CN111434025B

CN111434025B - Method for determining the rotor angle of an electric machine in a motor vehicle

Info

Publication number: CN111434025B
Application number: CN201880080283.1A
Authority: CN
Inventors: B.雷内克; J.穆勒; W.费舍尔; S.葛罗德
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2017-12-15
Filing date: 2018-11-29
Publication date: 2024-03-26
Anticipated expiration: 2038-11-29
Also published as: WO2019115238A1; DE102017222842A1; EP3724987A1; CN111434025A

Abstract

The invention relates to a method for determining the rotor angle of an electric machine (30)) The electric machine has a rotor (32) and a stator (33) with at least one phase winding (U, V, W), wherein the electric machine (30) is associated with a switched-mode charging regulator (LR) which is designed to regulate the electric machine (30) and to apply electric energy to an electrical store (S), wherein the charging regulator (LR) has a first switching state (S1) in which the application of electric energy to the electrical store (S) is performed, and a further switching state (S2) in which the application of electric energy to the electrical store (S) is at least partially inhibited, wherein in the first switching state (S1) the rotor angle is determined by means of a first determination rule (K1)) And determining the rotor angle in the other switching state (S2) by means of other determination rules (K2)). The invention also relates to a corresponding computing unit designed to perform the method, and to a computer program for performing the method.

Description

Method for determining the rotor angle of an electric machine in a motor vehicle

Technical Field

The invention relates to a method for determining the rotor angle of an electric machine, comprising a rotor and a stator having at least one phase winding, wherein the electric machine is associated with a switching charge regulator, which is designed to regulate the electric machine and to apply electrical energy to an electrical storage.

Background

The rotational angular position and rotational speed of the crankshaft of an internal combustion engine are important input variables for many functions of an electronic engine control device. In order to determine the rotational angle position and the rotational speed, marks may be provided at the same angular intervals on a main body that rotates together with a crankshaft of the internal combustion engine. The markings due to the rotation of the crankshaft are detected by a sensor and forwarded as an electrical signal to the evaluation electronics.

The electronic device determines the signal for the marker stored for the rotational angle position or measures the time difference between the two markers for the respective rotational angle position of the crankshaft, and can determine the angular velocity on the basis of the known angular separation of the two markers from one another and determine the rotational speed therefrom. In the case of motor vehicles, in particular in the case of motorcycles, scooters or motor-scooters, the marking can be provided, for example, by means of teeth of a metal gear, a so-called sensor wheel, which by their movement cause a change in the magnetic field in the sensor. The absence of some teeth may be used as a reference mark to identify absolute position.

Although 60-2 teeth are mostly used in the case of passenger cars (60 teeth are evenly distributed with two of them left empty), 36-2, 24-2 or 12-3 teeth are also used in the case of motorcycles or constraint motorcycles, for example. In this indirect principle of rotational speed determination or rotational angle position determination of the crankshaft, the resolution of the rotational speed signal or the absolute detection of the rotational angle position is determined by the number of teeth and by the reliable detection of the reference mark.

In modern vehicles each having an internal combustion engine, an electrical generator is installed, which is driven by the rotation of the crankshaft and provides electrical signals for supplying the vehicle with electrical energy and charging the vehicle battery. Without the generator, the vehicle may not operate as prescribed, or may only operate as prescribed for a short period of time. A regulator is used for regulating the battery voltage. Since the generator is implemented as a permanent excitation for many motorcycles or for limiting motorcycles, its excitation for regulating the battery voltage cannot be changed as is usual in passenger vehicles. Instead, the regulator regulates the battery voltage to a nominal value, for example by shorting the phases of the motor. The generator is typically used in addition to the sensor for detecting the rotational speed or for detecting the rotational angular position of the crankshaft.

The exact rotational angular orientation of the rotor of the motor can be read directly from the idling voltage of the unloaded motor, since the relative phase of the idling voltage coincides with the rotational angular orientation of the rotor. In the case of a loaded machine, the exact rotational angular position of the rotor can only be determined by additionally taking into account the rotor angle. Thus, it is possible to precisely determine the rotational angle position of the crankshaft from the signal of the motor only if the rotor angle can likewise be determined with sufficient accuracy. For a loaded motor this is not easily achieved. Furthermore, a corresponding voltage regulation, in particular a switched-mode voltage regulation in which at least one phase is short-circuited, may additionally make the determination of the rotor angle more difficult.

