIELD OF THE INVENTION

The present invention relates to a dynamic ignition level control system for an internal combustion engine. In particular, the present invention relates to arrangements and methods for reducing NVH (Noise, Vibration and Harshness) using adaptive control to generate a balancing torque that is applied to the drive train.

BACKGROUND

Most vehicles that are currently in operation are powered by internal combustion engines (VM). Internal combustion engines usually have several cylinders or other working chambers in which the combustion takes place. The power generated by the engine is dependent on the amount of fuel and air supplied to each working chamber and the engine speed.

The combustion process and the ignition of cylinders can introduce undesirable NVH. For example, the engine can transmit vibrations to the body of the vehicle, where they can be perceived by vehicle occupants. Noise can also be transmitted through the chassis into the passenger compartment of the vehicle. Under certain operating conditions, the ignition of cylinders creates undesirable acoustic effects through the exhaust system and exhaust tailpipe. Vehicle occupants can thus be exposed to undesirable NVH from vibrations transmitted through the structure or noise transmitted through the air. Thus, efforts are continually being made to reduce the amount of NVH generated by internal combustion engines.

SUMMARY OF THE INVENTION

Various methods and arrangements for reducing noise, vibration, and roughness (NVH) in a hybrid powertrain are described.

In one aspect, an electric motor / generator controller for operating an electric motor / generator used as an auxiliary torque source / sink in a hybrid powertrain that includes an internal combustion engine operating in an ignition sequence is described. The electric motor / generator control comprises an adapted basic function control module, which is set up to generate a control signal for the electric motor / generator on the basis of various input signals. Input signals for the matched basic function control module include a parameter indicating an ignition sequence length, a parameter indicating an ignition sequence phase, a crank angle signal and an NVH metric signal. The matched basis function control module may include a matched basis function generator that generates at least one sine and cosine pair. The NVH metric signal may be based on a crankshaft rotation signal or a time derivative thereof. The matched basis function control module includes a matched basis function coefficient module that can use a recursive least-squares algorithm or at least one least-squares algorithm to determine coefficients that weight the at least one sine and cosine pair.

In another aspect, a method of operating an electric motor / generator used as an auxiliary torque source / sink in a hybrid powertrain that includes an internal combustion engine that operates in an ignition sequence is described. The method includes generating a control signal for the electric motor / generator using a customized basis function control algorithm to provide an acceptable NVH metric and to minimize fuel consumption. The adjusted basis function control algorithm uses a sine and cosine signal that is adapted to the periodicity of an ignition sequence of an internal combustion engine that is part of the hybrid drive train. The NVH metric signal may be based on a crankshaft rotation signal or a time derivative thereof and may depend on the transmission gear.

In yet another aspect, powertrain control for operating an internal combustion engine in a manner according to dynamic ignition level modulation using an operational effective ignition component that is fuel efficient and has acceptable noise, vibration, and roughness (NVH) characteristics is described. The powertrain controller includes an ignition fraction calculation device configured to generate an operational effective ignition fraction to provide a requested engine torque and an ignition timing determination module configured to generate an ignition sequence used to operate the engine in a dynamic ignition level modulation manner. The firing order is based on the effective ignition component that can be used. An adapted basic function control module is configured to determine a compensation torque that is applied to a drive train by an energy storage / collection / release device, the compensation torque being designed to at least partially cancel out a torque variation generated by the Skip Fire firing sequence , reducing the NVH that would otherwise be generated by the Skip Fire firing order.

In yet another aspect, a powertrain controller for operating an internal combustion engine having multiple working chambers, so that a working chamber can be fired at a low torque or high torque ignition opportunity, which is fuel efficient and has acceptable noise, vibration and roughness (NVH) characteristics. Properties). The powertrain control includes an ignition fraction calculation device configured to generate a usable effective ignition fraction to provide a requested engine torque. An ignition timing determination module is set up to generate an ignition sequence used to operate the internal combustion engine. The firing order is based on the effective ignition component that can be used. The powertrain controller is an adaptive controller configured to determine a balance torque applied to a powertrain by an energy storage / catch / release device, the balance torque being designed to at least accommodate a torque variation generated by the Skip Fire firing order partially overrule, reducing the NVH that would otherwise be generated by the firing order.

In yet another aspect, a method of operating an internal combustion engine having multiple working chambers is described, wherein one of the multiple working chambers can be ignited at a low torque or high torque ignition opportunity that is fuel efficient and has acceptable noise, vibration, and roughness characteristics ( NVH properties). The method includes generating an operational effective ignition component to provide a requested engine torque and generating an ignition sequence based on the operational effective ignition component. A compensating torque that is applied to a drive train by an energy storage / collection / release device is determined by an adaptive control. The balancing torque is designed to at least partially cancel out torque variations generated by the firing order, thereby reducing the NVH that would otherwise be generated by the firing order. The adaptive control can be an adaptive filter feedforward control or an adapted basic function control.

In yet another aspect, a computer readable storage medium is described that includes executable computer code that is implemented in a tangible form and is suitable for operating an internal combustion engine so that a working chamber is fired at a low torque or high torque ignition opportunity in one way that has acceptable noise, vibration and roughness (NVH) properties. The computer readable medium includes executable computer code to generate a usable effective ignition component to deliver a requested torque and executable computer code to generate an ignition sequence used to operate the engine. The firing order is based on the operation of the effective firing fraction. The executable computer code determines a compensation torque that is applied to a powertrain by an energy storage / capture / release device using adaptive control. The balancing torque is designed to at least partially cancel out torque variations generated by the firing order, thereby reducing the NVH that would otherwise be generated by the firing order.

list of figures

• 1 ein Blockdiagramm einer Antriebsstrangsteuerung in einem Skip Fire-Motorsteuerungssystem gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 2 ein Ablaufdiagramm ist, das ein Verfahren zum Reduzieren von Geräuschen, Vibrationen und Rauigkeit (NVH) in einem Skip Fire-Motorsteuerungssystem gemäß einer Ausführungsform der vorliegenden Erfindung diagrammatisch darstellt.
• 3 ein Schaubild eines Antriebsstrangs und einer Antriebsstrangsteuerung gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 4 und 5 Diagramme sind, die beispielhafte Methoden zur Optimierung der Reduzierung von NVH gemäß einer Ausführungsform der vorliegenden Erfindung darstellen.
• 6 ein Graph des verbrauchten Kraftstoffs als Funktion des Zündungsanteils gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 7 ein Diagramm einer Motordrehmomentwellenform gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 8 ein beispielhaftes Diagramm einer ersten Harmonischen ist, die der in 7 dargestellten Motordrehmomentwellenform überlagert ist.
• 9 ein beispielhaftes Diagramm einer ersten Harmonischen ist.
• 10 ein beispielhaftes Diagramm einer die erste und die zweite Harmonische umfassenden Wellenform ist, die der in 7 dargestellten Motordrehmomentwellenform überlagert ist.
• 11 ein beispielhaftes Diagramm einer zweiten Harmonischen ist.
• 12 ein Blockdiagramm einer Antriebsstrangsteuerung in einem Skip Fire-Motorsteuerungssystem gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 13 ein Blockdiagramm eines Adaptivfiltervorwärtsregelungssystems gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 14 eine beispielhafte Wellenform ist, die Zylinderaktivierungen und Zylinderdeaktivierungen zugeordnete Drehmomentsignaturen gemäß einer Ausführungsform der vorliegenden Erfindung zeigt.
• 15 ein Blockdiagram ist, das Anteile einer Antriebsstrangsteuerung mit einem Zündzeitpunktbestimmungsmodul und einer Adaptivfiltervorwärtsregelung gemäß einer Ausführungsform der vorliegenden Erfindung darstellt.
• 16 ein Ablaufdiagramm eines Verfahrens für Adaptivfiltervorwärtsregelung gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 17A ein Blockdiagramm eines Filtermoduls gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 17B ein schematisches Diagramm eines Anteils eines Hybrid-Antriebsstrangs unter Verwendung einer Angepasste-Basisfunktion-Regelung gemäß einer Ausführungsform der vorliegende Erfindung ist.
• 17C und 17D beispielhafte Bode-Plots der Größe und Phase des Fahrzeugantwortfilters und des NVH-Antwortfilters sind.
• 17E ein Blockdiagramm, das ein Winkelerzeugungsmodul darstellt, gemäß einer Ausführungsform der vorliegenden Erfindung ist.
• 18A ein Beispiel für einen beispielhaften Übergang des Zündungsanteils darstellt.
• 18B ein momentanes und ein mittleres Motordrehmoment vor, während und nach einem beispielhaften Zündungsanteilübergang zeigt.
• 18C eine Motor/Generator-Befehlskurve und ein momentanes Motor/Generator-Drehmoment zeigt, das von einem Antriebsstrang vor, während und nach einem beispielhaften Zündungsanteilübergang geliefert oder durch diesen abgezogen wird.
• 18D das Antriebsstranggesamtdrehmoment und das mittlere Antriebsstrangdrehmoment vor, während und nach einem beispielhaften Zündungsanteilübergang zeigt.
• 19 ein Beispiel einer Simulation einer Ausführungsform der vorliegenden Erfindung zeigt.

The invention and its advantages are best understood with reference to the following description in conjunction with the accompanying drawings, in which:

• 1 10 is a block diagram of powertrain control in a Skip Fire engine control system according to an embodiment of the present invention.
• 2 FIG. 10 is a flowchart diagramatically illustrating a method for reducing noise, vibration, and roughness (NVH) in a Skip Fire engine control system according to an embodiment of the present invention.
• 3 14 is a diagram of a powertrain and powertrain control according to an embodiment of the present invention.
• 4 and 5 14 are diagrams illustrating exemplary methods for optimizing NVH reduction, in accordance with an embodiment of the present invention.
• 6 FIG. 12 is a graph of fuel used as a function of ignition percentage according to an embodiment of the present invention.
• 7 4 is a diagram of an engine torque waveform according to an embodiment of the present invention.
• 8th is an exemplary diagram of a first harmonic that the in 7 engine torque waveform shown is superimposed.
• 9 is an exemplary diagram of a first harmonic.
• 10 FIG. 3 is an exemplary diagram of a waveform comprising the first and second harmonics, which is shown in FIG 7 engine torque waveform shown is superimposed.
• 11 is an exemplary second harmonic diagram.
• 12 10 is a block diagram of powertrain control in a Skip Fire engine control system according to an embodiment of the present invention.
• 13 Figure 3 is a block diagram of an adaptive filter feedforward control system according to an embodiment of the present invention.
• 14 FIG. 12 is an exemplary waveform showing torque signatures associated with cylinder activations and cylinder deactivations, according to an embodiment of the present invention.
• 15 FIG. 12 is a block diagram illustrating portions of a powertrain control with an ignition timing determination module and an adaptive filter feedforward control according to an embodiment of the present invention.
• 16 10 is a flow diagram of a method for adaptive filter feedforward control according to an embodiment of the present invention.
• 17A 4 is a block diagram of a filter module according to an embodiment of the present invention.
• 17B FIG. 10 is a schematic diagram of a portion of a hybrid powertrain using a matched basis function controller according to an embodiment of the present invention.
• 17C and 17D are exemplary Bode plots of the size and phase of the vehicle response filter and the NVH response filter.
• 17E 10 is a block diagram illustrating an angle generation module according to an embodiment of the present invention.
• 18A represents an example of an exemplary transition of the ignition component.
• 18B shows an instantaneous and an average engine torque before, during and after an exemplary ignition share transition.
• 18C 12 shows an engine / generator command curve and an instantaneous engine / generator torque delivered by or drawn from a powertrain before, during, and after an exemplary ignition share transition.
• 18D shows the powertrain total torque and the average powertrain torque before, during and after an exemplary ignition share transition.
• 19 shows an example of a simulation of an embodiment of the present invention.

In the drawings, reference numerals are sometimes used to designate the same structural elements. It is also understood that the representations in the figures are schematic and not to scale.

DETAILED DESCRIPTION

The present invention relates to a Skip Fire engine control system. In particular, the present invention includes methods and arrangements for using compensating torque to reduce noise, vibration, and roughness (NVH) in a Skip Fire engine control system.

The Skip Fire engine controller takes into account the selective deactivation of the ignition of certain cylinders during selected ignition opportunities. Thus, for example, a particular cylinder can be activated during one ignition opportunity and can then be deactivated during the next ignition opportunity and then selectively deactivated or activated during the next. This is in contrast to the conventional operation of a variable displacement engine in which a fixed set of cylinders are deactivated during certain low load operating conditions.

The Skip Fire engine control offers several advantages, including significant improvements in fuel economy. However, a challenge with Skip Fire engine control is noise, vibration, and roughness. In particular, there are certain firing sequences, or firing fractions, that produce significant amounts of NVH. NVHs of this type are undesirable because they can be felt by vehicle occupants.

One approach to addressing these problems is to not use certain firing proportions or firing sequences that are known to produce unacceptable levels of NVH. Instead, other firing proportions or firing sequences are used and the cylinder delivery is adjusted accordingly (e.g. by adjusting the intake vacuum, the ignition advance, etc.), so that the target cylinder delivery is provided. Different approaches of this kind are transmitted simultaneously U.S. Patent Application No. 13 / 654,244 described, which is included here for all purposes in its entirety.

This application describes another approach to dealing with NVH. Various implementations include generating a balancing torque that is applied to a drive train of a vehicle. The balancing torque is any torque that is applied to help cancel or reduce a torque variation generated by an engine. The compensation torque can be generated by any suitable energy storage / collection / release device. An example would be an electric motor / generator with a battery and / or a capacitor for storing and releasing energy. Alternatively, any system or device that stores the energy and mechanically, pneumatically, or hydraulically collects / releases it can be used. For example, a flywheel with a variable mechanical coupling or a high pressure fluid container with valves that control fluid flow to and from a turbine, or a similar device for collecting / releasing energy from a powertrain can be used. The compensation torque is applied in such a way that the noises and vibrations generated by the Skip Fire firing sequence are at least partially reduced or eliminated.

The balancing torque system above is used selectively in various approaches. This means that many firing proportions and firing sequences that provide a target torque produce acceptable NVH levels, and therefore the balancing torque need not be applied under these circumstances. In other circumstances, an appropriate firing rate or firing order may produce undesirable NVH levels, but the energy consumption or equivalent energy costs involved in weakening the NVH may be too high. A different ignition component or a different ignition sequence is thus used to provide the target engine output. In yet other circumstances, the use of balancing torque may allow the use of firing proportions or firing sequences that would otherwise have been unacceptable due to their associated NVH dimensions and may result in energy savings rather than losses. In various embodiments, the balancing torque system is designed to analyze the energy cost of the options available and to choose the most fuel efficient approach that also brings the NVH to an acceptable level.

First, referring to 1 a powertrain control 100 according to a particular embodiment of the present invention. The powertrain control 100 includes an ignition component calculation device 112 , an ignition timing determination module 120 , an NVH reduction module 121 , a powertrain parameter adjustment module 116 , a sensor unit 122 and an ignition control unit 140 , The ignition share calculation device 112 , the ignition timing determination module 120 and the NVH reduction module 121 coordinate their operations to determine a suitable operational ignition component and a skip fire firing order for the engine. Depending on the part of the ignition that can be used, you can also determine a suitable compensation torque to reduce the NVH generated by the ignition sequence.

The ignition share calculation device 112 receives a torque request signal 111 based on the current accelerator pedal position (APP), engine speed and other inputs. The torque request signal can be from the APP 163 through an optional preprocessor 105 be passed before the ignition component calculation device 112 reached. The torque request signal 111 indicating a target engine charge may be from an accelerator pedal position sensor or other suitable sources, such as e.g. B. a vehicle speed controller, a torque calculation device, an ECU, etc. are related. An optional preprocessor 105 can the accelerator pedal signal before the supply to the ignition component calculation device 112 modify. However, it is understood that in other implementations, the accelerator pedal position sensor is directly connected to the ignition component calculation device 112 can communicate.

Based on the torque request signal 111 determines the ignition proportion calculation device 112 a skip fire firing portion that would be suitable for providing the target torque for selected engine operations and that has acceptable NVH characteristics (with or without the use of a balancing torque). Any share of ignition 112 indicates the percentage or percentage of ignition under the current (or commanded) operating conditions required to provide the target tax. In some preferred embodiments, the firing fraction may be determined based on the percentage of optimized firings required to provide the engine torque requested by the driver (e.g., when the cylinders fire at an operating point that is substantially fuel efficiency optimized). However, in other cases, non-dimensional ignitions, ignitions that are optimized for factors other than fuel efficiency, current engine settings, etc. can be used in determining the firing rate. In certain embodiments, the ignition component is selected from a set or a collection of predetermined ignition components.

The ignition component determination process can take into account various factors, including NVH, fuel efficiency, and target torque. For example, in some situations there is a certain amount of ignition that provides a target torque in the most fuel efficient manner at the current engine speed (e.g., using optimized ignitions). If this firing fraction is available for use by the firing fraction calculator and is also attributed to acceptable NVH levels, the ignition fraction calculator chooses 112 this ignition component and transmits it to the ignition timing determination module 120 , so that a suitable operational ignition sequence can be generated based on the ignition component. The ignition share calculation device 112 shows the NVH reduction module 121 also assumes that no NVH attenuation is required, and thus the energy storage / release device 124 no compensating torque to the driveline during this sequence to operate the engine 150 is used.

In spite of everything, if the above ignition component is known to produce unacceptable NVH dimensions, the ignition component calculation device may select this ignition component as the operational ignition component. Such a choice is based on a determination that the NVH can be reduced to acceptable levels by applying an appropriate balancing torque to the powertrain. The selection is also based on the determination that there are no other alternatives of more fuel efficient ignition components, even if the energy costs associated with the NVH weakening are taken into account. In this case, the ignition share calculation device transmits 112 the selected operational ignition component to the ignition timing determination module 120 , so that a suitable operational ignition sequence can be generated based on the ignition component. The firing rate calculator also indicates to the NVH reduction module that NVH attenuation is required. As a result, the NVH reduction module operates the energy storage / release device 124 to apply an appropriate amount of balance torque to the driveline to reduce the NVH generated by the firing order.

In still further circumstances, the ignition proportion calculator 112 choose an operational ignition component that is less optimal for providing the target torque, that is, an ignition component that is more suitable for providing a torque level that is different from the target torque, but which has acceptable NVH characteristics. Thus, the cylinder delivery must be set (e.g. by setting the MAP, advance ignition timing and other engine parameters) so that the target torque is provided. The operational ignition component is nonetheless more fuel efficient than the other alternatives, which may include an ignition component with poor NVH characteristics where NVH attenuation is not possible or leads to excessive energy consumption.

The ignition share calculation device 112 is designed to store and / or access data to assist in the implementation of the above provisions and energy efficiency comparisons. Any suitable data structure or algorithm can be used to perform the determinations. In some embodiments, the firing rate calculator uses 112 for example, a lookup table to determine a suitable operational ignition component and to determine whether balancing torque should be applied. In still further embodiments, the ignition component calculation device makes such determinations by dynamically calculating and comparing the various energy components associated with the different ignition components and / or ignition sequences in question. Some of these approaches are described in more detail during the registration process.