Furthermore, it is also known from EP 0664887 B1 to use an electrical output variable of an electric motor driven via a crankshaft for determining the rotational speed. For this purpose, a phase of the generator is provided as a reference, to which a pulsating direct voltage is applied. This arrangement can also be considered for determining an estimate of the rotational angle position of the rotor of the electric machine and thus also of the crankshaft of the internal combustion engine, which are coupled to one another directly or via a transmission, respectively, also on the basis of the respective phase signals.

A corresponding voltage regulation, which influences the electrical output variable of the electric machine at least during the switching phase of the voltage regulator, which is common in the case of motor cycles, for example, in the region of short-circuit regulation, is unsuitable here, since the characteristic signal for determining the rotational speed or rotational angle position of the shaft cannot be used reliably to determine the rotational speed or rotational angle position of the rotor. In addition, a high-resolution determination of the rotational speed of the crankshaft or of the rotor of the electric machine or of the rotational angle position is not achieved here.

It is therefore desirable to specify a possibility for determining the rotor angle of an electrical machine with a switching voltage regulator over the switching state.

Disclosure of Invention

According to the invention, a method is proposed having the features of claim 1. Advantageous configurations are the subject matter of the dependent claims and described below.

THE ADVANTAGES OF THE PRESENT INVENTION

The invention relates to a method for determining the rotor angle of an electric machine having a rotor and a stator with at least one phase winding, wherein the electric machine is associated with a switched-mode charge regulator which is designed to regulate the electric machine and to apply electrical energy to an electrical storage. The charging regulator has a first switching state in which electrical energy is applied to an electrical storage, and further switching states in which the application of electrical energy to the electrical storage is at least partially, preferably completely, inhibited, wherein the rotor angle is determined in the first switching state by means of a first determination rule and the rotor angle is determined in the further switching states by means of a further determination rule. In principle, it is possible within the scope of the invention to design the charging regulator in such a way that it is directly associated with the electric machine, but it is also possible to externally associate the charging regulator with a separate unit, in particular an engine control device, or to integrate it in the separate unit.

The method according to the invention has the advantage that by using different determination rules for the respective switching states of the charging states (charged or uncharged), the rotor angle can be determined at least stepwise in time, since the determination rules can be adapted to the current system parameters for determining the rotor angle, respectively. The determination rules comprise model-based determination rules for determining the rotor angle, wherein system parameters of the electric machine for different operating conditions can be used accordingly. For example, the internal resistance and coil inductance of the motors described herein, as well as their behavior at the desired idle or output voltage, may be used as system parameters. They may preferably be stored in the charging regulator or in a superordinate control device within the range of the characteristic map for determining the rotor angle. The above-described method therefore has the advantage that the rotor angle can be determined substantially continuously while the electric machine is running, despite a switching intervention by a voltage regulator or a charging regulator, which regulates the application of electric energy to a memory by a corresponding switching or regulating intervention on the electric machine.

In a preferred configuration of the method, the motor is actuated in a first switching state such that electrical energy is applied to the electrical storage, wherein the motor is actuated in the further switching state such that the current flow from the motor to the electrical storage is regulated, preferably inhibited, without load by shorting at least one phase winding or by leaving at least one phase winding free of current. In particular, it is particularly advantageous to regulate the electrical storage device by means of short-circuiting at least one phase or by leaving the respective phase free of current, since this can be achieved in a particularly simple and cost-effective manner. Such an adjustment is used in particular in motor vehicles, in particular in motor vehicles which are cost-effective, since the above-mentioned advantages play a particularly great role here. In particular, short-circuit regulation is particularly common, which has the disadvantage that the short-circuit regulation itself particularly strongly influences the phase signals of the motor, which makes it particularly difficult to determine the rotor angle from the phase signals.

In a further preferred embodiment of the method, the rotor angle is determined on the basis of a numerical model and/or on the basis of a characteristic curve family within the scope of the first determination rule and/or on the basis of other numerical models and/or on the basis of other characteristic curve families within the scope of the other determination rule. In principle, the determination rule may be implemented within the scope of a numerical model using global machine variables or based on a combination of numerical models using a family of characteristic curves in which a plurality of machine variables are stored. The characteristic map here comprises a rotor angle that depends on a respective parameter (for example the rotational speed or the output voltage of the generator). However, it is also preferred that a respective machine variable (e.g. the output voltage of the generator) is detected from the rotational speed or the edge time between the edges of at least one phase signal and stored in the characteristic map for further use in the numerical model.

In a further preferred embodiment of the method, in the case of the first determination rule, the voltage of the electrical storage is taken into account when determining the rotor angle of the electrical machine. In particular, when using a family of characteristic curves in which the respective machine parameters of the motor for determining the rotor angle are stored, the battery voltage can be used in determining the rotor angle. In principle, the battery voltage can only be regarded approximately as constant, so that the battery voltage can actually be important in determining the rotor angle. Depending on the operating point, fluctuations or level drops may occur. Since the operating voltage is usually measured continuously in the superordinate control device, these changes can be detected and correspondingly taken into account in the determination of the rotor angle. In a further embodiment, the rotor angle characteristic curve or the corresponding characteristic curve family for the battery charging operation in the first switching state can be corrected for a plurality of parameters, in particular by compensation parameters that depend on the battery voltage. In principle, further corrections can be made to the characteristic curve or to the corresponding characteristic curve family, for example tilting, stretching or compressing or other deformations of the characteristic curve or of the characteristic curve family.

In a further preferred embodiment of the method, the dynamic oscillation process caused by the switching on during a time change of at least one of the machine variables of the electric machine on which the rotor angle is determined is taken into account in determining the rotor angle of the electric machine during a time change of the at least one of the machine variables of the electric machine after the electric machine has been switched on in the range of the first switching state and/or after the electric machine has been switched on in the range of the further switching states. By switching on the motor for regulating the voltage, in particular after a corresponding switching process, a corresponding transient state with a large temporal dynamic change is produced, which returns to a substantially stationary state for the duration of the characterization of the machine by means of a starting process. If the respective machine parameters of the electric machine (e.g. output voltage, rotational speed, etc. of the generator) are now affected by such dynamics, such dynamics propagate into the determination of the rotor angle, which may lead to a corresponding error source. These error sources can be taken into account accordingly in order to determine the rotor angle, in particular also the rotor angle during a characteristic duration after switching from the first to the second switching state or from the second to the first switching state. This is particularly advantageous because now also unbalanced conditions can be taken into account when determining the rotor angle, and thus the rotor angle can be determined more precisely and in each operating point of the electric machine.

In a further preferred embodiment of the method, the dynamic oscillation starting process during the time-dependent change in the rotor angle is characterized in such a way that the amplitude of the rotor angle and/or the operating parameters of the electric machine are determined for the duration of the dynamic oscillation starting process and are used as a measure for taking into account the dynamic oscillation starting process when determining the rotor angle. In particular, the amplitude of the temporal dynamics of the rotor angle during the dynamic oscillation starting process can be used as a measure for whether a corresponding correction is made during the duration of the dynamic oscillation starting process. This can be based in particular on a threshold adjustment, in which a lower threshold value for the amplitude of the rotor angle oscillations in the dynamic time range is used as a basis, wherein a lower threshold value is not adapted or a higher threshold value is used to correct the rotor angle in the time range of the dynamic oscillation starting process. If the influence of the dynamic start-up process is too great, i.e. in particular the amplitude is greater than a certain threshold value, these amplitudes can be stored as application variables and used accordingly for calculating the rotor angle after the switching process. In particular, the application variables can be determined by means of reference measurements or suitable simulation models. Furthermore, depending on the influence of the rotational speed gradient on the suitability of the respective correction, the switching process can be placed at a suitable position in the operating cycle of the internal combustion engine driving the electric machine, for example when the rotational speed change process is as flat as possible. Accordingly, a flat rotational speed course is a rotational speed course with the smallest possible gradient.

In a further preferred embodiment, the switching process can be performed as a function of at least one rotational angular position of the rotor, in particular a switching process which results in a flow of current from the motor to the electrical storage device being inhibited. The advantage of this measure is that switching on the motor to regulate the voltage of the electrical storage and the corresponding use of motor parameters for determining the rotor angle are always performed with a time offset, so that the rotor angle can be determined without interference from the machine parameters, respectively, taking into account the minimum time interval from the switching process to ensure a possible starting process. Furthermore, it may be preferable to place the switching state in an area with low rotational speed dynamics. Alternatively, it may also be provided that the switching process is placed directly after the occurrence of a signal edge and/or zero crossing of at least one phase signal, in order to attenuate as much as possible the dynamics during the change in the measured variable for determining the rotor angle until the next edge in the corresponding phase signal comes during the change.