After choosing a suitable operational ignition component, the ignition component calculation device transmits 112 the ignition share 119 to the ignition timing determination module 120 , The ignition timing determination module 120 is designed to be a sequence of firing commands (e.g. control pulse signal 113 ) that cause the engine 150 the percentage of ignitions provided by a controlled ignition share 119 is specified. In some implementations, the spark timing module generates 120 For example, a bit stream where every 0 indicates a deactivation and every 1 an activation for the current cylinder ignition opportunity.

The ignition timing determination module 120 can generate the firing order in many different ways. For example, sigma-delta converters work well as the ignition timing determination module 120 , In still further embodiments, the ignition timing determination module selects an appropriate firing order from a firing order collection based on the received firing fraction.

If it has been determined that attenuation of NVH generated by the firing order is not required, the series of firing commands (which are sometimes used as a control pulse signal 113 is designated) by the ignition timing determination module 120 is output to the ignition control unit 140 are forwarded, which triggers and controls the actual cylinder firings. While executing the firing order in the engine 150 becomes no compensation torque by the energy storage / release device 124 applied to the drive train.

On the other hand, if it is determined that the firing order requires weakening, the firing proportion calculator transmits 112 and / or the ignition timing determination module 120 the ignition component and / or the ignition sequence to the NVH reduction module 121 before using the firing order to operate the engine. The NVH reduction module 121 is designed to determine, based on these inputs, an appropriate compensation torque that would bring the NVH generated by the firing order to acceptable levels. In various embodiments, the compensating torque is in the form of one or more substantially sinusoidal torque waveforms applied to the powertrain that counteract the particular torque variations generated by the Skip Fire firing order. The reason for the sinusoidal torque waveform can be based on the crankshaft angle of the engine. This means that the compensation torque as compensation torque = sin (f * θ + ϕ), where θ is the crankshaft angle, ϕ is the phase and f = 4 / N (for a V8 engine), where N is the denominator of the ignition component level, can be expressed.

In various embodiments, the balancing torque waveform is provided by the energy storage / release device 124 applied to the drive train. The balancing torque includes sequentially adding torque to the drive train and then withdrawing torque from the drive train. The timing, strength, and pattern of the balancing torque can be based on various factors, including the Skip Fire firing order, the engine speed, and the like Battery level or the level of charge with an alternative storage device, ie capacitor, and the current number of cylinders.

It is understood that various implementations of balancing torque, as described herein, differ from some prior art systems that use an electric motor to replace missing torque pulses from an exposed cylinder. Such a system requires the provision of high bandwidth and amplitude torque pulses from the energy storage / release device 124 , Various implementations of the present invention do not explicitly aim to fill a torque hole created by a missed ignition opportunity. Instead, such implementations take into account the entire torque signature that is generated by a specific ignition component or a specific ignition sequence. In these implementations, the control electronics strive to counteract the torque variation associated with one or more harmonics of the torque signature. This type of control algorithm advantageously requires an energy storage / release device 124 lower bandwidth because the controller does not attempt to override or imitate the high bandwidth components of the torque peak associated with an igniting cylinder. Likewise, the energy storage / release device 124 Provide lower amplitude compensation torque because the system does not attempt to mimic a torque spike associated with a firing cylinder. Both the lower bandwidth and the lower amplitude result in that of the energy storage / release device 124 assigned hardware is cheaper and easier to implement. Another advantage of this method is that low frequency harmonics are felt more by vehicle occupants, thereby maximizing the reduction in sensed vibrations for a given amount of balancing torque.

Any suitable algorithm or method can be used to generate the balancing torque waveform. In some embodiments, the NVH reduction module receives 121 for example an ignition sequence from the ignition timing determination module 120 that later used to operate the engine 150 is used. The NVH reduction module 121 determines engine torque variations that would be generated by the firing order. The variations in engine torque are responsible for the generation of NVH. The balancing torque includes variations that counter one or more types of engine torque variations or help to cancel them.

The balancing torque properties can vary widely depending on the requirements of a particular application. Consider an exemplary process, similar to that described above, in which the firing rate calculator 112 selects an operational firing component that is known to produce unacceptable levels of NVH. Thus, the NVH must be weakened. The ignition timing determination accordingly generates 120 a skip fire firing order based on the operational firing fraction used for analysis to the NVH reduction module 121 is transmitted.

The NVH reduction module 121 determines an expected engine torque based on the Skip Fire firing order, engine speed, firing rate, and / or any other suitable parameter. In some implementations, this motor torque is understood to include a fixed component (ie, a DC term) and a variable component that can be represented by sine functions of multiple harmonics, including a first harmonic (fundamental frequency) and other harmonics. The specified DC term drives the vehicle, and the harmonics are the inevitable result of the torque variation generated by an internal combustion engine as its cylinders move through the various strokes of a combustion cycle. These sine functions of harmonics or engine torque variations are considered the cause of the NVH. The NVH reduction module 121 determines a balance torque that is applied to the powertrain using a particular pattern or sequence. In various embodiments, the characteristics of the pattern or sequence of the compensating torque (e.g. frequency, strength and phase) are designed to at least partially counteract a selected set of one or more of the sine functions of harmonics.

An example of this concept is given in 7-11 shown. 7 FIG. 12 is a graph showing engine torque (N * m) applied to the crankshaft / drive train as a function of engine angle. That means the graph is a waveform 702 displays that represents engine torque generated by an exemplary Skip Fire firing order. In this example, the average torque is approximately 87 N * m. This average torque is the specified component of the motor torque, ie the DC Term. Using the methods of Fourier analysis, the engine torque waveform 702 are expressed as the sum of this fixed component and different harmonics each with a fixed amplitude (ie several harmonics, including a first harmonic, a second harmonic ... tenth harmonic, etc.).

In 8th becomes an exemplary waveform 802 for DC plus first harmonics that are above the motor torque waveform 702 is shown. The offset waveform 802 The first harmonic has a frequency that is at the fundamental frequency of the waveform 702 matches. How out 8th you can see that the offset waveform is correct 802 the first harmonic with a substantial portion of the engine torque waveform 702 match. The first harmonic component can be obtained by subtracting the average torque (e.g. DC offset of 87 N * m) from the waveform 802 be isolated. This leads to the waveform 902 the first harmonic of 9 , In various embodiments, a balancing torque waveform is generated around the waveform 902 counteract, ie, subtract torque from the powertrain when the waveform 902 Adding torque to the powertrain and adding torque to the powertrain when the waveform 902 Subtracts torque from the drive train.

Various approaches include a balancing torque that has properties (e.g., frequency) that are generally identical or substantially similar to those of a selected set of one or more of the sine functions of harmonics, except that the amplitude can be different and the phase is shifted (e.g. 180 °) so that the compensation torque reduces or cancels the selected or selected sine function (s) of the harmonics. In some embodiments, the compensation torque is designed to only counteract, base on, and / or have the same frequency as the first harmonic. This means that, in various embodiments, the compensating torque is not based on the other harmonics in the expected engine torque, has a different frequency than this and / or does not counteract it. The inventors have found that in various applications, only one or a few harmonic sine functions need to be canceled or reduced in order to bring the NVH to an acceptable level. In the in 7-9 Example shown, the compensation torque can simply be configured to the waveform 902 cancel or reduce the first harmonic to an acceptable level. The compensation torque can thus have the same frequency and amplitude as the waveform 902 of the first harmonic, but can simply be out of phase by 180 degrees. In still further embodiments, the compensation torque takes into account and counteracts multiple harmonics (e.g., the first harmonic and one or more other harmonics, etc.).

In other embodiments, a DC term can be added to the compensation torque. If the DC term is sufficiently large, the compensation torque will be uniform in one direction; this can eliminate or reduce the impact of nonlinear behavior (e.g., dead zone, play, etc.) that occurs when an energy storage / release device exceeds a net zero delivery torque value. The DC term can be in either direction, i.e. H. the energy storage / release device may store energy from the drive train or release energy to the drive train. The DC term can be zero. The strength and sign of the DC can depend on a number of factors, including battery or capacitor charge, torque requirement, or other operational characteristics.

10 represents a constant term and two harmonics (ie the first and second harmonics) represented by the waveform 1002 , from the exemplary engine torque waveform 702 separated and plotted over this. As by comparing 8th and 10 it can be seen that the two harmonics are even better in combination with the overall variation in the engine torque waveform 702 match than only with the offset first harmonic in 8th , 11 represents a waveform 1102 of the second harmonic, which is the second harmonic after the removal of the DC term and the first harmonic term. As by comparing 9 and 11 can be seen is the amplitude of the waveform 902 of the first harmonic is significantly larger than the amplitude of the waveform 1102 the second harmonic. That means the engine torque waveform 702 has a component of the first harmonic that is greater than the component of the second harmonic. The larger component of the first harmonic generally produces more undesirable NVH and thus control algorithms can focus on canceling or reducing this harmonic component. In various applications, the compensation torque is designed to be only the first harmonic (e.g. waveform 902 of 9 ) and not to counteract any other harmonics. It was found that this caused the calculation and Implementation of the balancing torque is facilitated and it is still sufficient to bring the NVH to acceptable levels. In yet other embodiments, the compensation torque is designed to be multiple harmonics (e.g., a combined waveform that represents the waveforms 902 and 1102 includes) to repeal or to counteract this. The cancellation of higher harmonics can be advantageous for the reduction of acoustic noises which result from the vibrations caused. For example, the first and any harmonics other than the second harmonic can be substantially canceled. In particular, harmonics that are close to the passenger compartment droning frequency can advantageously be substantially eliminated or reduced.

The magnitude (e.g. amplitude) of the compensation torque can vary depending on different conditions and applications. For example, in various embodiments, the magnitude of the balancing torque may be significantly less than the magnitude of the sine function (s) of the harmonics generated by the motor that it counteracts. In these embodiments, the level of balancing torque is designed to reduce and not eliminate the engine NVH, and to bring the NVH below a predetermined level that has been determined to be acceptable to vehicle occupants. What defines this predetermined NVH level can vary between different engine and vehicle designs. In various embodiments, user tests are performed to determine the acceptable NVH level. In addition, this predetermined level of acceptable NVH may also be based on various conditions such as: B. the accelerator pedal position, the rate of change in the accelerator pedal position, the road conditions, the gear engaged, the vehicle speed, the passenger compartment noise level, the presence of engine idling and any other suitable parameter, can be set dynamically. Such conditions can be detected by one or more suitable sensors.

In some implementations, the compensation torque is also based on the sensor unit 122 received feedback set. The sensor unit 122 includes one or more sensors that can detect various engine parameters, including crankshaft speed / acceleration, accelerometer data, vibrations, etc. For example, accelerometers can be positioned on a seat running rail, adjacent to it and / or in an ECU in order to detect vibrations felt by vehicle occupants. Based on that from the sensor unit 122 received feedback, the compensation torque is set dynamically. For example, the timing (phase) and magnitude of the compensation torque sequence can be changed based on the sensor input. It is understood that the above sensor feedback is not required and that in various embodiments the compensation torque generation system is a feed forward system.

As soon as the NVH reduction module prepares a suitable balancing torque, the NVH reduction module operates the energy storage / release device 124 to apply the balance torque to the powertrain. The application of the compensation torque is carried out with the execution of the corresponding firing sequence on the engine 120 coordinated. As a result, the compensation torque counteracts certain variations in the engine torque and the NVH generated by the Skip Fire firing sequence is reduced.

In the illustrated embodiment, an optional powertrain parameter setting module 116 provided that with the ignition component calculation device 112 interacts. The powertrain parameter adjustment module 116 directs the ignition control unit 140 to adjust selected powertrain parameters accordingly to ensure that the actual engine output corresponds essentially to the requested engine output for the triggered ignition component. For example, the powertrain parameter setting module 116 be responsible for determining the mass air charge (MAC), ignition timing control, and valve timing and / or other engine settings that are desirable to help ensure that the actual engine charge matches the desired engine charge. Of course, in other embodiments, the powertrain parameter adjustment module may be configured to directly control various engine settings.

It goes without saying that the valve control can be implemented in a variety of ways. In some embodiments, valve timing can be controlled by a cam. Some embodiments use a single cam with a fixed stroke plan, whereas in other embodiments the valve stroke can be selected by setting between different cams with different curve profiles. The valve lift can also be done by adjusting the mechanical relationship be changed between a single cam curve and its associated valve. Valve deactivation can be accomplished using a lost motion jack, a lift-free cam, or a change in the mechanical relationship between a cam and an associated valve. In some embodiments, a cylinder may have more than one intake valve and / or more than one exhaust valve, which may be actuated by either a single cam or many, may have separate cams with different profiles and timing. In some embodiments, electromagnetic valve actuation can be used, which gives almost complete flexibility in implementing various stroke and timing arrangements.

The ignition share calculation device 112 , the ignition timing determination module 120 , the NVH reduction module 121 , the powertrain parameter adjustment module 116 , the sensor unit 122 and the other components shown by 1 can be in a variety of forms and their functions can alternatively be integrated into an ECU or provided by other more integrated components, by groups of sub-components or by using a variety of alternative approaches. In various alternative implementations, these functional blocks can be achieved algorithmically using a microprocessor, the ECU, or other computing device, using analog or digital components, using programmable logic, using combinations of the foregoing, and / or in any other suitable manner ,

Although not required in all implementations, in some implementations the determination of an appropriate firing fraction and / or the balancing torque (ie a determination of whether balancing torque is used and what the balancing torque will be) can be made from ignition opportunity to ignition opportunity. This means that the currently desired ignition component and / or the currently desired compensation torque can be determined again before each ignition opportunity based on the accelerator pedal position or other operating parameters. This allows control 100 specifically respond to changing requirements (e.g., changing intake manifold pressure or other engine parameters) while maintaining the benefits of skip fire operation. In other implementations, the torque generated during the change in firing proportions can be predicted, and a control system based on adaptive filters or MPC (Model Predictive Control) can be used to improve NVH.

An example in which control from one ignition opportunity to another is advantageous is when the desired ignition ratio is changed. A special example is the change in the ignition component from ½ to 1. In this example, the MAP has to be reduced in order to generate the correct torque level, but this is done slowly, ie the MAP has a limited ability to change from ignition opportunity to ignition opportunity. A solution to this problem from the prior art, the z. B. in the U.S. Patent Application No. 13 / 654,244 is to set the ignition portion at a relatively low speed to match the expected change in MAP. By continually recalculating and providing adequate compensation torque, the NVH reduction module can remove excessive torque resulting from a MAP that is too high, allowing a faster transition.

In some embodiments, the compensation torque may be determined using a previously calculated future firing order with a short term calculation of the optimal control. This control method is particularly useful when the firing order is not repeated, such as. B. during a transition between ignition component levels. This can refer to the ignition decisions that have been made but have not yet been implemented. This can range from 4 to 20 ignition opportunities, but could be more or less. Since these decisions are known before they are implemented, the compensation torque can be calculated in advance. The balancing torque can include both negative and positive torques to achieve an optimal NVH fuel economy compromise depending on engine / generator and energy storage device requirements. Motor / generator requirements can include maximum permissible torque and power levels. Energy storage requirements may include current energy storage utilization as well as maximum power transfer from the energy storage device.

Next, referring to 2 a procedure 200 for determining a compensation torque according to a specific embodiment of the present invention. First, step 202 received an engine torque request. Determined in different implementations the ignition proportion calculator 112 a requested or desired engine torque based on the accelerator pedal position, engine speed, a cruise control setting, and any other suitable engine parameters.

The steps 203 . 204 . 206 and 208 relate to a process for evaluating various candidate ignition components to select an operational ignition component that provides the target torque and has acceptable NVH characteristics, either with or without any attenuation. In some embodiments, the powertrain controller performs these steps as needed when an operational ignition component needs to be selected. In other embodiments, however, the evaluation of various candidate firing fractions is instead integrated into an algorithm, a look-up table, or any other suitable decision-making mechanism. This means that the powertrain controller can consult a table or other mechanism that directly generates the operational ignition component based on different inputs instead of a hasty dynamic comparison of different ignition components in question. In this case, the procedure goes directly to step 210 about.

With reference back to step 203 of 2 determines the ignition proportion calculation device 112 after a torque target height has been reached, whether an available ignition component with acceptable NVH properties is the target torque when operating with an optimal cylinder load, e.g. B. the cylinder load, which maximizes fuel economy. In some embodiments, the firing fraction calculation device stores 112 for example, data indicating a set of such firing proportions that are known to have acceptable NVH characteristics when operating under optimal cylinder load under certain operating conditions. It will be appreciated that it is a function of engine speed and transmission that ignition levels produce acceptable NVH, as in the appended ones U.S. Patent Applications No. 13 / 654,244 and 13 / 963.686 to which reference is made in its entirety. If one of these ignition components can provide the target torque, the method goes to step 212 above and this ignition component becomes the operational ignition component.

If the ignition proportion calculator 112 Determining that there is no ignition component with acceptable NVH characteristics that can provide the target torque at optimal cylinder load, the method goes to step 204 about. At step 204 the ignition component calculation device obtains a set of candidate ignition components. The set of ignition components can include two types of ignition components. One type includes one or more candidate firing fractions with acceptable NVH characteristics that provide the target torque, but only when the cylinder output is set to a non-optimal load, as in the U.S. Patent Application No. 13 / 654,244 is discussed, to which reference is made here in its entirety. For the purposes of the present application, such an ignition component is referred to as a “low NVH ignition component”. The other type of ignition components include one or more candidate ignition components that can provide the desired torque with less or minimum cylinder load setting, but the NVH associated with such ignition components can be unacceptable without attenuation. For the purposes of the present application, such an ignition component is referred to as a "high NVH ignition component".

At step 206 the energy costs associated with the attenuation of NVH for the high NVH ignition share (s) are determined. This can be done in a number of ways. An example approach is described below.

wobei Tq(t) das Drehmoment als Funktion der Zeit ist, a₀ das durchschnittliche Drehmoment ist (DC-Term), an die der n-ten harmonischen Komponente zugeordnete Amplitude ist, T der Zeitraum der ersten Harmonischen ist (Grundfrequenz) und φ_n die Phase der n-ten harmonischen Komponente ist.In this example, the ignition timing determination module generates 120 an eligible Skip Fire firing order based on the eligible high NVH firing rate. The torque generated by the Skip Fire firing order and firing fraction can be represented as a periodic waveform in a model. This waveform can again be represented as a Fourier series: $${}_{\text{Tq}\left(\text{t}\right)={\text{a}}_0+}{\displaystyle \underset{n\text{= 1}}{\overset{\infty }{\Sigma }}\left(a_{\mathrm{<figure-callout\; id="2"\; label="n\; cos\; n"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">n\; </figure-callout>}}\text{cos}\frac{n}{}\frac{2\pi t}{T}+{\phi }_n\right)}$$

where Tq (t) is the torque as a function of time, a _{0 is} the average torque (DC term) to which the amplitude is assigned to the nth harmonic component, T is the period of the first harmonic (fundamental frequency) and φ _n is the phase of the nth harmonic component.