In addition, depending on the type and extent of the amplitude or the amplitude thereof, the type and extent of possible corrections can also be made in the time frame of the dynamic starting process when correcting the rotor angle to be determined.

In a further preferred embodiment of the method, the characteristic curve family or the further characteristic curve family has at least the rotor angle, the output voltage of the electric machine and the rotational speed or the time between two edges of at least one phase signal as reference variables. In principle, all motor parameters which are relevant for determining the rotor angle and parameters which depend on these motor parameters can be stored in the respective characteristic curve family, in order to be able to determine the rotor angle as precisely as possible from the respective operating parameters of the motor.

In a further preferred embodiment of the method, at least one rotor angle value stored in at least one characteristic curve family is used in a first revolution of the rotor, and the at least one rotor angle value is corrected by a rotor angle value determined based on measurements over the duration of the dynamic starting process, wherein the corrected rotor angle is used for further revolutions of the rotor, in particular for a time range of further dynamic starting processes. Instead of correcting or accumulating the rotor angle, in particular in the time frame of the dynamic starting process, using a corresponding correction term, it is also possible to learn such a correction term during operation of the electric machine. For this purpose, values outside these switching states are extrapolated from the motor measured variables and the time-dependent course of the measured variables between the first switching state and the further switching state or before the switching process between the further switching state and the first switching state, and the corresponding characteristic curve or characteristic curve family of the rotor angle is used here. Deviations of the extrapolated values from the measured values after the switching process can be taken into account in possible correction terms and can be stored for further use in determining the rotor angle using other revolutions of the rotor. This configuration is advantageous because the respective machine parameters, which are sometimes also affected by the respective fluctuations and degradation effects over time, can be used as a learning function in the case of determining the rotor angle while running.

In a further preferred embodiment of the invention, as an addition or replacement of the correction factors extrapolated using the time course of the motor measured variable preceding the switching process and the determination of the correction factors based on the extrapolation, it is also possible to learn the undisturbed course of the electrical variable at the same or similar operating points in terms of the course of the rotor angle change, at which no switching process occurs. Likewise, the correction factor can be determined by comparing an undisturbed (at least undisturbed by one of the switching processes) change with an electrical variable change following the switching process and using a corresponding characteristic curve or characteristic curve family of the rotor angle. This configuration is also advantageous, since corresponding operating or machine parameters, which are also sometimes influenced over time by corresponding fluctuations and degradation effects, can be used as a learning function in the case of determining the rotor angle while running.

In a further preferred embodiment of the invention, the rotor angle is used to determine the rotational angular position of the rotor. In order to reliably derive the respective rotational angular position of the rotor from the signals of the electric machine, it is necessary to ensure that the rotor angle is determined correspondingly precisely for the respective operating conditions of the electric machine. Within the scope of the above-described method, this is possible in particular, whereby the rotor angle can be determined with a high degree of accuracy independently of the respective operating state of the electric machine.

In a further preferred embodiment of the invention, at least one phase signal of the electric machine is processed by means of an electronic circuit, in particular an engine control device. By appropriate external processing of the phase signal or of the values associated with the phase signal and of the associated rising and falling edges and appropriate external processing of the regulation, in particular of the charge regulation of the electrical storage in the engine control device, additional control components can be dispensed with, since the engine control device is already present and can in principle also be used for this purpose. This is advantageous, since the corresponding adjustment architecture can thereby be simplified, whereby additional costs can be saved.

In principle, it is easily understood that by means of the method described above, the high-resolution rotational angle position or rotational speed of the rotor of the electric machine and thus also of the crankshaft of the internal combustion engine can be determined directly from the internal signal of the electric machine, whereby a corresponding sensor wheel for determining the rotational angle position or rotational speed and a sensor device connected to the sensor wheel can also be omitted. The rotational angle position or rotational speed of the rotor can always be determined during operation, since a corresponding charge regulation of the electrical storage device is correspondingly taken into account. However, it is also possible to decouple the switching intervention of the charge regulation in time from the determination of the edges of the phase signals, which are necessary for determining the rotational angle position or the rotational speed.

Thus, the determination of the rotor and thus the high-precision rotational speed and rotational angle position of the crankshaft and the corresponding voltage regulation of the electrical storage device can be ensured while running using only the phase phases of the electrical machine. Thereby, costs can be saved, which is particularly advantageous for scooters or scooters, which are more cost-effective. Further, control functions such as position calculation of injection, torque calculation, or learning function for accurately determining the OT position, etc. can be significantly improved.

It is furthermore easy to understand that the phase signals can in principle be obtained in different ways. For example, the phase voltages can be observed with respect to each other, the phase voltages at the diodes of the connected rectifiers with respect to the output terminal potential of the rectifiers, provided that the stator of the motor forms a star circuit with extractable star points, the output voltages of the observation branches with respect to the star points or similar analyses of the phase currents are performed.

In a further preferred embodiment of the method, the rotational angular position of the crankshaft is used for controlling the internal combustion engine. The detection and processing of the phase signals of the electric machine by the engine control device and the corresponding determination of the rotational angular position of the crankshaft as a function of the rotational angular position of the rotor and the possible angular offset derived from the rotor angle can be used in a control device of the internal combustion engine to control the ignition point or the torque of the internal combustion engine. Thus, the regulation of the charge of the battery, the control of the internal combustion engine and the improved determination of the rotational angle position or rotational speed of the crankshaft can be integrated in the engine control device, whereby a further synergistic effect is obtained. For this purpose, the used computing unit, which is preferably configured as an engine control device for the internal combustion engine, has a corresponding integrated circuit and/or a computer program stored on a memory, which is designed to execute the above-described method steps.

It is advantageous if the method is implemented or provided in the form of a computer program, in particular an ASIC (application specific integrated circuit), which is preferably stored in software on a data carrier, in particular on a memory, and which can be used in the computing unit to execute the method, since this results in particularly low costs, in particular if the execution control device is also used for other tasks and therefore is already present. As is generally known from the prior art, data carriers, in particular magnetic, optical and electrical memories, suitable for providing the computer program.

Drawings

Other advantages and configurations of the invention will be apparent from the description and drawings.

Fig. 1 schematically shows a sensor wheel with a sensor according to the prior art, in particular for determining the rotational speed;

fig. 2a to 2c show schematic diagrams (a, b) of an electric machine coupled to an internal combustion engine and the associated signal change process (c);

fig. 3 schematically shows an electric machine with corresponding belonging phase signals;

figures 4a and 4b show possible voltage variations of the phases of a three-phase motor;

fig. 5a and 5b show a single-phase simplified equivalent circuit diagram (a) of the motor, and an associated vector diagram (b) of the phase voltage vector;

6a-6f illustrate six different embodiments of regulator circuits connected downstream of the rectifier of the motor and designed to regulate the battery voltage;

figures 7a and 7b show the course of a phase signal with regulatory intervention according to a first embodiment and an alternative second embodiment of the method;

figures 8a and 8b show a variation of the phase signal with regulatory intervention according to another embodiment of the method and alternative another embodiment;

FIG. 9 shows a phase change process with clocked conditioning intervention according to yet another embodiment of the method;

fig. 10a to 10d show a time-varying course (a) of the rotor angle with respect to the rotational speed at a first output voltage of the motor, a time-varying course (b) of the rotor angle with respect to the rotational speed at a further output voltage of the motor, a time-varying course (c) of the rotor angle with respect to an edge time between two edges of a phase signal at the first output voltage of the motor, and a time-varying course (d) of the rotor angle with respect to an edge time between two edges of a phase signal at the further output voltage of the motor;

fig. 11 shows the time course of the rotor angle in two switching states of the regulator and the dynamic oscillation initiation process field arranged between the two switching states;

Fig. 12 shows a time-varying process of the voltage edges of the phase signals, based on which the learning method is explained; and

FIG. 13 schematically illustrates a method based on determining rotor angleA flow chart of the foregoing method.

Detailed Description

Fig. 1 schematically shows a sensor wheel 20 and an associated inductive sensor 10, which are used in the prior art for determining the rotational speed of a crankshaft or for approximately determining the rotational angular position of a crankshaft. The sensor wheel 20 is firmly connected to the crankshaft of the internal combustion engine, and the sensor 10 is mounted in a stationary manner in place.