Human perception of NVH varies with frequency. As a rule, lower frequencies, below approximately 8 Hz, are perceived as more disturbing than higher frequency vibrations. The relative contribution of each harmonic component to the NVH can be defined by a weighting factor w _n . If w _{n is} the weighting of the nth harmonic, the total NVH can be determined by finding the root mean square of the product of the weighting functions and the strength of the different harmonic frequencies: $${}_{\text{NVH}=}\sqrt{\frac{1}{2}{\displaystyle \underset{n=1}{\overset{\infty }{\Sigma }}{w}_n^2{a}_{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">n</figure-callout>}}^2}}$$

If an energy storage / release device 124 is included in the powertrain, equation 2 must be modified to include one by the energy storage / release device 124 compensating torque applied to the drive train. The compensation torque can be expressed by a Fourier expansion similar to equation 1, the nth harmonic component having a strength e _n . Equation 3 represents the NVH, including the effect of the balancing torque, assuming that the phase of each harmonic term of the balancing torque is shifted 180 degrees from the engine torque: $${}_{\text{NVH}=}\sqrt{\frac{1}{2}{\displaystyle \underset{n=1}{\overset{\infty }{\Sigma }}{w}_n^2{\left(a_n-e_n\right)}^2}}$$

hierbei ist η der Umlaufwirkungsgrad der Energiespeicherungs-/-freigabevorrichtung. Anders ausgedrückt gibt Gleichung 4 die Menge an Energie, die von der Energiespeicherungs-/-freigabevorrichtung 124 zur Erzeugung des entsprechenden Ausgleichsdrehmoments erforderlich ist, an. Typische Werte für η sind 0,7-0,9 für eine Energiespeicherungs-/-freigabevorrichtung basierend auf einem Motor/Generator und kapazitiver Energiespeicherung. Andere Energiespeicherungs-/-freigabevorrichtungen können höhere oder niedrigere Wirkungsgrade aufweisen.The power required to generate the above attenuating waveform or balancing torque is as follows: $$P=\left(1-\eta \right)\sqrt{\frac{1}{2}{\displaystyle \underset{n=1}{\overset{\infty }{\Sigma }}{\left(e_n\right)}^2}}$$

here η is the circulating efficiency of the energy storage / release device. In other words, equation 4 gives the amount of energy released by the energy storage / release device 124 is required to generate the corresponding compensation torque. Typical values for η are 0.7-0.9 for an energy storage / release device based on a motor / generator and capacitive energy storage. Other energy storage / release devices may have higher or lower efficiencies.

It is understood that Equation 4 assumes that the orbital efficiency is constant for all harmonics and that a single energy source / sink is used. In general, these are valid assumptions because an internal combustion engine is typically the ultimate source of all energy to power the vehicle and there is only a single energy storage / release device in the vehicle. Although this is generally the case, there are vehicle architectures where this cannot be the case. For example, plug-in hybrids draw energy from the power grid. Likewise, vehicles with regenerative braking can store energy in an energy storage / release device independently of the internal combustion engine. In these cases, a monitoring module can access the relative energy costs from different sources and use the optimal source or mix sources to apply the balancing torque. It should be noted that the orbital efficiency of storing and releasing drive train energy is always less than one. The energy deficit associated with this energy transfer can be taken into account in NVH mitigation, management of the energy level of a capacitor, a battery, etc.

In Equations 3 and 4, it should be noted that the harmonic components e _n at the compensating torque need not have the same strength as their corresponding motor-generated harmonic components. This means that the compensation torque does not have to eliminate all NVH, but instead can bring it to an acceptable NVH target level. At the NVH target level, the NVH can consist of two components, NVH of harmonics that are not attenuated, that is, the higher harmonics, and NVH of harmonics that may not be completely canceled.

die der folgenden Auflage unterliegt: $${}_{{\text{NVH}}_{\text{target}}\ge }\sqrt{\frac{1}{2}{\displaystyle \sum _{n=1}^{\infty }{w}_n^2{\left(a_n-e_n\right)}^2}}$$

Thus, the difficulty is in determining the lowest level of energy consumption required to achieve acceptable target NVH. This optimization problem can be expressed as a cost function, which is captured by the following equation: $${}_{{\text{min e}}_{\text{n}}\text{P}=}\sqrt{\frac{1}{2}{\displaystyle \underset{n=1}{\overset{\infty }{\Sigma }}{\left(e_n\right)}^2}}$$

which is subject to the following requirement: $${}_{{\text{NVH}}_{\text{target}}\ge }\sqrt{\frac{1}{2}{\displaystyle \underset{n=1}{\overset{\infty }{\Sigma }}{w}_n^2{\left(a_n-e_n\right)}^2}}$$

This optimization problem can be represented graphically. Two simplified examples are given in 4 and 5 shown. 4 represents a set of circles 402a . 402b and 402c and a set of ellipses 404a and 404b represents the energy cost function, ie equation (5), or the circulation function, ie equation (6), for a certain ignition component in question. This exemplary diagram only contains the first two harmonics. The strength of component e _{1 of} the first harmonic of the compensating torque is presented along the horizontal axis, and the strength of component e _{2 of} the second harmonic of the compensating torque is presented along the vertical axis. The values of components a ₁ and a _{2 of} the motor-generated first and second harmonics are shown. Each cylinder load, ignition component, engine speed, and transmission gear has an associated set of a ₁ and a ₂ , which can be determined by vehicle calibration or any other means.

In 4 represents each concentric circle in the set of circles 402a - 402c represents a constant amount of energy used to mitigate torque, assuming that the efficiency of the energy storage / release device 124 is the same for the frequencies of the first and second harmonics. The smaller the circle, the less energy used. The center of the set of circles, the origin, indicates a point at which no energy is used, ie e ₁ = e ₂ = 0. Any concentric ellipse in the set of ellipses 404a - 404b represents an NVH target amount generated by the components of the first and second harmonics. Each point on or in the selected target NVH ellipse creates an acceptable NVH amount. The eccentricity of the ellipses 404a and 404b is determined by the ratio of the weighting factors w ₂ / w ₁ . With the same weighting factors, the ellipses shrink into circles. Generally people are more sensitive to the low frequency first harmonic and thus the ellipses are in 4 extended vertically. E ₁ requires less variation to change the NVH by a fixed amount than e ₂ . The smaller the ellipse, the lower the NVH allowed. The middle-point 406 the ellipses 404a and 404b represents a situation in which all NVHs related to the first and second harmonics have been eliminated. At point 406 , a ₁ = e ₁ and a ₂ = e ₂ , the compensating torque exactly cancels the first and second harmonics of the torque variation generated by the engine.

To optimize energy costs, it is desirable to use as little energy as possible to reduce NVH to an acceptable level. Assuming that the acceptable NVH level is given by the ellipse 404b is defined, this goal is reached at point A, where the ellipse 404b and the circle 402c to cut. Point A achieves acceptable NVH because it is on the ellipse 404b located, and minimizes energy consumption because this point is on the ellipse 404b is closest to the origin, ie the circle 402c is consistent with the intersecting ellipse 404b as small as possible.

Provides for comparison purposes 5 is a diagram that contains another vehicle operating point, ie cylinder load, ignition component, engine speed and / or gear. For example, the engine torque, engine speed, and transmission gear may be those of 4 be identical, but the ignition component and the cylinder load can be different. This operating point has significantly different NVH properties than the ignition component and the cylinder load accordingly 4 on. 5 points those of 4 similar axes on and the concentric circles 502a . 502b and 502c provide constant energy expenditure from the energy storage / release device 124 represent. Similarly, the ellipses 504a and 504b different acceptable levels of NVH generated by the first and second harmonic components 5 are the first and second harmonics a ₁ and a ₂ generated by the motor. When e ₁ = a ₁ and e ₂ = a ₂ , powertrain operation is at point 506 and no NVH are generated by the first and second harmonics. Assuming that the acceptable NVH level is given by the ellipse 504b is defined, there is no compensation torque from the energy storage / release device 124 Required to meet the NVH target because point B, corresponding to e ₁ = e ₂ = 0, is within Ellipse 504b lies. If the ellipse 504a would indicate the acceptable level of NVH, some weakening of the balancing torque would be required to achieve the target value.

It is understood that the graphic explanation given in 4 and 5 is shown is appropriate in the case where the first two harmonics can be attenuated by the energy storage / release device. If only the first harmonic is taken into account, the two-dimensional circles and ellipses become lines. Likewise, if the first, second, and third harmonics are considered, the circles become spheres and the ellipses become ellipsoids. In general, the number of optimization variables corresponds to the number of harmonics that are potentially weakened. Any number of harmonics can be attenuated if desired, but generally only one or two harmonic attenuations are required to achieve acceptable NVH behavior, as discussed above.

The above approach assumes that an acceptable level of NVH has been set. The acceptable level of NVH can be determined in any suitable manner. For example, extensive user tests can be performed to determine the level of vibration that is acceptable to vehicle occupants. It is understood that the acceptable level of NVH may vary dynamically based on various conditions. In some embodiments, the acceptable level of NVH is determined based on road conditions, user selection, gear engaged, gear change, vehicle speed, cabin noise level, engine idling, accelerator pedal position (e.g., rate of change in accelerator pedal position) and / or any other suitable engine parameters or criteria.

With reference back to 2 determines the NVH reduction module 121 using any of the above methods, the energy cost required to mitigate the NVH of each candidate high NVH firing fraction so that the associated NVH is reduced to acceptable levels. The total energy cost associated with the high NVH ignition component in question is the sum of the attenuation costs and costs associated with operating the engine with the ignition component in question and the cylinder load. It is to be understood that any known method can be used and the energy cost estimation process is not limited to the examples, diagrams and equations provided above.

At step 208 the NVH reduction module compares the energy costs associated with each of the ignition components in question. The manner in which this comparison is made may vary depending on the properties of each ignition component in question. Consider an example where it is believed that each cylinder is ideally fired under optimal conditions, e.g. B. in which the throttle valve position, the air mass charge, the ignition timing advance, the valve timing and other engine parameters are essentially optimized for fuel efficiency. It is further contemplated that in this example both a low NVH firing portion and a high NVH firing portion may be used to provide the target torque. The high-NVH ignition component is able to provide the target torque under almost optimal cylinder conditions. However, mitigation is required to reduce the resulting NVH. The low NVH firing fraction, on the other hand, has the opposite problem - although it has acceptable NVH characteristics, it cannot achieve the desired torque without certain adjustments to the cylinder output, that is, by deviating from the above optimal conditions, resulting in a loss in fuel efficiency , provide. Thus, comparing the energy costs of these two candidate firing fractions includes comparing the energy cost (losses) of adjusting the cylinder output associated with the low NVH firing fraction with the energy cost of weakening the NVH associated with the high NVH Ignition share related. Such comparisons can be made between any number and any types of ignition components in question.

Based on the above analyzes and / or comparisons, the NVH reduction module and / or the firing fraction calculation device selects the candidate firing fraction that provides the target torque in the most fuel efficient manner (ie, with the lowest energy cost). at In some embodiments, other factors are considered in the selection process. The selected ignition component in question becomes the operational ignition component (step 210 ).

An exemplary process for choosing an operational ignition component from multiple candidate ignition components is shown in 6 described. 6 FIG. 12 is a graph showing fuel consumption, inversely proportional to fuel efficiency, as a function of the ignition percentage. The graph assumes an engine speed of 1200 rpm and an engine torque fraction (ETF - Engine Torque Fraction) of 0.2. (In this example, the ETF represents a target engine torque. For example, ETF = 1 full engine output.)

The vertical axis of the graph represents fuel consumption (grams per second). The horizontal axis represents eligible ignition fractions. In this figure, dots marked by a circle in a square indicate a low NVH ignition fraction with no NVH weakening is necessary to achieve acceptable NVH. Data points with an x in the circle indicate a high NVH firing fraction where the NVH is unacceptable without attenuation. Immediately above these points are points marked with a square, indicating the total fuel consumption associated with both operating the engine and balancing the torque using the aforementioned methods to bring the NVH to an acceptable level ,

Without any NVH weakening, there is a point 604 the choice of the most fuel efficient ignition component, ie an ignition component of 0.5 that provides the target torque, has acceptable NVH characteristics and has a fuel consumption rate of approximately 0.93 g / s 606 however, is a better choice versus point 604 because point 606 requires less energy (about 0.87 g / s) and uses an ignition ratio of 0.4, while also providing the target torque. Although the firing fraction of 0.4 is known to produce unacceptable NVH levels at 1200 engine speed, the NVH can be mitigated using a compensating torque. The calculated energy costs of 0.87 g / s take into account the energy costs of the weakening and are still below the point 604 associated energy costs. Thus, in this simplified example, the ignition component 0.4 is selected as the operational ignition component. The fuel savings in this case ((0.93-0.87) / 0.93) are about 6.5%, which shows the advantage of using the control method described here.

It goes without saying that the choice of the operational ignition component can be based on factors other than fuel efficiency. For example, in some embodiments, the status of the energy storage / release device plays 124 a role in the selection process. That is, consider an example in which it is determined that a certain high NVH ignition component is suitable for providing the target torque. In addition, the NVH reduction module determines 121 that the NVH related to the ignition component with a compensating torque, e.g. B. can be weakened sufficiently using the methods described above. The NVH reduction module 121 also determines the amount of energy required to generate an appropriate balancing torque. However, the NVH reduction module 121 and / or the ignition component calculation device 112 also determine that the ignition portion cannot be selected as the operational ignition portion because the energy storage / release device is currently unable to generate the required balancing torque (e.g. based on battery charge level, lack of stored energy, inability to provide the determined amount of energy, etc.). Conversely, the cost of the attenuation when the energy storage device is nearly full, possibly due to regenerative braking, may be reduced compared to the previous calculation.

With reference back to 2 transmits the ignition proportion calculator 112 at step 212 the selected operational ignition component to the ignition timing determination module 120 , The ignition timing determination module generates based on the operational ignition component 120 a Skip Fire firing order (step 212 ). At step 214 a determination is made as to whether the operational ignition component requires NVH mitigation. If it is not (e.g., it is a low NVH firing fraction), then the process goes to step 222 about. At step 222 the engine is operated in a skip fire manner based on the firing order.

If it is determined that the operational firing fraction requires NVH mitigation, the NVH reduction module determines 121 a suitable compensation torque (step 216 ). The balancing torque may be any suitable balancing torque or waveform that is provided by the energy storage / release device 124 on the powertrain is applied to help reduce the NVH generated by the firing order. The compensation torque may be determined using any suitable algorithm, method, or mechanism (e.g., any of those associated with FIG 1 described methods) are generated.

One approach can be described as follows. After the ignition proportion calculator 112 has selected an operational ignition component and determined that a suitable compensation torque must be generated, transfers the ignition component calculation device 112 the operational ignition component to the ignition timing determination module 120 , The ignition timing determination module then generates a skip fire ignition sequence based on the operational ignition component.

The firing order becomes the NVH reduction module 121 transfer. The NVH reduction module analyzes the Skip Fire firing order and identifies one or more selected engine torque variations that are generated by the sequence. This can be done in a number of ways. For example, in some embodiments, the torque may be characterized as a torque waveform that has a fixed component and a variable component (e.g., from variations / sine functions of multiple harmonics). Some approaches involve choosing the sine function of harmonics, the frequency of which is the fundamental frequency. Other approaches involve selecting sine functions from multiple harmonics whose associated frequencies include the fundamental frequency and one or more other frequencies (e.g., the second harmonic, etc.).

The NVH reduction module 121 then generates a compensation torque based on the selected variations / sine functions. As previously discussed, in various approaches, the compensation torque is in the form of one or more sinusoidal waveforms at substantially the same frequency as the harmonics generated by the internal combustion engine. In some approaches, the balancing torque waveform would have the same frequency as the selected variations, but would be out of phase (e.g., 180 ° shifted) to cancel out the torque variations generated by the motor. The compensation torque is determined such that it at least partially, but not necessarily completely, cancels the selected variation (s) that are the cause of at least a portion of the NVH. In various applications, the magnitude of the balancing torque waveform is determined to bring the NVH generated by the firing order below a predefined level.

At step 218 the engine will be based on that at step 210 selected operational ignition component and its corresponding ignition sequence operated Skip Fire-like. At step 220 it will at step 216 certain compensation torque by the energy storage / release device 124 applied to the drivetrain while tuning the Skip Fire firing order on the engine. Thus, the compensation torque helps reduce the NVH generated by the Skip Fire firing order. In various embodiments, the NVH reduction module receives 121 any suitable inputs (e.g., firing order, engine speed, current cylinder, etc.) necessary to properly coordinate the application of the balancing torque and the execution of the firing order.

The above steps of procedure 200 can be done from ignition opportunity to ignition opportunity. Alternatively, one, some, or all of the above steps may be performed a little less frequently, such as: B. one or more times per engine cycle.

Regarding 3 becomes a powertrain system 300 according to a particular embodiment of the present invention. The powertrain system 300 includes powertrain control 100 , an internal combustion engine 304 , an energy storage / release device 124 , a crankshaft 308 , a gear 312 and wheels 314 , The motor 304 and / or the energy storage / release device 124 are designed to apply torque to the crankshaft 308 to put on the wheels 314 about the transmission 312 drives. The powertrain control, which in 1 is described is designed to operate the engine 304 and the energy storage / release device 124 to coordinate. The powertrain system can be used in conjunction with any of the 1 . 2 . 12 . 13 . 15 and 17B described methods are operated. It is understood that, although in 3 a particular powertrain configuration is presented, the components of the figure can be positioned in any suitable arrangement.

The energy storage / release device 124 is designed to add torque to the drive train or to subtract torque from it. In various embodiments, the Energy storage / release device 124 a waveform of pulses of the balancing torque. The waveform of the equalizing torque pulses generated by the energy storage / release device 124 is essentially the sum of one or more sinusoidal waveforms that apply torque at one moment and subtract torque at another moment. In general, the waveform of the compensation torque pulses is designed to cancel out a selected torque variation generated by the engine (e.g., as in conjunction with the NVH reduction module 121 of 1 and step 216 of 2 is discussed).

The energy storage / release device 124 may be any suitable device or devices that can absorb or subtract torque from the powertrain, store the resulting energy, and / or use the energy to add torque to the powertrain. In various implementations, the energy storage / release device includes 124 a motor / generator and a battery or a capacitor. In further implementations, the energy storage / release device stores 124 Energy and releases it mechanically (e.g. a flywheel), pneumatically or hydraulically.

Some embodiments include an energy storage / release device 124 which is designed to have multiple applications, ie other applications in addition to generating a compensating torque. In some applications, the energy storage / release device subtracts 124 for example, in the same manner as any modern hybrid vehicle to improve fuel efficiency (e.g. using regenerative braking, etc.) torque from the powertrain and add torque to the powertrain. This means that the energy storage / release device supplies a DC component to the drive train torque in addition to supplying an oscillating balancing torque. This DC component can be positive or negative depending on the operating conditions, the amount of energy currently stored in the energy storage / release device and other variables. The DC component can be selected in part to compensate for inefficiencies associated with the storage and release of energy from the energy storage / release device. The energy storage / release device 124 can also be an integrated starter generator that is used as part of a start / stop engine system to restart an engine.

In various approaches, the energy storage / release device 124 also used to smooth transitions between different ignition components. For example, if the engine is operated as a skip fire and is switched from a lower ignition component to a higher ignition component, it may take the intake manifold pressure to adjust from a higher to a lower level. This means that if the change is made immediately, the vehicle will jump forward because the cylinder output is too high. In various applications and under such circumstances, the energy storage / release device is 124 designed to absorb torque from the drive train / supply torque to the drive train, thereby helping to ensure a smoother transition between the ignition components. For example, the powertrain control 100 and the energy storage / release device 124 using any of the methods or steps described in the U.S. Patent Application No. 13 / 654,244 and the preliminary U.S. Patent Application No. 62 / 053,351 are described, to which reference is made here in their entirety.

The invention has been described primarily in the context of controlling the ignition of four-stroke reciprocating engines suitable for use in automotive vehicles. However, it is understood that the described skip fire approaches are well suited for use with a wide variety of internal combustion engines. These include engines for almost any type of vehicle - including passenger cars, trucks, boats, construction equipment, aircraft, motorcycles, scooters, etc .; and almost every other application involving the ignition of working chambers and using an internal combustion engine. The various approaches described work with engines that operate under a variety of different thermodynamic cycles - including almost every type of two-stroke reciprocating piston engine, diesel engine, gasoline engine, double cycle engine, miller engine, Atkinson engine, Wankel engine and other types of rotary piston engine, mixed cycle engine (such as double Otto - and - diesel engines), radial engines, etc. It is further believed that the approaches described will work well with newly developed internal combustion engines regardless of whether they are operated using currently known or later developed thermodynamic cycles.

In some preferred embodiments, the ignition timing determination module uses sigma-delta conversion. While it is believed that sigma-delta converters are well suited for use in this application, it should be understood that the converters can use a variety of modulation schemes. For example, pulse width modulation, pulse height modulation, CDMA-oriented modulation or other modulation schemes can be used to provide the control pulse signal. Some of the described embodiments use first order transducers. However, higher order converters or a collection of predetermined firing sequences may be used in other embodiments.

It is understood that the powertrain control designs that are contemplated in the present application are not limited to those in FIG 1 and 3 special arrangements shown are limited. One or more of the modules shown can be integrated with one another. Alternatively, the features of a particular module can instead be divided among multiple modules. The controller may further include additional features, modules, or steps based on other pending patent applications, including U.S. Patents No. 7,954,474 ; 7,886,715 ; 7,849,835 ; 7,577,511 ; 8,099,224 ; 8,131,445 ; and 8,131,447 ; U.S. Patent Applications No. 13 / 774,134 ; 13 / 963.686 ; 13 / 953.615 ; 13 / 886.107 ; 13 / 963.759 ; 13 / 963.819 ; 13 / 961.701 ; 13 / 963.744 ; 13 / 843.567 ; 13 / 794.157 ; 13 / 842.234 ; 13 / 004,839 . 13 / 654.244 ; 13 / 004,844 ; 14 / 207.109 ; and 13 / 681.378 and preliminary U.S. Patent Applications No. 61 / 952,737 and 61 / 879.481 , which are referred to here in their entirety. Any of the features, modules, and steps described in the above patents can be used by the controller 100 to be added. In various alternative implementations, these functional blocks of algorithms can be achieved using a microprocessor, ECU, or other computing device, using analog or digital components, using programmable logic, using combinations of the foregoing, and / or in any other suitable manner ,

A method for reducing vibrations using an electrical machine is described in U.S. Patent No. 8,015,960 , although the method differs from various embodiments of the present invention in some aspects. First, the '960 patent focuses on a control system for a variable displacement engine, not a control system for a Skip Fire engine. Furthermore, the '960 patent describes the following process: 1) determining the torque applied to a crankshaft; 2) extracting components from the torque attributable to the uneven cylinder firings in a variable displacement mode; 3) removing a specified component (a specified target torque) from the extracted component so that only the variable component remains in the extracted components; 4) Generate a vibration damping torque that counteracts the variable component generated in the third step. In other words, the vibration damping torque is designed to counteract any variation (ie after removal of the target torque) attributable to uneven cylinder firings. In various embodiments of the present invention, compensation torque is generated that does not necessarily counteract any variation attributable to uneven cylinder firings. Instead, in some embodiments, the compensating torque counteracts only part of the variation (e.g., the compensating torque can only counteract one or more of the harmonic sine functions at certain frequencies, such as the fundamental frequency, etc.). In various implementations, the compensation torque does not counter certain types of variations attributable to uneven cylinder firings, e.g. B. it cannot counteract one or more other sine functions of the harmonics. Various embodiments of the present invention further describe a control algorithm that only applies a sufficient amount of compensation torque to achieve an NVH target value. The control algorithm also selects an operational ignition component that maximizes fuel efficiency, taking into account the energy costs associated with generating the balancing torque. Another difference between the present invention and the prior art is that the frequency of the equalizing torque cannot correspond to the firing frequency. For example, the ignition frequency at 1500 rpm and an ignition share of 40% is 40 Hz, but the compensation torque can have a frequency of 20 Hz.

Although the invention has been described generally in terms of using an ignition component to characterize the firing order, this is not necessary. 12 shows an embodiment of a powertrain control 1200 , Many of the different elements of powertrain control 1200 are those who are in relation to the in 1 Powertrain control shown 100 shown and described, similar or identical to these. In contrast to powertrain control 100 shows 12 the generation of the control pulse signal 113 directly from a torque request signal 111 without reference to an ignition component. Instead, a firing order generator 1202 the control pulse signal 113 produce. The control pulse signal 113 may consist of a bit stream where every 0 indicates a deactivation and every 1 an activation for the current cylinder firing opportunity that defines the firing order. The firing decision associated with each firing opportunity is generated prior to the firing opportunity by a sufficient amount of time for the ignition control unit 140 to properly configure the engine, such as deactivating a cylinder intake valve in the event of a missed ignition opportunity. Each ignition opportunity has a known torque signature depending on whether the ignition opportunity corresponds to a deactivation or an activation and the settings of the powertrain parameters by the powertrain parameter setting module 116 To be defined.

The firing order and the compensation torque can be determined using various methods. In one embodiment, short-term model predictive control that balances requested and provided torque, NVH, and energy costs associated with generating the balancing torque as part of the optimization problem may be used. In various embodiments, model predictive control is an optimal control method that easily treats systems with multiple performance criteria using a short-term calculation of the optimal control. In various implementations of this method, discrete optimization is performed each time new system sizes are used to compute the best system inputs to apply the regulated system at the current time. The process repeats this optimization every time a new input is requested. Inputs to the model may include the requested torque, torque signature associated with deactivations and activations, the appropriate NVH level, the acceptable level of emissions and energy costs, and energy / performance constraints associated with generating a balancing torque. Model variables can include engine speed, gear position, engine and ambient temperature, road conditions, and engine parameters such as: B. MAP, valve timing and ignition timing control include.

Applying this control method can determine and evaluate various combinations of powertrain parameters, balancing torques, and firing sequences that provide the requested torque on every firing occasion in the firing sequence generator 1202 include. The firing order generator 1202 can then, depending on the system requirements, generate an ignition sequence that provides optimal / improved fuel economy with acceptable NVH. This control method is particularly useful when the firing order is not periodic, e.g. B. during a transition between firing sequences, with changing torque requirements 111 related, but of course also applies to uniform torque requirements. This may refer to ignition decisions that have been made but have not yet been implemented. This can range from 4 to 8 firing opportunities, since these decisions are known prior to their implementation, the compensation torque can be calculated in advance. Compensatory torque requirements may include maximum allowable torque levels and frequency provision restrictions. Energy storage requirements may include current energy storage utilization as well as maximum power transfer from the energy storage device.

In another embodiment, an adaptive filter (AF) feed forward (FF) system can be used to dampen undesirable torque vibrations caused by combustion events. In some embodiments, the AF-FF control can take advantage of the fact that the firing order and the resulting torque disturbances are clearly defined. 13 shows a schematic diagram of an AF-FF controller. The task of the AF-FF controller is to mitigate interference in a system of interest, and the AF-FF controller accomplishes this task by generating a noise-canceling signal that counteracts the interference when applied to the system. In this case, the disturbance is the variation of the engine torque 1310 from its mean. A noise filter signal output 1314 is through a digital filter 1304 based on an input interference-correlated signal 1312 generated. The interference correlated signal 1312 may have an average of zero so as not to change the average total powertrain output torque. The interference correlated signal 1312 contains information regarding the expected disturbance with a certain lead time. This signal 1312 can be based on firing order and estimated torque signatures associated with deactivations and activations. The firing order can be derived using a torque request, a firing fraction, a sigma-delta filter, a lookup table, a state machine, or by any other means. The filter output 1314 can be in a balancing torque module 1302 can be entered, which is a compensation torque 1316 generated. The compensation torque module 1302 represents the dynamic reaction of a motor / generator or one any other system that applies a compensation torque that includes any response delays or restrictions to produce the desired compensation torque 1316 The compensation torque 1316 becomes with the engine torque 1310 at a summing point 1318 combined. The summing point 1318 gives a delivery torque 1320 to the drive train. The summing point 1318 shows the subtraction of the compensation torque 1316 from the engine torque 1310 , It is understood that the compensation torque may have the opposite polarity in further embodiments and the compensation torque is added to the motor torque.

Adaptive filter parameters called weights can be adjusted by a weight update module 1306 be updated using an adaptive algorithm to reduce the differences between the balancing torque 1316 and the disturbance to a minimum, with the engine torque 1310 is less than the mean to the delivery torque 1320 compensate. The weight update module 1306 uses a model of a balancing torque module 1302 and inputs of both the delivery torque 1320 as well as the interference-correlated signal 1312 to determine the appropriate weights. Reducing to a minimum may include reducing an average squared difference between the signals to a minimum, although other minimizing criteria can be used.

Graphs showing the time behavior of the different signals in 13 represent, for better understanding and better explanation of the operation of the exemplary powertrain control 1300 shown. The graph 1311 of engine torque represents vibrations in engine torque output similar to those previously described in 8th were shown. The interference-correlated signal curve 1313 shows an estimated signal of the disturbance in engine torque, which is used to reduce variations in the torque provided 1320 to a minimum. This estimate reflects the skip fire nature of the ignitions, making it the digital filter 1304 provides the necessary frequency component information, which results in the filter output 1314 has the proper frequency components. The digital filter controls based on the various weightings that are related to the size and phase response of the filter 1304 adaptive the filter output 1314 in that, disturbances in the torque provided 1320 to a minimum. The graph 1315 the filter output represents how the noise correlated signal 1312 through the digital filter 1304 is modified. The filter output 1314 is in the compensation torque module 1302 that includes a motor / generator or similar system that can generate or absorb torque. The compensation torque module 1302 gives a compensation torque 1316 from that in the graph 1317 is shown. The graph 1317 represents how the compensation torque 1316 with the variations in engine torque 1310 matches and cancels them. If the compensation torque and the motor torque in the summing point 1318 be combined, the resulting delivery torque 1320 relatively small torque variations, as in graph 1321 is shown.

An advantage of various implementations of the AF-FF control is that, since it is a feedforward control, it can eliminate or at least minimize any disturbance in the powertrain torque within the bandwidth of the compensation torque module, provided that the Advance in time of the interference-correlated signal 1312 related to the disturbance is greater than the delay caused by the balancing torque module 1302 and filter calculations in the digital filter 1304 is caused. Setting the weights in the digital filter 1304 are used is comparatively slower than the change in the disturbance, however, this will affect the ability of the controller 1300 not limited to dampening the interference. The adaptive algorithm determines the powertrain characteristics with respect to the disturbance-correlated signal 1312 and the actual disturbance (variation in engine torque 1310 ) as well as the secondary path dynamics with the compensation torque module 1302 associated, which are fixed or whose rate of change is much slower than that of the disturbance.

An input to both the adaptive filter feed forward control and the short term model predictive control can be the torque signature associated with cylinder deactivations and activations. 14 shows exemplary torque signatures with the curve 1410 for activations and the curve 1420 for no activations (deactivations) related. These exemplary curves represent the normalized torque output assigned to a cylinder over a working cycle of 720 degrees crankshaft rotation. These exemplary torque signatures can be scaled and set based on the engine parameters. The total engine torque results from the sum of the torque generated by all cylinders. The estimated total engine torque can then be used as part of a short term model predictive control or an adaptive filter feed forward system.

adaptive control

The related to 4-11 The control methods described work quite well with stable operation; however, they depend on the accuracy of the underlying engine torque and auxiliary torque model to provide optimal fuel efficiency with acceptable NVH. An adaptive control procedure overcomes these restrictions. With adaptive control, a controller adapts or modifies its control characteristic based on the observed characteristic of the control task, which varies or is initially uncertain. In the present invention, adaptive control uses an NVH level to modify the control signals to the auxiliary torque source / sink to bring NVH to an acceptable level while offering fuel efficient operation.

One type of adaptive control is an adaptive filter feedforward control or AF-FF control. In AF-FF, a filter that influences the control system response is modified, i. H. adapted based on a captured output. The system characteristics vary based on the detected output, i.e. H. of the adaptive filter. Typical changes to the filter characteristics are relatively slow compared to variations in the captured output.

In one embodiment, AF-FF control is used in a micro-hybrid vehicle that has an internal combustion engine and a motor / generator that are connected to a common drive train. The motor / generator can counteract torque vibrations that are caused by combustion events in ignited working chambers and can therefore dampen the torque vibrations in the drive train. The motor / generator can also counteract vibrations other than driveline loads, such as those from skipped working chambers, a camshaft drive, etc; however, these vibrations tend to be smaller in strength than those resulting from an activated work chamber. In one embodiment, the compensation torque can be generated, for example, by an electric motor / generator that has a capacitor that serves as an energy storage / collection / release device. An advantage of an AF-FF control is that it can compensate for any inaccuracy in dynamics from a weakening command to the compensation torque actually applied to the crankshaft. It can also help compensate for changes in powertrain performance due to wear or other factors. AF-FF control also works seamlessly in both stable and transient conditions, such as during a spark transition.

In one example of such a micro-hybrid system, an internal combustion engine may be a four-cylinder, four-stroke engine capable of delivering a maximum output of 200 horsepower (≈150 kW). Aid can be supplied / subtracted from an electric motor / generator, a maximum stable output of 10% or less than this value, i. H. ≤ 15kW. The electric motor / generator can be electrically connected to a capacitor for energy storage. Under normal operation, the capacitor energy storage / release capacity can only be sufficient to attenuate the variation in torque that is generated by an ignition sequence. An engine running at 1000 rpm has an ignition opportunity every 30 ms. The maximum energy that a 15 kW motor can deliver during this period is 450 J. In order to reduce torque fluctuations, the motor / generator would in practice not work with its maximum output for most of this period, so that the actually delivered or stored energy can be less than this value, for example 1/3 or 150 J. In order to store this energy in a capacitor, there is a trade-off between the capacitor voltage, the permissible amount of energy drawn off and the capacitor capacity. For example, the capacitor may have a capacitance of 0.3 F and be operated at a peak voltage of 125 V. During normal operation, the capacitor voltage may fluctuate between 120 V and 125 V when energy is stored in and released from the capacitor, a net change of approximately 184 J, slightly more than the 150 J estimate from the previous calculation. Operation with a relatively small drop in capacitor voltage from its peak value is advantageous since it generally improves the efficiency of the motor / generator. For example, the capacitor voltage cannot vary by more than 2%, 4%, or 10% from its maximum value because it stores and releases energy to dampen motor torque vibrations. It is understood that the values given here are only exemplary and the current invention can be used with both larger and smaller internal combustion engines, motor / generators and storage capacitors. The capacitor peak voltage can be higher or lower than 125 V, such as 300 V, 48 V, or any other voltage.

In addition, a battery can be connected in parallel to the capacitor in order to enable greater energy storage. The battery can be used to store energy from a regenerative Brake system are used and / or can be part of a stop / start system that switches off the engine when there is no torque request. In some embodiments, an electrical system that operates with peak voltages less than 60 V can be used. These low voltage systems are advantageous because the costs associated with isolating the electrical system elements are lower due to the lower voltage. In particular, a battery that works with a nominal voltage of 48 V can be used. Such a battery can have a battery chemistry that allows quick and efficient charging and discharging. Exemplary types of battery chemistry include, among others, lithium titanate, lithium iron phosphate, lithium iron manganese phosphate, nickel manganese cobalt, or any other type of fast-charging / discharging battery chemistry. The battery components can be arranged as small particles, for example nanoparticles, in order to enlarge the surface area available for chemical reactions in order to increase the charge / discharge rate. An exemplary battery that can be used is an 8 Ah UltraPhosphate battery available from A123-Systems of Livonia, Michigan. Such a battery has a total energy storage capacity of approximately 1.3 MJ, so that it can be used both to start combustion, for regenerative braking, and as the energy source / sink when balancing torque is applied.

Although an electrical system using only a battery as the energy storage element can be used, an advantage of using a capacitor for energy storage is that the circulating efficiency of storing and releasing energy from a capacitor is typically higher than in a battery , for example 80% versus 50%. Thus, the battery can be used when the energy storage needs are greater, and the capacitor can be used when the energy storage needs are moderate, such as damping the torque vibrations of an internal combustion engine.

The exact arrangement of the motor / generator and its relationship with the drive train can vary widely. For example, the motor / generator can either be mechanically coupled into the drive train between the motor and the wheels or on the motor side opposite the wheels. The motor / generator can be coupled via a belt-pulley system, a chain-pinion system, a gearwheel system or can be connected directly to the drive train. Systems with a motor / generator coupled to the crankshaft by an accessory drive belt, commonly referred to as a front end accessory drive belt (FEAD belt), can be inexpensive when combined with a 48 V battery system in a micro or mild hybrid system ,

The motor / generator can be an AC induction electric motor / generator that can advantageously generate high torque and can work symmetrically as a motor or a generator. Other types of electric motor / generators can be used; for example, a brushless DC motor / generator with buried permanent magnets, a brushless DC motor / generator with surface permanent magnets, a switched reluctance motor / generator, or any other type of motor / generator. All of these motor / generator types are very efficient in converting mechanical energy into electrical energy and vice versa. Conversion efficiencies are generally higher than 80%. Advantageously, a brushless DC motor / generator with buried permanent magnets provides very high efficiency operation, typically in the range of 92-95%. Another consideration when choosing an electric motor / generator is its operating speed range. A switched reluctance motor / generator can advantageously be operated over a wider speed range than some of the other motor / generator types. This is particularly advantageous in a PO hybrid architecture, in which the internal combustion engine and the motor / generator are connected by means of a belt in a front-end accessory drive (FEAD) and rotate at the same speed.