The sensor wheel 20, which is usually made of ferromagnetic material, has teeth 22, which teeth 22 are arranged on the outside with a space 21 between two teeth 22. At one of said outer positions the sensor wheel 20 has a recess 23 over the length of a predetermined number of teeth. The recess 23 serves as a reference mark for identifying the absolute position of the sensor wheel 20.

The sensor 10 has a bar magnet 11 on which a soft magnetic pole contact pin 12 is mounted. The pole contact pin 12 is in turn surrounded by an induction coil 13. As the sensor wheel rotates, the teeth 22 and the space between each two teeth alternately pass the induction coil 13 of the sensor 10. Since the sensor wheel is made of ferromagnetic material and thus the teeth 22 are also made of ferromagnetic material, a signal is induced in the coil during the rotation, with which signal the teeth 22 and the air gap can be distinguished.

By correlating the time difference between two teeth with the angle enclosed by the two teeth, the angular or rotational speed, and in addition the corresponding angular position of the crankshaft, can be approximately calculated.

At the gap 23, the signal induced in the induction coil has a different course than at the teeth 22 that otherwise alternate with the gap. In this way, absolute position marking can be achieved, but only with reference to a complete revolution of the crankshaft.

In fig. 2a, an internal combustion engine 112 is depicted, with an electric machine 30 being connected to the internal combustion engine 112 directly or via a drive coupling, wherein the electric machine 30 is driven by a crankshaft 17' of the internal combustion engine 112. Thus, the rotational speed n of the motor 30 _Gen And the rotational speed n of the crankshaft 17 _BKM And the angular position alpha of the rotor of the electric machine 30 ₁ And the rotational angle position α of the crankshaft 17' have a fixed relationship with each other. Furthermore, the electric machine 30 is associated with a charging regulator LR, which supplies energy to an electrical storage S (in the present case, battery B) in the on-board electrical system 110 in accordance with the remaining capacity of the battery B. Depending on the state of charge of battery B or electric storage S, they have a varying voltage U _Bat . Furthermore, a computing unit, in particular an engine control device 122, is provided, which exchanges data with the electric machine 30 or with the internal combustion engine 112 via a communication connection 124 and is designed to control the internal combustion engine 112 and the electric machine 30 accordingly.

In fig. 2b, the motor 30 is again schematically shown in an enlarged form. The electric machine 30 has a rotor 32 with field windings and a stator 33 with stator windings, wherein the rotor 32 has a shaft 17. This is therefore an excited machine, which is common in particular in the case of motor vehicles. However, in particular for motor-limited vehicles, especially for small and light motor-limited vehicles, engines with permanent magnets, i.e. permanent magnet excited motors, are mostly used. In principle, both types of electric machines can be used within the scope of the invention, wherein in particular the method according to the invention does not depend on the use of electric machines of the respective type (permanent-magnet excited machines or separately excited machines).

The electric machine 30 is embodied as a three-phase generator, in which three phase voltage signals are induced, which are 120 ° phase-shifted to each other. Such a three-phase generator is generally used as a generator in modern motor vehicles and is suitable for performing the method according to the invention. Within the scope of the invention, it is in principle possible to use all motors independently of their phase numbers, wherein in particular the method according to the invention is independent of the use of the corresponding type of motor.

Three phases of the three-phase generator 30 are denoted by U, V, W. The voltage dropped across the phases is rectified via rectifying elements configured as a positive diode 34 and a negative diode 35. Thus, there is a generator voltage U between poles B+ and B _G The negative electrode is grounded in the case of this generator voltage. Such a three-phase generator 30 supplies power to, for example, a battery B or other consumer in the vehicle electrical system 110.

Fig. 2c shows three diagrams, which show the associated voltage course with respect to the rotation angle of the rotor 32 of the electric machine 30. The voltage course of the phases U, V, W and the associated phase voltages U are plotted in the upper graph _P . It is generally understood that the numbers and value ranges illustrated in this and the following figures are merely exemplary and thus do not limit the invention substantially.

The generator voltage U is shown in the middle diagram _G Which is formed by the envelopes of the positive and negative half waves of the voltage course U, V, W.

Finally, the rectified generator voltage U is shown in the following chart _G- (see FIG. 2 a) and the generator voltage U _G- Effective value U of (2) _Geff The power generationA pole voltage is applied between b+ and B-.

In fig. 3 a stator 33 with phases U, V, W is schematically shown, together with a positive diode 34 and a negative diode 35 from fig. 2 b. In principle, it is easy to understand that the rectifier elements depicted here in the form of positive diode 34 and negative diode 35 can also be configured as transistors (not shown), in particular MOSFETs (metal oxide semiconductor field effect transistors), in the case of active rectifiers. The terms used below also show the voltages and currents that occur.

U _U ，U _V ，U _W Instead, the phase voltages of the associated phases U, V, W are represented, which fall between the outer conductor and the star point of the stator 33. U (U) _uv ，U _vw ，U _wu Representing the voltage between the two phases or the external conductors to which they belong.

I _U ，I _V ，I _W Representing the phase currents from the respective outer conductors of the phases U, V, W to said star point. I represents the total current of all phases after rectification.

In fig. 4a, three phase voltages U with potential references on B-are now shown in three diagrams versus time _U ，U _V ，U _W These phase voltages occur in generators having an outer pole rotor with six permanent magnets. The illustration of the electric machine 30 with the three-phase stator winding 33 is only exemplary to be seen, wherein in principle, without limiting the generality, the method according to the invention can also be implemented on a generator with a corresponding number of phases or permanent magnets or field coils as required. Instead of a star connection of the stator coils, a delta connection or other connection can likewise be selected.

In the case of a motor 30 having a current output, the phase voltage U _U ，U _V ，U _W Is rectangular in a first approximation. This is explained in particular by the fact that: the positive or negative diode is made conductive in the flow direction by the generator voltage and therefore measures approximately 15-16 volts (battery charge voltage in the case of a 12V lead-acid battery and voltage on the positive diode) or negative 0.7-1 volts (negative two) Voltage on the pole tube). The measured reference potentials are each ground. Other reference potentials, such as the star point of the stator, may also be selected. Although these reference potentials result in different signal profiles, the analyzable information, its acquisition and analysis are not altered.

In principle, the phase signal (U _U ，U _V ，U _W ，I _U ，I _V ，I _W ). For example, the phase voltages (U _UV ，U _UW ，U _WU ) Phase voltages are determined at the diodes of the connected rectifier with respect to the output terminals (b+, B-) of the rectifier, provided that the stator of the motor forms a star circuit with extractable star points, the observation branch being connected with respect to the star points (U _U ，U _V ，U _W ) Or a similar analysis of the phase currents.

In fig. 4b, the phase voltages U from fig. 4a are plotted together in a graph _U ，U _V ，U _W . A uniform phase shift can be clearly detected here.

During a complete revolution of the rotor 32 of the motor 30, the voltage signal is repeated six times through six magnets (in particular permanent magnets), so-called pole pairs. Thus, each phase of each revolution of the rotor 32, i.e., each phase voltage U _U ，U _V ，U _W Six falling edges FL occur _D And six rising edges FL _U (FL for the corresponding phase _UU ，FL _VU ，FL _WU And FL (field effect transistor) _UD ，FL _VD ，FL _WD )。

These edges set angular segments, that is to say angular segments which are exactly covered by the magnets along the radial periphery of the stator. Thus, the corresponding edge FL can be identified _U Or FL _D If the absolute reference point of each revolution is known, it is determined, for example, on the basis of the phase voltage U _U ，U _V ，U _W Is characterized by a reference magnet having a characteristic different from the other magnets.

Now utilizeThe falling edge FL can be identified by suitable means _D And rising edge FL _U Both of which are located in the same plane. For example, a TTL signal can be generated for each phase voltage by means of a so-called schmitt trigger and transmitted to the control device. The required schmitt triggers can be integrated in the control device or in the control electronics, for example in the control device, the battery voltage regulator and/or in the case of an active rectifier in the respective generator regulator, or can also be associated with them externally. The respective TTL signals can be transmitted via one line each, in particular using a control device, in particular an engine control device 122 (see fig. 2 a), or via only one data line 124 (see fig. 2 a) in a suitable combination by means of upstream combination electronics or other means.