Much of the discussion so far has focused on optimizing fuel efficiency during stable or nominally stable operating conditions. An acceptable level of NVH or any other parameter can be selected and an algorithm that maximizes energy efficiency by summing up the contributions of the engine and the electric motor / generator has been implemented. However, in a normal driving cycle, the torque required to operate the vehicle varies almost continuously with changing driving conditions. In a conventionally controlled Otto cycle internal combustion engine, throttle control is the primary mechanism for controlling engine output torque. In a gasoline engine with skip fire control, controlling the amount of ignition or the ignition density is the primary mechanism for controlling the engine output torque. In order to minimize brake specific fuel consumption (BSFC) and thus optimize fuel efficiency, it is generally desirable to operate the engine at or near a wide open throttle to minimize pumping loss. As described above, during Skip Fire operation, operation becomes active certain ignition components that provide desired NHV characteristics preferred. These desirable firing fractions tend to be simple fractions with a small denominator, such as ½, 1/3, 2/3, ¼, ¾, 1/5, 2/5, 3/5, 4/5, 1/6, etc Much of the discussion so far has focused on optimizing fuel efficiency by selecting the spark and balance torque combinations that provide the requested torque using one of these spark levels. However, in order to adapt to changes in the required engine output torque, the engine has to switch or transition between these different ignition component levels. During a normal driving cycle, such as an FTP (Federal Test Protocol) fuel efficiency test cycle, the engine may spend approximately 20% of its time transitioning between different levels of ignition. It is therefore important to optimize fuel efficiency during these transitions while providing acceptable NVH performance.

In previous U.S. patents and patent applications, the applicant has described various strategies for providing acceptable NVHs during ignition share level transitions in non-hybrid vehicles. These strategies include slowing the ignition fraction transition to match the air intake manifold dynamics, matching ignition patterns through a transition, using a linear rise time for the ignition fraction in a transition, using firings with different output levels, coordinating the ignition fraction change with movement of different actuators, using additional firings into the firing order and using forward and feedback control based on firing order induced torque variations during the transition. The present invention extends these ideas to use in a hybrid vehicle that uses an auxiliary torque source / sink to dampen driveline vibrations.

In one embodiment, the feedback signal may correspond to a sensed output of the engine, such as an engine torque. However, the feedback signal can be any detected signal that correlates with the torque actually applied to the crankshaft, such as crankshaft acceleration, that can be obtained from a crankshaft speed measurement. The speed is often measured in rpm, revolutions per minute. The crankshaft angle acceleration is simply the time derivative of the speed. Thus, the crankshaft angle acceleration measurement can be used to derive the torque applied to the drive train by the engine and the auxiliary power source / sink. The crankshaft acceleration signal can be filtered to reduce noise, low frequency components of engine torque, and variations in load torque. Although load torque and other torques exerted on the wheels (which are usually unknown) affect acceleration, the frequencies of these torques are often low compared to those of the engine torque vibrations (2 to 50 Hz) which a micro-hybrid system is designed to combat , The attenuation frequency range from 2 to 50 Hz is only an example and larger and smaller frequency ranges can be used. The attenuation frequencies should generally be above the frequencies that affect driver torque requests or load torque variations. Therefore, crankshaft acceleration can be used instead of direct torque measurement for the purpose of damping torque vibration since it reflects the engine torque pulses associated with firing and exhausting working chambers. By extending this idea, in one embodiment, measurements directly related to NVH metrics are used instead of or in addition to crankshaft acceleration, such as an acceleration measurement of a seat rail or a sound measurement near an ear of the driver, such as a seat headrest. The NVH metric includes varying human perceptions of different vibration and noise frequencies. In some embodiments, the NVH metric may be based on a crankshaft rotation signal or a time derivative thereof.

15 shows the relationship between 1 and 13 according to one embodiment. The ignition share calculation device 112 and the ignition timing determination module 120 each generate an ignition component 119 and a firing order 113 based on the torque request 111 , The firing order, or alternatively the control pulse signal, 113 is known before the actual combustion events of the cylinder, so that a planned firing order as the basis for the fault-correlated signal 1312 can be used. In general, the disturbance-correlated signal is an input to a control system with regard to a likely undesirable fluctuation on the output that is at least partially canceled by the control system. The disturbance correlated signal is fed to the control system prior to the fluctuation so that the control system can minimize or reduce the fluctuation using feedforward control. The noise correlated signal is in the noise correlated signal generation module 1590 generated. The module 1590 can be the ignition sequence as inputs 113 and the powertrain adjustment parameters 109 which allow estimates of the torque profile associated with ignited and skipped work chambers. These signals can optionally be through a Delay compensation module 1596 be directed to help synchronize the fault correlated signal with the fault, since the firing / skip decision is generally known some firing opportunities before it is executed. Using the delay compensation module 1596 can be the number of weights required in the digital filter 1304 reduce, which reduces calculation time and complexity. The interference correlated signal generation module 1590 can be a torque model 1592 or the filter 1594 use to generate the noise-correlated signal 1312 to help. The torque model 1592 can the torque profile associated with ignited and skipped work chambers, as in 14 with the powertrain setting parameters 109 is shown scaled. The torque model 1592 may include firing order and other engine operating parameters such as MAC, spark timing and air / fuel mixture. Based the interference correlated signal 1312 on the torque scaling model can improve transient performance because of the weights in the digital filter 1304 will vary less by any transition. The bias torque can be removed so that only the vibration components of the torque profile in the interference-correlated signal 1312 be included. That is, in some embodiments, the interference correlated signal is substantially bias free with a mean of substantially zero. Essentially here means that the DC component of the interference-correlated signal is small compared to the AC components, for example the DC component of the interference-correlated signal is less than 10% of the root mean square (rms) of the AC components. An overall engine torque model is described in greater detail in U.S. Provisional Patent Application No. 62 / 379,357, which is hereby incorporated by reference.

The interference correlated signal 1312 is in the filter module 1340 directed. The filter module can have two elements, the digital filter 1304 and the weight update module 1306 that works as previously described. It is understood that the filter module 1340 can take a wide variety of forms, some of which are described in more detail below. The filter module 1340 can also have an NVH signal 1309 from the NVH reduction module 121 comes. The NVH signal 1309 helps optimize the trade-off between NVH and fuel efficiency.

The advance of the interference-correlated signal 1312 in terms of the disorder 1310 is greater than that due to the secondary path dynamics of the torque compensation module 1302 and any filter calculations of the digital filter 1304 caused delay. Here, secondary path dynamics relate to the frequency response of the torque compensation module 1302 , The secondary path dynamics include the dynamics of the motor / generator and its mechanical coupling to the crankshaft, which can introduce differences between the commanded torque and the actual torque delivered / absorbed by the motor / generator. Advancing the disturbance-correlated signal with respect to the disturbance can be used to improve the response and the stability of the control system.

One task of the AF-FF control is to attenuate interference in a system of interest, and the AF-FF control solves this problem by generating a noise-canceling signal that counteracts the interference when applied to the system, as in FIG 13 is shown. In this case, the disturbance is the variation of the engine torque 1310 from its mean. A filter output 1314 is through an adaptive digital filter 1304 based on an input interference-correlated signal 1312 generated. The filter output may have an average of zero or substantially zero so as not to change the average total powertrain output torque. In some embodiments, filter output 1314 have a small offset to compensate for inefficiencies in the motor / generator system. Alternatively, the torque requirement may be increased slightly to compensate for these inefficiencies.

The interference correlated signal 1312 contains information regarding the expected disturbance with a certain lead time. The signal 1312 can be based on a firing order. The firing order can be derived using a torque request, a firing fraction, a sigma-delta filter, a lookup table, a state machine, or by any other means. The interference correlated signal 1312 may also include estimated torque signatures related to the firing and skipping of the working chambers in the firing order, as previously described. Including this information will be the convergence times of the weights in the adaptive digital filter 1304 reduce, but is not a requirement.

The filter output 1314 can be in a torque compensation module 1302 can be entered, which is a compensation torque 1316 generated. The torque compensation module 1302 may be a motor / generator coupled to a capacitor, a battery, or any other energy storage device. The torque compensation module 1302 may have response delays or others Have restrictions that cause the generated compensation torque 1316 from the filter output 1314 differs. The compensation torque 1316 becomes with the engine torque 1310 at a summing point 1318 combined. The summing point 1318 gives a delivery torque 1320 to the drive train. The summing point 1318 shows the subtraction of the compensation torque 1316 from the engine torque 1310 , It is understood that the compensation torque may have the opposite polarity in further embodiments and the compensation torque is added to the motor torque. The delivery torque 1320 is used to power the vehicle and becomes the weight update module 1306 fed back as part of a feedback loop. The vibrations of the delivery torque 1320 can at least partially be used in the digital filter 1304 modify weights used.

Adaptive filter parameters called weights can be adjusted by a weight update module 1306 be updated using an adaptive algorithm to reduce the differences between the balancing torque 1316 and the disturbance to a minimum, with the engine torque 1310 is less than the mean to the delivery torque 1320 compensate. The weight update module 1306 determines the appropriate weights according to the procedures described in more detail below. In some embodiments, the difference can be minimized; however, this can consume energy unnecessarily. In some embodiments, the vibrations of the delivery torque 1320 only reduced to a level that delivers acceptable NVH, resulting in optimal fuel efficiency.

Graphs showing the time behavior of the different signals in 13 represent, for better understanding and better explanation of the operation of the exemplary powertrain control 1300 shown. The graph 1311 of engine torque represents vibrations in engine torque output similar to those previously described in 8th were shown. The interference-correlated signal curve 1313 shows an estimated signal of the engine torque disturbance, which is used to minimize variations in delivery torque 1320 is used. This estimate reflects the skip fire nature of the ignitions, hence the digital filter 1304 the necessary frequency component information is supplied, which leads to the filter output 1314 has the correct frequency components. Estimating the disturbance is in the disturbance-correlated signal 1312 not necessary since the signal only has to have the same frequency components as the interference in order to enable convergence; however, an estimate of the perturbation is often desirable as it will reduce the convergence time.

By setting the weights in the digital filter 1304 becomes the regulatory system 1300 the filter output 1314 adaptively control to avoid disturbances in the delivery torque 1320 to reduce or minimize. The filter output curve 1315 represents how the noise correlated signal 1312 through the digital filter 1304 is modified. The filter output 1314 is in the torque compensation module 1302 , which includes a motor / generator or any similar system that can generate and / or absorb torque. The torque compensation module 1302 gives a compensation torque 1316 from that in graph 1317 is shown. If the compensation torque and the motor torque in the summing point 1318 be combined, the resulting delivery torque 1320 relatively small torque variations, as in graph 1321 is shown.

In one embodiment, one or more aspects of the AF-FF control design improve control stability and / or improve the convergence of the adaptive filter weights. For example, the AF-FF control is modified in one embodiment to take advantage of the fact that the firing pattern can be generated by a sigma-delta filter. In particular, the input to the sigma-delta filter represents the requested average engine torque without any delay (in fact, it is advanced relative to the torque delivery). The signal is determined by the torque request 111 represents, optionally in the interference correlated signal generation module 1592 can be entered as in 15 is shown. Since the average torque is known, it can be removed, resulting in an averaged, noise-correlated signal. This facilitates the generation of a balancing or weakening torque that more precisely neutralizes the torque disturbance produced by the engine. Also, including information regarding the torque signatures of ignited and skipped work chambers and engine operating parameters, ie, cam phase, MAC, MAP, ignition timing, etc., will generally improve convergence in the noise-correlated signal. In the 13 shown example curve 1313 provides inclusion of this information in the interference-correlated signal 1312 represents.

An alternative method of removing the bias from a torque oscillation signal is to use a filter that has a low pass nature, pull the bias out, and remove it from that Subtract original signal, effectively forming a high pass filter. A simple conventional filter can be used, but introduces a delay in the filtered signal. However, some filters can be used that have minimal delays. The vibration components of the engine torque are primarily at the fundamental frequency and the harmonics of the ignition frequency. A comb filter, such as an adaptive averaging filter, can be used to remove all harmonics with reduced delay.

In one embodiment, scaling the noise correlated signal is used to reduce the convergence time of the adaptive filter during an engine transition. Scaling the noise correlated signal is arbitrary because the adaptive filter adjusts the amplitude. In order to reduce the convergence time, however, the interference-correlated signal can be scaled to the actual torque for a given operating state, which is obtained from an estimate of the engine parameters that influence the working chamber delivery, such as MAC, MAP, cam phase, ignition timing, etc. This estimate can be done in the torque model module 1592 be performed.

16 10 is a flowchart illustrating one embodiment of a method for adaptive filter feed forward control. The firing order is through the fault correlated signal generation module 1590 receive 1702 , The skip firing sequence is used to generate the interference correlated signal 1704 , The interference-correlated signal is used via AF-FF to determine the compensation torque 1706 ,

$$=w_c\left(k\right)+\left[w_1\left(k\right)w_2\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}{x}_1\left(k\right)\\ x_2\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

$$u\text{'}\left(k\right)=u\left(k\right)-w_c\left(k\right)=\left[w_1\left(k\right)w_2\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}{x}_1\left(k\right)\\ x_2\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

Additional examples of adaptive FIR filters that are described in 13 and 15 used as adaptive digital filters will now be described, according to one embodiment. The purpose of an adaptive FIR filter is to filter the noise correlated signal to adjust the phase and strength to generate a noise neutralization command for the motor / generator that neutralizes or reduces the noise. The adaptive FIR filter helps compensate for secondary path dynamics associated with differences between the motor / generator command and the torque actually applied to the crankshaft. In one embodiment of the present invention, an additional method for separating the bias from the noise neutralization command, called a bias estimate, can be used. The following is an example formulation of the FIR filter with bias estimation. $$u\left(k\right)=\left[w_c\left(k\right)w_1\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}1\\ x_1\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

$$=w_c\left(k\right)+\left[w_1\left(k\right)w_2\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}{x}_1\left(k\right)\\ x_2\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

$$u\text{'}\left(k\right)=u\left(k\right)-w_c\left(k\right)=\left[w_1\left(k\right)w_2\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}{x}_1\left(k\right)\\ x_2\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

The components of the column vector are the last N samples (x (k)) of the noise-correlated signal 1312 and the components of the row vector (w _c (k) and w _i (k)) are the N weights of the FIR filter. Here k is the sampling index. The scanning can be done in the crank angle domain or the time domain, the two domains being related by the known engine speed. The number of weights in the FIR filter can be chosen to provide an acceptable trade-off between filter performance and calculation speed. In practice, the number of weights can range from 5 to 100, with 10 being a value suitable for many motors. In some cases, downsampling can be used to reduce the computational load. More weightings are generally required when damping low frequencies, ~ 2 Hz, is required. The first component of the column vector, 1, is the constant introduced for bias estimation and w _c is the weight for the bias. The signal u (k) is the noise neutralization command before subtracting the bias term w _c (k). The signal u '(k) is the noise neutralization command after subtracting the bias term w _c (k). This can be the filter output signal 1314 be in the torque compensation module 1302 in 13 and 15 is directed.

In one embodiment, the filter output 1314 as in the previous section, by removing w _c (k) from u (k), which removes the bias. The bias component of the filter output is constrained to the bias weight, which can be easily separated from the rest of the filter components. For micro-hybrid applications where the total stored energy available is small, the bias component can be removed from the filter output so that the system's energy storage / release capabilities are not exceeded. Alternatively, a slight bias may be included to compensate for inefficiencies in the energy collection / storage / release system. In other embodiments, the bias term may simply be repacked to mitigate the bias error if the auxiliary power source / sink has sufficient capacity to address the delivery torque bias error. However, in this case the torque must be estimated accurately and the difference between the requested torque and the estimated torque must be determined.

In one embodiment of the present invention, the bias is removed from the filter output and only the signal u '(k) from Eq. 7c is applied to the motor / generator to dampen the frequencies related to the motor torque vibrations. Although it is possible to correct the bias deviation from the requested torque using the motor / generator, in some embodiments it is more practical to limit the corrective action by the motor / generator to the oscillating torque pulse component due to the limited energy storage capacity of the energy storage - / - release device. Another reason for this limitation is the accuracy of the torque estimate. It is often difficult to accurately estimate the bias component due to the variability of combustion dynamics and estimation errors of the various engine parameters, such as MAC. The uncertainty of the load torque also complicates the estimation of the engine torque from the drive train acceleration.

In some cases, it may be desirable due to hardware limitations, the strength or frequency of the filter output 1314 limit that is entered into the motor / generator. This can be done, for example, using a loop shape filter, such as a notch filter, for certain harmonic components from the filter output 1314 to remove or dampen. The same method used for bias removal as described below with reference to Eq. 10 can be applied to these distant or damped harmonics to ensure stability. Flexibility is another advantage of embodiments of the present invention.

wobei A(k) das gemessene oder abgeleitete Motordrehmoment ist, R ein Strafgewichtungsterm ist und u(k) das gefilterte Ausgabebefehlssignal in das Drehmomentausgleichsmodul 1302 ist. Der Term R * u(k)^2 repräsentiert den mit der zum Produzieren des Ausgleichsdrehmoments erforderlichen Energie verknüpften Strafterm. R kann in Abhängigkeit von den in der Motordrehmomentabweichung vorhandenen Frequenzkomponenten und dem akzeptablen Niveau des NVH variieren. Bei manchen Ausführungsformen kann R unter Verwendung einer Nachschlagetabelle bestimmt werden, die als Indices Motordrehzahl, Zündungsanteil und Getriebegang beinhaltet. Minimieren von J(k) maximiert die Kraftstoffeffizienz, während gleichzeitig ein akzeptables NVH-Niveau geliefert wird.There are many optimization methods to minimize fuel consumption while delivering the required torque with acceptable NVH. One method is to minimize a function J such that $$J\left(k\right)=A\left(k\right)2\text{+}R\text{*}u\left(k\right)\text{2}$$

where A (k) is the measured or derived engine torque, R is a penalty weight term and u (k) is the filtered output command signal into the torque balancer 1302 is. The term R * u (k) ^ 2 represents the penalty term associated with the energy required to produce the balancing torque. R can vary depending on the frequency components present in the engine torque deviation and the acceptable level of NVH. In some embodiments, R can be determined using a look-up table that includes engine speed, ignition percentage, and transmission gear as indices. Minimizing J (k) maximizes fuel efficiency while delivering an acceptable NVH level.

$$\left[\begin{array}{c}{w}_c\left(k+1\right)\\ w_1\left(k+1\right)\\ ⋮\\ w_N\left(k+1\right)\end{array}\right]=\left[\begin{array}{c}{w}_c\left(k\right)\\ w_1\left(k\right)\\ ⋮\\ w_N\left(k\right)\end{array}\right]+\lambda \left(A\left(k\right)-Ru\left(k\right)\right)\left[\begin{array}{c}1\\ x_1\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

One of the simplest and most common methods for minimizing J is a Least Mean Squares (LMS) algorithm, the difference equation of which is given below. $$\left[\begin{array}{c}{w}_c\left(k+1\right)\\ w_1\left(k+1\right)\\ ⋮\\ w_N\left(k+1\right)\end{array}\right]=\left[\begin{array}{c}{w}_c\left(k\right)\\ w_1\left(k\right)\\ ⋮\\ w_N\left(k\right)\end{array}\right]+\mathrm{<figure-callout\; id="1"\; label="\lambda "\; filenames="DE112018001304T5\_0033.png,DE112018001304T5\_0040.png"\; state="\{\{state\}\}">\lambda </figure-callout>}\frac{1}{2}\frac{\partial J\left(k\right)}{\partial w\left(k\right)}$$

$$\left[\begin{array}{c}{w}_c\left(k+1\right)\\ w_1\left(k+1\right)\\ ⋮\\ w_N\left(k+1\right)\end{array}\right]=\left[\begin{array}{c}{w}_c\left(k\right)\\ w_1\left(k\right)\\ ⋮\\ w_N\left(k\right)\end{array}\right]+\lambda \left(A\left(k\right)-Ru\left(k\right)\right)\left[\begin{array}{c}1\\ x_1\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

The LMS method minimizes J (k), which is an approximation to the statistical expectation of J. The adaptation gain λ determines the stability and the convergence rate. For the LMS method, there is a trade-off when choosing the value of λ. For large λ, the weights can vibrate with a large amplitude when converged, and for small λ the convergence rate can be slow. Other algorithms, such as a recursive least squares method or RLS method, can deliver faster convergence and a smaller stable state error, but require more calculation.

Hier ist A_m (k) der gemessene Drehmomentfehler (oder Antriebsstrangbeschleunigung mit angemessener Skalierung) und P{} repräsentiert das Filtern des Signals mit einem anhand des Sekundärpfadmodells geschätzten Filters. Die Schätzung der Sekundärpfadantwort auf den Bias-Term,w_c (k)x_c(k), wird von dem Messfehler entfernt, um A(k) zu erhalten, welcher der Gewichtungenaktualisierung zugeführt wird. Das Sekundärpfadmodell muss nicht sehr genau sei, da es nur verwendet wird zum Entfernen der Auswirkung des Bias-Terms von dem gemessenen Fehler und irgendein Modellierfehler P{} von dem tatsächlichen Sekundärpfad wird durch die adaptive Gewichtung w_c kompensiert. Der Zweck dieser Modifikation besteht darin, eine virtuelle Rückkopplungsschleife für den Bias-Term zu bilden, um die Systemstabilität zu verbessern. Ansonsten weist die Bias-Schätzung eine Möglichkeit auf, unbegrenzt anzuwachsen, und hat eine negative Auswirkung auf die Konvergenz anderer Gewichtungen.As previously described in one embodiment of the present invention, the bias can be removed from the noise neutralization command. To improve stability of this embodiment, it may be necessary to make the following modification to the measured error. $$A\left(k\right)=A_m\left(k\right)-P\left\{w_c\left(k\right)x_c\left(k\right)\right\}$$

Here A _m (k) is the measured torque error (or powertrain acceleration with appropriate scaling) and P {} represents filtering the signal with a filter estimated from the secondary path model. The estimate of the secondary path response to the bias term, w _c (k) x _c (k), is removed from the measurement error to obtain A (k) which is applied to the weighting update. The secondary path model need not be very accurate since it is only used to remove the effect of the bias term from the measured error and any modeling error P {} from the actual secondary path is compensated for by the adaptive weighting w _c . The purpose of this modification is to create a virtual feedback loop for the bias term to improve system stability. Otherwise, the bias estimate has an opportunity to grow indefinitely and has a negative impact on the convergence of other weights.

In other embodiments, the noise correlated signal used for weighting updates can be filtered by a model that represents the secondary path dynamics to improve stability. This can be referred to as a Filtered-X implementation. Here the weighting update algorithm module points 1306 an input of the interference-correlated signal 1312 that is filtered by the secondary path dynamics.

For x _c = 1, as in Eq. 9, the change in bias is represented by w _c (k). For a steady state, the average torque is stationary and w _c (k) converges to a constant. In transient operation, it follows the bias changes. For a large torque change, such as a change in the ignition component, it will take time for w _c (k) to converge to the correct value. Although the bias term will not be used in the noise neutralization signal, the convergence of w _c (k) affects the convergence of other weights.

$$\left[\begin{array}{c}{w}_c\left(k+1\right)\\ w_1\left(k+1\right)\\ ⋮\\ w_N\left(k+1\right)\end{array}\right]=\left[\begin{array}{c}{w}_c\left(k\right)\\ w_1\left(k\right)\\ ⋮\\ w_N\left(k\right)\end{array}\right]+\lambda \left(A\left(k\right)-Ru\left(k\right)\right)\left[\begin{array}{c}{x}_c\left(k\right)\\ x_1\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

To improve convergence, one embodiment of the present invention adjusts the value of x _c (k) based on the estimated average engine torque instead of using a constant value for x _c (k). Assuming that x _c (k) is approximately proportional to the change in the bias term during a transient, using x _c (k) reduces the convergence time. The modified FIR filter and weight update formulas are given below. $$u\left(k\right)=w_c\left(k\right)x_c\left(k\right)+\left[w_1\left(k\right)w_2\left(k\right)\cdots w_N\left(k\right)\right]\left[\begin{array}{c}{x}_1\left(k\right)\\ x_2\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

$$\left[\begin{array}{c}{w}_c\left(k+1\right)\\ w_1\left(k+1\right)\\ ⋮\\ w_N\left(k+1\right)\end{array}\right]=\left[\begin{array}{c}{w}_c\left(k\right)\\ w_1\left(k\right)\\ ⋮\\ w_N\left(k\right)\end{array}\right]+\lambda \left(A\left(k\right)-Ru\left(k\right)\right)\left[\begin{array}{c}{x}_c\left(k\right)\\ x_1\left(k\right)\\ ⋮\\ x_N\left(k\right)\end{array}\right]$$

The bias term is removed from the interference-correlated signal, but is added as x _c (k) here in Equation 12. This is necessary in order to explicitly separate the slow bias from the combustion pulses at a higher frequency, which is important for managing the frequency components concerned.

17A is a schematic diagram of a control system using a Filtered-X implementation with which in Eq. 10 described modification to the feedback term, according to an embodiment of the current invention. Entries in the filter module 1340 can use the interference-correlated signal 1312 , the delivery torque 1320 and the estimated average engine torque 1344 include. As discussed previously, other perturbation parameters, such as crankshaft acceleration, can be used in place of the delivery torque. The interference correlated signal 1312 becomes a secondary path dynamics module 1334a entered. The secondary path dynamics module 1334a filters the interference-correlated signal 1312 to get a filtered noise-correlated signal 1342 to result in the weight update module 1306 is entered. The estimated average engine torque 1344 is also in the weight update module 1306 entered that the requested torque 111 may be or may be derived from the requested engine torque and any torque model. Another in the weight update module 1306 input signal can be the modified delivery torque signal 1338 that represents the delivery torque modified by removing one or more selected components through the virtual feedback. The modified delivery torque signal 1338 can be determined by the difference between the delivery torque 1320 and the correction term 1336 is taken what the impact of removed components through the loop shape filter 1324 represented on the delivery torque. The weight update module 1306 may use a least squares algorithm, a recursive least squares algorithm, or any other algorithm to determine appropriate weights. The weight update module 1306 can updated weights 1326 output that through the digital filter 1304 be used. The digital filter filters the interference-correlated signal 1312 to a fault neutralization command without a bias signal 1330 to create. The fault neutralization command without bias 1330 can through the digital filter 1304 are output and to a loop shape filter 1324 be directed. The output of the loop shape filter 1324 can use a filtered fault neutralization command 1314 output that is sent to the torque compensation module. The weight update module can also do a bias estimate 1328 output that represents the estimated average torque value. The bias estimate 1328 can output the fault neutralization command without a bias signal 1330 can be added to a fault neutralization command with a bias signal 1332 to obtain. The difference between the bias neutralization command 1332 and the filtered fault neutralization command 1314 can in the summing node 1346 be determined. The output of the summing node 1346 is 1348 by the filtered fault neutralization command 1314 removed component into a secondary path dynamics module 1334b can be entered. The secondary path dynamics module 1334b gives a correction term 1336 from which the signal 1348 filters based on the secondary path dynamics model. The output of the secondary path dynamics module 1334b is the correction term 1336 which represents the noise neutralization command modified by the secondary path dynamics and the loop shape filter. The correction signal 1336 can from the delivery torque 1320 at the summing node 1350 which is the difference of the two signals 1338 returns to be subtracted. The signal 1338 can be in the weight update module 1306 can be entered where it is based on the delivery torque 1320 provides a modified feedback to help reduce NVH to an acceptable level.

It is understood that the filter module 1340 can be implemented in a number of ways. This in 17A shown filter module 1340 works in the time domain. Filters with similar filter characteristics can also be implemented in the frequency domain. The filter 1340 has an input signal 1320 that represents the delivery torque. It is possible that the filter has no feedback signal and only the interference-correlated signal 1312 used to generate the balancing torque. In this case, the interference-correlated signal 1312 at least partially on the torque model 1592 because there is no feedback to help reduce torque disturbances.

In one embodiment, a controller can be switched between online and offline adaptation. For example, the weights can be updated adaptively online, i. H. as in Eq. 9, can be calculated recursively in real time. Alternatively, pre-calculated values can be stored offline in a look-up table that is used by the controller at various operating points when the control algorithm is run.

Adjusted basis function control

Another type of adaptive control that can be used in a hybrid powertrain control is a custom basis function control. With adjusted basic function control, a control signal is expressed as one or more functions of time, for example trigonometric functions, such as sine and cosine functions. Arguments of the trigonometric functions are chosen to match one or more frequencies expected in the perturbation, i. H. the trigonometric functions are adapted to the disturbance. The use of adapted basic function control is particularly advantageous if an engine is operating with a fixed ignition component, since in this case the periodicity of the ignition sequence is fixed and the frequency components of interest are well-defined.

17B shows part of a hybrid powertrain 1700 that uses a custom basis function control. This method is similar to that previously described, but instead of using a fault correlated signal to control the motor / generator, an adapted base function is used to determine the motor / generator control signal. The adjusted base function can be any function of time and the weakening or balancing torque can be created from the base functions. In one embodiment, the base function is composed of cosine and sine function pairs from one or more selected frequencies. In general, the frequencies of a fundamental frequency that matches the periodicity of the firing order and harmonics will correspond to that frequency. The phase and strength of the adjusted basis function can be adaptively adjusted by a well-known algorithm, such as a recursive least squares (RLS) algorithm, an average square deviation (LMS) algorithm or any other algorithm to minimize a constraint objective function. The target function may be fuel efficiency and the constraint may be an NVH metric, such as the root mean square of the estimated driver seat rail acceleration or crankshaft acceleration. The strength of the mitigation is adjusted through a weighting factor that determines the relationship of the control action and the NVH metric in a target function, which ensures that the control effort is minimized while the NVH constraint is met.

Different aspects of 17B are those in 15 similar. As previously described, the control pulses 113 by an ignition timing module based on an ignition percentage 119 certainly. These two signals, control pulse signal 113 and the ignition component 119 , can be in an angle generation module 1780 be entered, the part of a basic function generator module 1740 is. The ignition component indicates the length of the repetitive firing sequence length in steady-state operation. The control impulses 113 indicate whether an ignition opportunity should be ignited or missed. The control impulses 113 indicate the phase of the repeating firing order. The control impulses 113 also become an ignition control unit 140 entered the operation of the engine 150 orchestrated. The motor 150 gives a crank angle signal 1710 that specifies the crank angle orientation, which is also in the angle generation module 1780 is entered. Operation of the angle generation module 1780 is described in detail below.

The base pair module 1740 generates a pair of sine and cosine functions on one or more in the control pulses 113 existing frequencies. The outputs of the sine and cosine pair generation module 1740 are an input basis function 1355 and an output basis function 1357 , The input basis function 1355 and the output basis function 1357 can be vectors composed of one or more pairs of sine and cosine functions. The input basis function 1355 and the output basis function 1357 have similar dependencies, but can differ by a phase angle and / or the relative strength of the sine and cosine functions. These differences arise from the secondary dynamics. The input basis function 1355 becomes a multiplier 1360 entered. The output base function 1357 is converted into a matched basis function coefficient module 1750 entered.

Based on the input from the ignition control unit 140 the engine generates 150 an engine torque 1310 , The engine torque 1310 is mechanically in an adder 1318 with the issue 1316 of the torque compensation module 1302 combined. The resulting delivery torque 1320 can be used to supply the vehicle with power and can drive any vehicle auxiliary units. The delivery torque 1320 or a signal related to this is placed in a vehicle response filter module 1722 and an NVH estimation module 1720 feedback that determine an NVH metric based on vehicle response and vehicle occupant sensitivity to NVH. The NVH metric signal 1380 is in the adjusted basis function coefficient module 1750 entered. The adjusted basis function coefficient module 1750 determines coefficients 1390 with the corresponding sine or cosine function in the input basis function 1355 in the multiplier 1360 be multiplied. The output is a motor / generator control signal 1301 that in the torque compensation module 1302 is entered to the compensation torque 1316 to deliver. The sine and cosine pair generation module 1740 , the multiplier 1360 and the matched basis function coefficient module 1750 can all be in a customized basic function control module 1730 be installed.

Bode diagrams showing the sample filter responses for the strength and phase of the vehicle response filter module 1722 and the NVH filter module 1720 represent are in 17C and 17D shown. 17C shows a graph 1760 , which is an exemplary strength of the vehicle response 1762 and the NVH response 1764 represents. The vehicle response can be measured in the driver's seat or anywhere else in the vehicle interior. The vehicle response strength 1762 can have one or more resonance peaks 1766 have powertrain resonance frequencies or any other vehicle resonance. The NVH filter response 1764 is controlled by human perception of shock and vibration. There have been a number of studies involving the effects of vibration on vehicle occupants. ISO 2631, for example, provides guidance on the influence of vibration on vehicle occupants. Generally, as in 17C vibration, from the perspective of passenger comfort, at frequencies between 0.2 and 8 Hz is considered to be among the worst types of vibration (of course there are a number of competing theories on the most relevant limits). Therefore, these frequency components are preferred by the NVH filter 1720 let through so that they can be dampened by applying a compensating torque.

17D shows a graph 1770 which is an exemplary phase of vehicle response 1772 and the NVH response 1774 represents. The characteristics of the vehicle response filter module 1722 may vary based on the gearbox or the effective gear ratio for a continuously variable transmission. Other factors, such as ambient temperature, can also modify the characteristics of the vehicle response filter. That means that each in 17C and 17D shown graphs of vehicle response strength 1762 and phase 1772 vary depending on the transmission gear and possibly other factors.

17E Figure 3 is a block diagram showing an angle generation module 1780 represents, according to an embodiment of the present invention. The purpose of the angle generation module 1780 is the crank angle signal 1710 to unfold and generate a periodic function that has a periodicity that matches the firing order. The crank angle signal 1710 can vary between 0 and 720 degrees. It can be in 6-degree increments, but this is not a requirement. The crank angle signal 1710 is split and part of the signal goes through a delay element 1781 , The difference between the crank angle signal 1710 and the delayed signal 1782 is through the adder 1783 determined which is a comparison signal 1784 produced. The comparison signal 1784 is generally a small positive value because the crank angle increases over time. An exception occurs when the crank angle signal 1710 720 crosses and is reset to zero, in which case the comparison signal becomes approximately equal to a negative 720 degrees.

The crank angle signal 1710 is also in the comparator 1785 entered. The comparator 1785 compares the current version of the crank angle signal 1710 with its previous value. If the current crank angle signal is larger, the comparator output is 1797 zero. If it is smaller, then the comparator output is one. The comparator output signal 1797 gets into the multiplier 1787 entered. The multiplier 1787 multiplies the input signals 1786 and 1797 , Because the comparator signal 1797 is usually zero, is the output of the multiplier 1787 usually zero. If the crank angle signal 1710 being folded back to zero, or near zero, the comparator output signal 1797 to one and the output of the multiplier 1787 becomes 720 , The output of the multiplier 1787 is in the adder 1788 with the comparison signal 1784 summed. This output becomes the crank angle difference between successive measurements, for example 6 degrees. The output of the adder 1788 is in the reinforcing element 1789 entered. The size of the reinforcement in the reinforcement element 1789 is determined by the number of ignition opportunities per crankshaft revolution. A four-cylinder four-stroke engine would have a gain of 2. The output of the reinforcement element 1789 is a set crank angle increment 1790 that in the multiplier 1791 is entered.

The other input into the multiplier 1791 is the ignition component 119 , For Skip Fire controlled engines operating in a steady state, the denominator of the ignition portion is the length of the periodic firing order. Multiply the crank angle increment 1790 in the multiplier 1791 is equivalent to dividing it by the length of the repeating firing order. The output of the multiplier 1791 is in the adder 1792 entered. The adder 1792 sums up the output 1779 with a signal 1778 , The signal 1778 is from the delay block 1795 output the output as an input 1796 a modulo unit 1794 having. The entries in the modulo unit 1794 are the adder output 1793 and 360 which is the modulo argument. The collective elements in block 1793 are used to integrate the divider output signal 1779 and return a value between 0 and 360. The edition 1796 can be used as the argument in the trigonometric functions that define the fitted base function. The functions are periodic with a period length that matches the length of the repeating firing order length.

To ensure the stability of the algorithm, some information about the secondary path may be necessary, such as belt transmission and motor / generator dynamics. An accurate model of the secondary path is not required because the actual phase tracking is adaptively performed using the measured feedback signal to minimize the defined error that compensates for any uncertainty in the secondary path model.

In contrast to some of the previously described embodiments, matched basis function control addresses a finite number of dominant harmonics in the engine torque profile, rather than addressing all frequencies within a certain bandwidth. Similar to the previously described methods, it uses feedback to adaptively modify parameters in the control system to reduce NVH to an acceptable level while delivering maximum fuel efficiency. Custom basis function control may use a pair of sine and cosine signals or other orthogonal functions that may represent engine torque pulses generated by DSF operation. The trigonometric function can advantageously be generated directly on the basis of a crank angle measurement in order to avoid any delay associated with filtering.

Hierbei ist N_rpm die in Upm ausgedrückte Motordrehzahl, ist N_cyl die Anzahl der Zylinder in dem Motor, ist N_Denom der Nenner eines einfachen Bruchs und ist f die Grundfrequenz des Motordrehmomentimpulses. Alle Werte sind ganze Zahlen und die Grundfrequenz ist für Zünddichten, die denselben Nenner aufweisen, dieselbe. Harmonische der Grundfrequenz sind ganzzahlige Vielfache der Frequenz f.The relationship between the engine configuration and the ignition frequency of a four-stroke engine is given by the following formula. $$f=\frac{N_{rpm}}{60}\frac{N_{\mathrm{<figure-callout\; id="2"\; label="c\; y\; l"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">c\; </figure-callout>}yl}}{2}\frac{1}{N_{DenOm}}\left(Hz\right)$$

Here, N _{rpm is} the engine speed expressed in rpm, N _{cyl is} the number of cylinders in the engine, N _{denom is} the denominator of a simple fraction, and f is the fundamental frequency of the engine torque _pulse . All values are integers and the fundamental frequency is the same for ignition densities that have the same denominator. Harmonics of the fundamental frequency are integer multiples of the frequency f.

To attenuate such harmonics, it is necessary to generate sinusoidal torque of the corresponding frequencies, and this can be done using trigonometric functions that oscillate at this angular velocity 2πf (rad). However, a measured engine speed is very noisy and some type of filter is needed to remove high frequency noise from the engine speed measurements. The disadvantage of this approach is that a filter introduces a delay into the estimated angular velocity trajectory, which leads to an attenuation error.

Üblicherweise wird der Kurbelwinkel in 0 - 720 Grad gefaltet und muss vor Anwenden der obigen Formel entfaltet werden. Ein beispielhaftes Modul zum Durchführen dieses Entfaltens wurde in Verbindung mit 17E und dem Winkelerzeugungsmodul 1780 beschrieben.Since what is really needed is not the frequency, but the argument of the trigonometric functions, ie an integral of the angular velocity, a crank angle measurement can be used instead, since the integral required is an integer multiple of the crank angle. A crank angle measurement is generally less noisy and no delay is introduced because it does not require filtering. The following formula, Eq. 14 shows the equivalent sine function, the arguments of which are obtained from the angular velocity (left side of the equation) and the crank angle (right side of the equation). $$sin\left({\displaystyle {\int }_{r=0}^t\omega \left(\tau \right)d\tau }\right)=\text{sin}\left(\frac{N_{\mathrm{<figure-callout\; id="2"\; label="c\; y\; l"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">c\; </figure-callout>}yl}}{2}\frac{1}{N_{DenOm}}\phi \left(t\right)\right).\phi \left(t\right):{}_{Kurbelwinkel}$$

Usually the crank angle is folded in 0 - 720 degrees and has to be unfolded before applying the above formula. An exemplary module for performing this deployment was in connection with 17E and the angle generation module 1780 described.

In Gl. 15 und in anderen Gleichungen bezieht sich der Exponent T auf die Transponierte der Ausgabebasisfunktion, H(k), und wird dazu verwendet, den Ausdruck von H(k) kompakter zu machen. In Gl. 15 weist die Basisfunktion nur eine Frequenz auf, die Grundschwingungsfrequenz; allerdings können andere Terme zu der Basisfunktion für andere Frequenzen von Interesse, wie etwa unter anderem 2ω und 3ω, hinzugefügt werden. Die Ausgabebasisfunktion entspricht dem Signal 1357 in 17E.In one embodiment of the matched basis function control, a pair of sine and cosine functions are used as the basis function, the frequency of which corresponds to the started frequency. An output basis function, H (k), can be defined as follows. $$H\left(k\right)={\left[\text{cos}\left(\omega kT\right)\text{sin}\left(\omega kT\right)\right]}^T$$

In Eq. 15 and in other equations, the exponent T refers to the transpose of the output basis function, H (k), and is used to make the expression of H (k) more compact. In Eq. 15, the basic function has only one frequency, the fundamental frequency; however, other terms may be added to the base function for other frequencies of interest, such as 2ω and 3ω, among others. The output base function corresponds to the signal 1357 in 17E ,

hierbei sind θ und ρ die Phase und die Stärke der Frequenzantwort einer Übertragungsfunktion F, welche jegliche Dynamik zwischen dem Drehmomentabschwächungsbefehl und der gemessenen NVH-Metrik repräsentiert, was eine Zwangsbedingung liefert. Die Eingabebasisfunktion entspricht dem Signal 1355 in 17E.An input basis function, G (k), which includes compensation for secondary path dynamics, can be defined as follows: $$G\left(k\right)={\left[\frac{1}{\rho \left(\omega \right)}\text{cos}\left(\omega kT-\theta \left(\omega \right)\right)\frac{1}{\rho \left(\omega \right)}\text{sin}\left(wkT-\theta \left(\omega \right)\right)\right]}^T$$

where θ and ρ are the phase and strength of the frequency response of a transfer function F, which represents any dynamics between the torque weakening command and the measured NVH metric, which provides a constraint. The input base function corresponds to the signal 1355 in 17E ,

In some embodiments, the input basis function can be obtained by filtering the basis function with a transfer function F. A disadvantage of this approach is that it results in some phase and power lag. Instead, the input basis function in Eq. 16 are obtained by the phase and strength of the frequency response of F at the frequency being approached, which can be provided as a table or as a formula. The transfer function F can be expressed in several forms, as in Eq. 17 is shown. $$\left[\begin{array}{c}\rho \left(\omega \right)\text{cos}\left(\omega kT+\theta \left(\omega \right)\right)\\ \rho \left(\omega \right)\text{sin}\left(\omega kT+\theta \left(\omega \right)\right)\end{array}\right]=F\left\{\left[\begin{array}{c}\text{cos}\left(\omega kt\right)\\ \text{sin}\left(wkt\right)\end{array}\right]\right\}=\rho \left(\omega \right)\left[\begin{array}{c}\text{cos}\left(\theta \left(\omega \right)\right)-\text{sin}\left(\theta \left(\omega \right)\right)\\ \text{sin}\left(\theta \left(\omega \right)\right)-\text{cos}\left(\theta \left(\omega \right)\right)\end{array}\right]\left[\begin{array}{c}\text{cos}\left(wkT\right)\\ \text{sin}\left(\omega kT\right)\end{array}\right]$$

The transfer function F contains the secondary path dynamics and filters with regard to NVH metrics and vehicle response. Since the effect of the transfer function F will be identified and compensated for implicitly by an adaptive algorithm, the knowledge of the transfer function need not be very precise as long as it is accurate enough to make the system stable. The tolerances for the transfer function F are very lax.

In some cases, the input basis function can be obtained in the manner shown below. $$\frac{1}{\rho }\left[\frac{\text{cos}\left(\omega kT-\theta \right)}{\text{sin}\left(\omega kT-\theta \right)}\right]=\frac{1}{\rho }\left[\begin{array}{c}\text{cos}\left(\theta \right)\text{sin}\left(\theta \right)\\ -\text{sin}\left(\theta \right)\text{cos}\left(\theta \right)\end{array}\right]\left[\begin{array}{c}\text{cos}\left(\omega kT\right)\\ \text{sin}\left(\omega kT\right)\end{array}\right]$$

The derivative shown above has the advantage of not introducing deceleration because it uses the measured crank angle directly.

In operation of the adjusted basis function control algorithm, the input basis function G (k) is transformed from the output basis function H (k) by the strength and phase estimation of the transfer function F for the given frequency. The difference between the true strength and phase of the transfer function F is adaptively compensated by the control algorithm.

Die Koeffizienten α(n) können durch einen adaptiven Algorithmus bestimmt werden, wie etwa ein rekursives Kleinste-Quadrate(RLS)-Verfahren oder Quadratischer Mittelwert (LMS oder Gradientenalgorithmus), welcher eine Zielfunktion J minimiert. Die Zielfunktion J für RLS kann ausgedrückt werden als $$\begin{array}{l}J=\frac{1}{2}{e}_{NVH}^T{e}_{NVH}+\frac{1}{2}{u}_c^{T}R{u}_c=\frac{1}{2}{\left(y_{NVH}-T\text{'}\alpha \right)}^T\left(y_{NVH}-T\text{'}\alpha \right)+\frac{1}{2}{\alpha }^T{T}^{T}RT\alpha \\ u_c={\left[h{\left(n\right)}^T\alpha h{\left(n-1\right)}^T\alpha \cdots h{\left(n-N+1\right)}^T\alpha \right]}^T\\ y_{NVH}={\left[h\text{'}{\left(n\right)}^T{w}_{0}h{\left(n-1\right)}^T{w}_0\cdots h{\left(n-N+1\right)}^T{w}_0\right]}^T\\ h\left(n\right)={\left[\text{cos}\left(n\omega T\right)\text{ sin}\left(n\omega T\right)\text{ cos}\left(2n\omega T\right)\dots \right]}^T\text{, }T={\left[h\left(n\right)h\left(n-1\right)\dots h\left(n-N+1\right)\right]}^T\\ h\text{'}\left(n\right)={\left[q_1^c{g}_1^c\text{cos}\left(n\omega T\right)q_1^s{g}_1^s\text{ sin}\left(n\omega T\right)q_2^c{g}_2^c\text{cos}\left(2n\omega T\right)\dots \right]}^T,T\text{'}={\left[h\text{'}\left(n\right)h\text{'}\left(n-1\right)\dots h\text{'}\left(n-N+1\right)\right]}^T\\ R=r^{2}I\end{array}$$

wobei e_NVH und u_c die Vorgeschichte der NVH-Metrik und des Steuerbefehls ist, w₀ und α die entsprechenden Basisfunktionskoeffizienten sind, I die Einheitsmatrix ist und r² eine Gewichtung ist zum Anpassen hinsichtlich der Strafe für eine Regelaktion, in diesem Falle derKraftstoffverbrauch. $g_m^c,g_m^s,q_m^c{}_{und}{q}_m^s$

sind jeweils die Projektion der Frequenzantwort der Sekundärpfad-Übertragungsfunktion und der NVH-Metrik/Fahrzeugantwort-Übertragungsfunktion bei der m^th Harmonischen auf die Basisfunktion.The input basis functions G (k) can be multiplied by an input basis coefficient vector α (n) to produce an attenuation command u (k), as in Eq. 19 is shown. $$u\left(k\right)=G^T\left(k\right)a\left(k\right)$$

The coefficients α (n) can be determined by an adaptive algorithm, such as a recursive least squares (RLS) method or quadratic mean (LMS or gradient algorithm), which minimizes an objective function J. The objective function J for RLS can be expressed as $$\begin{array}{l}J=\frac{1}{2}{e}_{NVH}^T{e}_{NVH}+\frac{1}{2}{u}_c^{T}R{u}_c=\frac{1}{2}{\left(y_{NVH}-T\text{'}\alpha \right)}^T\left(y_{NVH}-T\text{'}\alpha \right)+\frac{1}{2}{\alpha }^T{T}^{T}RT\alpha \\ u_c={\left[H{\left(n\right)}^T\alpha H{\left(n-1\right)}^T\alpha \cdots H{\left(n-N+1\right)}^T\alpha \right]}^T\\ y_{NVH}={\left[H\text{'}{\left(n\right)}^T{w}_{0}H{\left(n-1\right)}^{\mathrm{<figure-callout\; id="0"\; label="T\; w"\; filenames="DE112018001304T5\_0035.png,DE112018001304T5\_0036.png"\; state="\{\{state\}\}">T\; </figure-callout>}}{w}_{}{}_0\cdots H{\left(n-N+1\right)}^T{w}_0\right]}^T\\ H\left(n\right)={\left[\text{cos}\left(n\omega T\right)\text{sin}\left(n\omega T\right)\text{cos}\left(2n\omega T\right)...\right]}^T\text{.}T={\left[H\left(n\right)H\left(n-1\right)...H\left(n-N+1\right)\right]}^T\\ H\text{'}\left(n\right)={\left[q_1^c{G}_1^c\text{cos}\left(n\omega T\right)q_1^s{G}_1^s\text{sin}\left(n\omega T\right)q_2^c{G}_2^c\text{cos}\left(2n\omega T\right)...\right]}^T.T\text{'}={\left[H\text{'}\left(n\right)H\text{'}\left(n-1\right)...H\text{'}\left(n-N+1\right)\right]}^T\\ R=r^{2}I\end{array}$$

where e _NVH and u _{c is} the history of the NVH metric and control command, w ₀ and α are the corresponding basic function coefficients, I is the unit matrix and r ^{2 is} a weight to adjust for the penalty for a control action, in this case the fuel consumption. $G_m^c.G_m^s.q_m^c{}_{und}{q}_m^s$

are the projections of the frequency response of the secondary path transfer function and the NVH metric / vehicle response transfer function at the m ^th harmonic onto the base function.

$$e_{NVH}\left[n\right]=F_{NVH}\left\{y\left[n\right]-u\left[n\right]\right\}$$

wobei y[n] das Motordrehmoment ist, u[n] das Abschwächungslieferdrehmoment mit irgendeinem Fehler aufgrund von Unsicherheit im Sekundärpfad ist, λ ein Vergessenfaktor ist, r² wie zuvor der Gewichtungsfaktor ist, F_NVH die Übertragungsfunktion vom Motordrehmoment zur NVH-Metrik ist und e_NVH [n] die gemessene NVH-Metrik ist.The RLS update formula for the coefficient vector α [n] can be given as follows: $$\begin{array}{l}\alpha \left[n\right]=\alpha \left[n-1\right]+P\left[n\right]\left(H\text{'}\left[n\right]e\left[n\right]-r^{2}H\left[n\right]H{\left[n\right]}^T\alpha \left[n-1\right]\right)\\ P\left[0\right]={\left(T\text{'}{\left[0\right]}^{T}T\text{'}\left[0\right]+r^{2}T{\left[0\right]}^{T}T\left[0\right]\right)}^{-1}\\ P\left[n\right]=\left(\lambda P{\left[n-1\right]}^{-1}+r^{2}H\text{'}\left[n\right]H\text{'}{\left[n\right]}^T+r^{2}H\left[n\right]H{\left[n\right]}^T\right)\end{array}$$

$$e_{NVH}\left[n\right]=F_{NVH}\left\{y\left[n\right]-u\left[n\right]\right\}$$

where y [n] is the engine torque, u [n] is the attenuation delivery torque with some error due to uncertainty in the secondary path, λ is a forgetting factor, r ^{2 is} the weighting factor as before, F _{NVH is} the transfer function from the engine torque to the NVH metric and e _NVH [n] is the measured NVH metric.

sind die Projektion des Motordrehmomentimpulses auf die Kosinus- und Sinusbasisfunktion der m^th angegangenen Frequenz und der Komponente des Vektors w₀. $$\frac{1}{2}{\displaystyle \sum _{m=1}^M{w}_m^{c^2}{q}_m^{c^2}}{\left(\frac{r^2}{g_m^{c^2}{q}_m^{c^2}+r^2}\right)}^2+w_m^{s^2}{q}_m^{s^2}{\left(\frac{r^2}{g_m^{s^2}{q}_m^{s^2}+r^2}\right)}^2={\sigma }_{NVH}^2$$

By including the penalty for regulatory action in the objective function, the strength of the regulatory effort can be controlled. The weighting value can be determined by solving the equation below for r, with the desired NVH metric expressed as σ _NVH . $w_m^{c}und{w}_m^s$

are the Projection of the engine torque pulse onto the cosine and sine basis function of the m ^th started frequency and the component of the vector w ₀ . $$\frac{1}{2}{\displaystyle \underset{m=1}{\overset{M}{\Sigma }}{w}_m^{c^2}{q}_m^{c^2}}{\left(\frac{r^2}{G_m^{c^2}{\mathrm{<figure-callout\; id="2"\; label="q\; m\; c"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">q\; </figure-callout>}}_{m{c}^{}}^{{}^2}+r^2}\right)}^2+w_m^{s^2}{q}_m^{s^2}{\left(\frac{r^2}{G_m^{s^2}{q}_m^{s^2}+r^2}\right)}^2={\sigma }_{\mathrm{<figure-callout\; id="2"\; label="N\; V\; H"\; filenames="DE112018001304T5\_0037.png"\; state="\{\{state\}\}">N\; </figure-callout>}VH}^2$$

In some embodiments, powertrain control may switch between an adaptive filter forward algorithm and an adjusted basis function control algorithm, depending on whether the vehicle is operating with a fixed firing order or whether the vehicle is in transition between two firing order levels. With ignition share transitions, the ignition pattern no longer consists of a repeating ignition sequence; however, the length of the transition, ie the number of firing opportunities in the transition, may be known and this may be an input to the custom basis function control module 1730 be used.

Although an embodiment of a matched basic function controller and control algorithm has been described above, it will be understood that the controller and control method can be reconfigured while maintaining their basic functionality.

Results

18A-D shows a simulation result of an embodiment of the present invention. 18A shows the firing fraction command starting at 0.33 and moving to 0.25 over a period of about 20 ms. 18B shows two curves. The curve 2105 is the engine torque and the curve 2120 is the average motor torque, which is the bias level that was removed when determining the noise correlated signal. Initially there is an engine ignition approximately every 30 ms. Since the ignition component is 0.33, this means that there is an ignition opportunity approximately every 10 ms. This exemplary engine torque output may come from a four-cylinder engine operating at 3000 rpm, an eight-cylinder engine operating at 1500 rpm, or any other combination of number of cylinders and engine speed that gives ignition opportunities with this interval. It should be noted that the change in engine ignition density as a result of the ignition share transition slightly after 0.1 s or about 60 ms after the change in the in 18A Ignition share command signal shown begins to occur. This delay, in this case about 6 firing opportunities, represents the associated delay between determining an ignition sequence based on an ignition percentage and executing that sequence in the engine. This level of deceleration is typical of engines that control valve movement via a rotating cam and deactivate omitted cylinders by disabling valve movement. 18C shows a curve 2140 which is the motor / generator command curve and a curve 2150 , which is the torque that is supplied / drawn by the motor / generator. The motor / generator command curve 2150 is an example of a filter output 1314 of 13 or the motor / generator command signal 1301 of 15 , The difference between the curves 2140 and 2150 reflects the impact of the previously discussed secondary path dynamics. 18D shows the total powertrain torque as a curve 2110 and the average total torque as a curve 2120 , Curve 2110 is an example of the delivery torque 1320 , shown in 13 . 15 . 17A and 17B ,

19 shows the frequency spectrum of the torque error versus a simulation result. Curve 2205 represents the filtered engine crankshaft acceleration without applying any compensating torque. Curve 2210 illustrates the combination of the motor and motor / generator responses using an adaptive filter feed forward method as described herein. The torque error for the fundamental and second harmonic frequencies is almost eliminated, as by the inserts in FIG 19 will be shown. The third harmonic is not significantly attenuated, at least in part due to bandwidth limitations of the motor / generator system; however, the frequency components of ~ 100 Hz and above generally have little effect on NVH, so damping is not required.

Dynamic ignition level modulation

The description so far has focused on control methods for a torque compensation system applicable to an internal combustion engine controlled by Skip Fire. It is understood that the control methods described here are also applicable to engines that work with different cylinders with different outputs, as in U.S. Patent 9,399,964 and the US patent application 14/705, 712 , both of which are incorporated in their entirety by reference. Both of these references describe engine control methods in which, in some cases, rather than skipping some cylinders and producing no net torque, some cylinders have a delivery that is significantly different from that of other cylinders. A high cylinder delivery may correspond to a stroke optimized for performance and a low cylinder delivery may correspond to a stroke optimized for fuel efficiency. These high torque and low torque strokes can be arranged in varying sequences to provide the desired engine output. The valves can be controlled in such a way that the ignitions with a higher torque output have a greater amount of air introduced than ignition with a lower torque output. In some cases, omissions can also be added to the pattern so that any given ignition opportunity can be a high level ignition, a low level ignition, or an omission. Collectively, these types of engine control systems can be referred to as dynamic firing level modulation, in which delivery of any ignition opportunity may differ significantly from delivery to adjacent ignition opportunities. When operating with dynamic ignition level modulation, a finite number of effective ignition components, periodic patterns or sequences can be used in the steady state and each effective ignition component, pattern or sequence has a fundamental frequency and its harmonics. The effective firing fraction may be based in part on the firing fraction and the ratio fraction, which is the fraction of low torque firings to high torque firings. It gives an indication of the proportion of the maximum available engine displacement that is used.

For engines with dynamic ignition level modulation, which can have three or more different working chamber outputs, the ignition sequence length can differ from the ignition component. Thus, instead of a single firing component input into the basic function generator 1740 two inputs may be required to determine the repetitive firing sequence length, such as the ratio of skip to firing and the ratio of firing to firing.

In some embodiments, powertrain control for use in operating control of an internal combustion engine having multiple working chambers to provide a desired output is described. Each working chamber has at least one inlet valve that is cam-operated and at least one outlet valve. The powertrain includes an auxiliary torque source / sink. An ignition level determination unit identifies selected low work cycles to be fired on a low torque delivery and selected high work cycles to be fired on a high torque delivery. The ignition level determination unit may be configured to make the high or low torque output determinations based on ignition opportunity to ignition opportunity during operation of the engine. The adaptive filter feedforward controller is configured to determine a compensation torque that is applied to a drive train by the auxiliary torque source / sink, the compensation torque being designed to at least partially cancel out a torque variation generated by the high / low firing order, thereby reducing the NVH, otherwise would be generated by the firing order. In some cases, some ignition opportunity may be missed, in addition to the possibility of high or low torque delivery.

Other characteristics

A number of controls / regulators and control elements have been described in the foregoing patent and included applications, including, for example, powertrain controls, engine controls, engine control units / units (ECU), hybrid powertrain controls, skip fire engine control systems, ignition controls, adaptive filter feedforward controllers, customized basic functions- Regulators, ignition control units, engine / generator controls, feedforward systems and other control elements. The various controls described here and control methods can be implemented using software or firmware running on a processing unit, such as a microprocessor, using programmable logic, using application specific integrated circuits (ASICs), using discrete logic, etc and / or using a combination of the foregoing. These elements can be implemented, grouped and laid out in different architectures in different embodiments of the present invention. For example, in some embodiments, a hybrid powertrain controller or an ECU that consists of a single processor may be configured to perform all of the described activities related to vehicle control. Alternatively, a powertrain or engine control may include multiple processors that are placed together as part of a powertrain or engine control module or that are distributed at different locations within the vehicle. The specific functionalities provided by any of the processors or any of the Control units are carried out can be widely varied. For example, the engine / generator controller may be integrated into an ECU or hybrid powertrain controller, or it may be a separate element that is placed adjacent to the engine / generator. In another example, components such as the firing fraction calculation device, the ignition timing determination module, the adaptive filter feedforward controller, and other components may be functional blocks of an engine or powertrain control unit, as parts of a skip fire or ignition level modulation controller, as part of an NVH mitigation module, as discrete control unit components, or on other suitable ways, including any suitable combination thereof, may be implemented.

The control methods described herein can be implemented using software and firmware that executes in an engine control unit, a powertrain control module, an engine control module, or any suitable processor that is programmed with appropriate control algorithms. Alternatively, if desired, the functionality can be implemented in the form of programmable logic or using application specific integrated circuits (ASICs) or a combination of any of the foregoing.

In one embodiment, a method for operating a hybrid vehicle having an internal combustion engine and a motor / generator that are connected to a common drive train is described. The method includes determining a requested powertrain torque output and operating the engine in a skip fire manner. A balancing torque is applied to the drive train with the motor / generator. The compensation torque counteracts torque variations that are introduced by igniting working chambers of the internal combustion engine and is determined by an adaptive filter feedforward control. In various embodiments, the adaptive filter feedforward control may use a noise correlated signal that may be based on the firing order. The interference-correlated signal can have an essentially vanishing mean value.

In other embodiments, a hybrid vehicle includes a motor / generator mechanically coupled to the drive train. An energy storage element is electrically coupled to the motor / generator. A skip fire controlled internal combustion engine is also mechanically coupled to the drive train and the electric motor / generator exerts a compensating torque on the drive train which counteracts a torque which is exerted by an igniting working chamber of the internal combustion engine. The balancing torque applied is determined using adaptive filter feed forward control. In some embodiments, the adaptive filter feedforward control uses a noise-correlated signal and the noise-correlated signal may be based on the firing order. The interference-correlated signal can have an essentially vanishing mean value. The energy storage element can be a capacitor, a battery or a parallel combination of a battery and a capacitor.

In still other embodiments, a method of controlling a hybrid vehicle having a powertrain with a powertrain controller that controls a skip fire controlled internal combustion engine and an auxiliary torque source / sink is described. The powertrain controller receives a torque request and determines an ignition sequence that provides the requested torque based on the torque request. A fault correlated signal can be generated based on the firing order. An adaptive filter can be used to filter the noise-correlated signal. The filtered noise correlated signal can be input as a control signal to the auxiliary torque source / sink. Balancing torque may be applied to the driveline from the auxiliary torque source / sink to thereby at least partially dampen torque vibrations resulting from the Skip Fire engine operation. An NVH metric can be determined based on the acquired signal. The NVH metric can be fed back to the adaptive filter in order to modify its filter characteristics. In some embodiments, the sensed signal may be crankshaft rotation or a time derivative thereof, or an accelerometer. The filter can be a digital FIR filter (Finite Impulse Response). The digital FIR filter can have weights that are modified based on the NVH metric. The NVH metric can be regulated so that it does not exceed a predefined target level, which would lead to an unacceptable NVH if exceeded.

While only a few embodiments of the invention have been described in detail, it should be understood that the invention can be implemented in many other forms without departing from the spirit or scope of the invention. For example, the drawings and the embodiments sometimes describe specific arrangements, operations, and control mechanisms. It is understood that these mechanisms and steps can be modified to fulfill the Requirements of different applications. For example, some or all of the steps and features of the NVH reducer module are not required, and instead, some or all of these steps may be applied to other modules, such as, as appropriate. B. the ignition component calculation device and / or the ignition timing determination unit. In some embodiments, one or more of the operations described may be rearranged, replaced, modified, or removed. Although the invention is applicable to all forms of hybrid vehicles, it is particularly applicable to micro hybrids that have a relatively low energy storage and motor / generator capacity that is insufficient to provide all of the driving force to drive the vehicle. The invention is also applicable to engines with any number of cylinders. Various embodiments of the invention are in compact vehicles with relatively small motors, such as. B. 2-, 3- or 4-cylinder engines, where the NVH associated with a Skip Fire engine with a small number of cylinders can be weakened by a compensating torque, particularly advantageous. Thus, the present embodiments are to be considered exemplary and not restrictive, and the invention is not to be limited to the details provided here.

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• US 62/470487 [0001]
• US 13654244 [0017, 0048, 0052, 0053, 0086, 0089]
• US 13963686 [0052, 0089]
• US 62053351 [0086]
• US 7954474 [0089]
• US 7886715 [0089]
• US 7849835 [0089]
• US 7577511 [0089]
• US 8099224 [0089]
• US 8131445 [0089]
• US 8131447 [0089]
• US 13774134 [0089]
• US 13953615 [0089]
• US 13886107 [0089]
• US 13963759 [0089]
• US 13963819 [0089]
• US 13961701 [0089]
• US 13963744 [0089]
• US 13843567 [0089]
• US 13794157 [0089]
• US 13842234 [0089]
• US 13004839 [0089]
• US 13004844 [0089]
• US 14207109 [0089]
• US 13681378 [0089]
• US 61952737 [0089]
• US 61879481 [0089]
• US 8015960 [0090]
• US 9399964 [0168]
• US 14705712 [0168]

Claims (56)

An electric motor / generator controller for operating an electric motor / generator used as an auxiliary torque source / sink in a hybrid powertrain, the hybrid powertrain including an internal combustion engine that operates in an ignition sequence, the motor / generator controller following includes: a matched basic function control module configured to generate a control signal for the electric motor / generator, the matched basis function control module being designed to receive a parameter indicating an ignition sequence length, a parameter indicating an ignition sequence phase, a crank angle signal and an NVH -Metric signal as inputs and for generating the control signal for the electric motor / generator based at least in part on the received parameter, which specifies the firing sequence length, the firing sequence phase, the crank angle signal and the NVH metric signal. Electric motor / generator after Claim 1 , wherein the matched basic function control module includes a sine and cosine generation module and the control signal output by the matched basis function control module is composed of at least one pair of cosine and sine functions, each pair of cosine and sine functions having an associated frequency , Electric motor / generator after Claim 2 , wherein the sine and cosine generation module includes an angle generation module that produces arguments for the sine and cosine generation module that have a periodicity that matches a periodicity of the firing order. Electric motor / generator after Claim 3 , wherein the periodicity of the firing order is a function of a denominator of an ignition component. Electric motor / generator after Claim 1 , wherein the NVH metric signal is based on a crankshaft rotation signal or a time derivative thereof. Electric motor / generator after Claim 5 wherein the crankshaft rotation signal or a derivative thereof is filtered by an NVH filter module and a vehicle response filter module to obtain the NVH metric. Electric motor / generator after Claim 5 , with the NVH metric signal depending on the gear ratio. Electric motor / generator after Claim 1 , wherein the adjusted basis function control module includes an adjusted basis function coefficient module. Electric motor / generator after Claim 8 , wherein the fitted basis function coefficient module uses a recursive least squares algorithm or a root mean square algorithm to determine the coefficients. Electric motor / generator after Claim 1 , wherein the firing order has multiple firing opportunities and some of the firing opportunities produce a low output torque and some of the firing opportunities produce a high output torque. A method of operating an electric motor / generator used as an auxiliary torque source / sink in a hybrid powertrain, the hybrid powertrain including an internal combustion engine operating in an ignition sequence, the method comprising: Generating a control signal for the electric motor / generator using a customized basis function control algorithm to provide an acceptable NVH metric. Procedure according to Claim 11 , wherein the control signal output by the adapted basis function control algorithm is composed of at least one pair of cosine and sine functions, each pair of cosine and sine functions having an associated frequency. Procedure according to Claim 12 , the sine and cosine signals having a periodicity that matches the firing order. Procedure according to Claim 13 , wherein the ignition sequence is based on an ignition component and the periodicity of the ignition sequence is based in part on a denominator of the ignition component. Procedure according to Claim 11 , wherein the NVH metric signal is based on a crankshaft rotation signal or a time derivative thereof. Procedure according to Claim 15 wherein the crankshaft rotation signal or a derivative thereof is filtered by an NVH filter module and a vehicle response filter module to obtain the NVH metric. Procedure according to Claim 15 , with the NVH metric signal depending on the gear ratio. Procedure according to Claim 11 , wherein the matched basis function control algorithm uses matched basis function coefficients to generate the control signal for an electric motor / generator. Procedure according to Claim 18 , wherein the fitted basis function coefficients are determined using a recursive least squares algorithm or a quadratic mean algorithm. Powertrain control for operating an internal combustion engine in a dynamic ignition level modulation manner using an operational effective ignition component that is fuel efficient and has acceptable noise, vibration and roughness (NVH) characteristics, the powertrain control comprising: an ignition fraction calculation device configured to generate a usable effective ignition fraction to provide a requested engine torque; an ignition timing determination module configured to generate an ignition sequence used to operate the engine in a dynamic ignition level modulation manner, the ignition sequence based on the operational effective ignition fraction; and an adapted basic function control module that is configured to determine a compensation torque that is applied to a drive train by an energy storage / collection / release device, the compensation torque being designed to at least partially cancel out a torque variation generated by the ignition sequence, thereby reducing the NVH that would otherwise be generated by the firing order. Powertrain control for operating an internal combustion engine in a dynamic ignition level modulation manner using an operational effective ignition component that is fuel efficient and has acceptable noise, vibration and roughness (NVH) characteristics, the powertrain control comprising: an ignition fraction calculation device configured to generate a usable effective ignition fraction to provide a requested engine torque; and an ignition timing determination module configured to generate an ignition sequence used to operate the engine in a dynamic ignition level modulation manner, the ignition sequence based on the operational effective ignition fraction; wherein the powertrain controller is an adaptive controller configured to determine a compensation torque applied to a drive train by an energy storage / catch / release device, the compensation torque being configured to at least partially cancel a torque variation generated by the firing order , reducing the NVH that would otherwise be generated by the firing order. Powertrain control after Claim 21 , wherein the control characteristics of the adaptive controller are modified based on a detected parameter indicating NVH. Powertrain control after Claim 21 , wherein the dynamic firing level modulation mode has two torque levels associated with each firing opportunity. Powertrain control after Claim 21 wherein the dynamic firing level modulation manner has three torque levels associated with each firing opportunity, the three torque levels being a high torque output, a low torque output level, and an omitted output level. Powertrain control after Claim 21 wherein the dynamic firing level modulation mode is a skip fire operation with all firing work chambers fired at substantially the same delivery level and some work chambers omitted. Powertrain control after Claim 21 , wherein the powertrain control uses an adaptive filter feedforward controller. Powertrain control after Claim 26 wherein the powertrain controller uses a fault-correlated signal as an input to determine the compensation torque and the fault-correlated signal is generated based in part on the firing order. Powertrain control after Claim 27 , wherein the interference correlated signal is generated based on a filtered version of the firing order. Powertrain control after Claim 27 , wherein the disturbance-correlated signal is generated based on an engine torque model that estimates engine torque as a function of the firing order. Powertrain control after Claim 27 , wherein the disturbance-correlated signal is generated by scaling a signal that is correlated with an overlay of a predicted torque profile from all working chambers of the engine. Powertrain control after Claim 27 wherein the adaptive filter feedforward controller includes an adaptive filter with a finite impulse response or FIR filter that has weights that are updated at least in part based on a feedback signal. Powertrain control after Claim 31 wherein the filter weights are updated recursively in a weight update module. Powertrain control after Claim 32 , wherein the FIR filter weights are updated using a Least Squares Average Algorithm or a recursive Least Squares Average Algorithm. Powertrain control after Claim 31 , wherein the adaptive forward filter uses the filter weights determined from a look-up table in certain operating modes. Powertrain control after Claim 31 , wherein the filter weights are based at least in part on a penalty term that includes energy costs associated with providing the balancing torque. Powertrain control after Claim 32 , wherein a bias is estimated and removed from the noise correlated signal so that it is essentially devoid of bias. Powertrain control after Claim 36 , wherein the adaptive filter feedforward controller comprises an additional loop shape filter for damping frequency components of the compensation torque. Powertrain control after Claim 37 wherein the adaptive filter feedforward controller includes a virtual feedback loop through which the estimated effects of the removed bias term and the frequency components attenuated by the loop shape filter are subtracted from a feedback signal input to the weighting update module. Powertrain control after Claim 21 where the powertrain control minimizes fuel consumption while delivering the requested torque at an acceptable NVH level. Powertrain control after Claim 21 wherein the energy storage / catch / release device comprises an electric motor / generator and a capacitor. Powertrain control after Claim 40 , wherein the capacitor energy storage capacity is only sufficient to dampen the torque variation generated by the firing order. A method of operating an internal combustion engine in a dynamic ignition level manner using an operational effective ignition component that is fuel efficient and has acceptable Noise, Vibration and Roughness (NVH) characteristics, the method comprising: generating an operational ignition component to provide one requested engine torque; Generating a firing order used to operate the engine in a dynamic firing level manner, the firing order based on the operational effective ignition fraction; Determine, by means of an adaptive control, a compensation torque that is applied to a drive train by an energy storage / collection / release device, the compensation torque being designed to at least partially cancel out torque variations generated by the firing order, thereby reducing the NVH, the otherwise would be generated by the firing order. Procedure according to Claim 42 , wherein the control characteristic of the adaptive control is modified on the basis of a detected parameter that indicates NVH. Procedure according to Claim 43 , wherein the adaptive control uses an adaptive filter feedforward control. Procedure according to Claim 44 wherein determining the compensation torque includes generating a disturbance-correlated signal based at least in part on the firing order and using the disturbance-correlated signal as an input to determine the compensating torque. Procedure according to Claim 45 , wherein generating the interference-correlated signal comprises generating a filtered version of the firing order. Procedure according to Claim 45 wherein generating the disturbance-correlated signal includes using an engine torque model to estimate engine torque as a function of the firing order. Procedure according to Claim 45 wherein generating the disturbance-correlated signal includes scaling a signal correlated with an overlay of a predicted torque profile from all of the working chambers of the engine. Procedure according to Claim 43 wherein determining using adaptive filter feedforward includes performing adaptive filtering with finite impulse response or FIR filtering that has FIR weights that can be updated at least in part based on a feedback signal. Procedure according to Claim 43 wherein generating the firing order includes performing sigma-delta filtering to determine the firing order; and generating the disturbance-correlated signal includes removing a bias by subtracting an input of the sigma-delta filtering. Procedure according to Claim 43 , wherein the adaptive filter feedforward control includes a loop shape filter to control frequency components of the compensation torque. Procedure according to Claim 43 , with the adaptive filter feed forward controller minimizing fuel consumption while delivering the requested torque and an acceptable NVH level. Powertrain control after Claim 43 , wherein the adaptive control uses an adapted basic function control. Computer readable storage medium comprising executable computer code that is executed in tangible form and for operating an internal combustion engine in a manner of dynamic ignition level modulation using an operational effective ignition component that is fuel efficient and has acceptable noise, vibration and roughness (NVH) - Features, is suitable, wherein the computer readable storage medium includes: executable computer code for generating a usable effective ignition component for delivering a requested torque; executable computer code to generate a firing order used to operate the engine in a dynamic firing level modulation manner, the firing order based on the operational effective firing fraction; and executable computer code for determining, by means of an adaptive control, a compensation torque, which is applied to a drive train by an energy storage / collection / release device The compensating torque is designed to at least partially cancel out torque variations generated by the firing order, thereby reducing the NVH that would otherwise be generated by the firing order. Computer-readable storage medium Claim 54 , wherein the adaptive control uses an adaptive filter feedforward control. Computer-readable storage medium Claim 54 , wherein the adaptive control uses an adapted basic function control.

Images