In fig. 4b, the phase supply voltage U _U ，U _V ，U _W Respectively assigned the value W at the end of the corresponding falling edge of (a) _U ，W _V ，W _W These values are also referred to as W _Ud ，W _Vd ，W _Wd . Likewise, the rising edge FL can be given _U Assigning a corresponding value W _Uu ，W _Vu ，W _Wu . These values can be used to identify the rotational angular orientation α of the rotor 32 ₁ Or an angular increment set by the pole pairs of the stator 33. The rotational angular orientation alpha of the rotor 32 may also be identified based on plateau regions of the phase signals or other regions in between ₁ . Also, these values can be used for time-difference Δt-based ₁ ，Δt ₂ ，Δt ₃ A rotational speed of the generator is determined.

In this case, in the case of a uniform arrangement of six permanent magnets in the motor 30, a total of 18 falling edges FL occur _d And thus 18 associated values occur at equal intervals to each other, respectively, per revolution. Thus at a time difference Deltat ₁ ，Δt ₂ Or Deltat ₃ During this time, an angle of 360 °/18=20° is swept. As already mentioned at the beginning, this can also be used to identify the rotational angular position α of the rotor 32 ₁ Wherein an exemplary determined 20 deg. represents a detectable angular increment. In addition, the angular velocity ω can also be determined from this _i . The angular velocity isFrom omega _i ＝20°/Δt _i Obtained, relative rotation speed n _i Is made up of n _i ＝ω _i Obtained in units of revolutions per minute per 360 DEG.60 s/min.

In principle, it is easy to understand that the falling edge FL is _D Alternatively, the rising edge may also be used to determine the rotational angular orientation α of the rotor 32 ₁ And determining the instantaneous rotational speed n of the motor 30 _Gen . Thus, by doubling the number of values per revolution, the rotational angular orientation α of the rotor 32 is obtained ₁ And a rotational speed n _Gen Is a higher resolution of (a). In addition, the edges of the phases may be analyzed in a number of other ways, such as by the rising edges FL of the same or corresponding phases to each other _U And falling edge FL _D Time interval between, or by rising edges FL of the same phase or of all phases together _U Or falling edge FL _D Is a time interval of (a) for a time period of (b).

Except for rising edge FL _U And falling edge FL _D In addition, phase signal U _U ，U _V ，U _W May also be used to improve the determination of the rotational angular orientation alpha of the rotor 32 ₁ Or to identify the rotational speed n _Gen Is a single-layer structure.

Based on electric signals, in particular phase signals U, of the motor 30 _U ，U _V ，U _W Or the associated phase current I _U ，I _V ，I _W The actual rotational angular orientation alpha of the rotor 32 and its shaft 17 can only be determined with insufficient accuracy ₁ And thus the rotational angle position α of the crankshaft 17', since the phase signal U is generated in the event of a current flow causing a load on the motor 30 _U ，U _V ，U _W Or I _U ，I _V ，I _W Is relative to the actual rotational angular orientation alpha of the rotor 32 ₁ A systematic error in the form of an angular offset between them. This is explained in more detail in the following figures.

A schematic of a single-phase simplified equivalent circuit diagram of the motor is shown in fig. 5a, and the respective voltages or currents and their relative phase offsets are shown in relation to each other in a vector diagram in fig. 5 b. The knowledge determined from this single-phase equivalent circuit diagram can in principle also be transferred to a polyphase electric machine, which is shown, for example, in the preceding description. The voltage equation for a loaded motor can be derived from the motor single-phase equivalent circuit diagram in fig. 5a and the associated vector diagram shown in fig. 5b, as follows:

U _P =jx+u, where U corresponds to the output voltage of the motor 30, U _P Corresponds to the idle voltage of an unloaded electric machine, and ijx corresponds to the voltage drop U in the generator due to the current flowing through the electric machine and due to the reactance X of the electric machine _X 。

Here, the idle voltage U of the motor 30 _P Corresponding to the ideal induced voltage, which corresponds to the angular orientation alpha of the rotor 32 with respect to phase ₁ And consistent. In this case, an angular offset corresponding to the rotor angleAnd accordingly equal to zero. Thus, the idling voltage U _P Precisely reflecting the geometric movement of the rotor 32 and thus illustrating the precise angular orientation of said rotor in the unloaded state of the motor 30.

Due to the load of the electric machine 30 and the resulting current flow I, the output voltage U of the generator 30 of the load is relative to its induced idle voltage U _P Is then changed rapidly by an angular offsetSo-called rotor angle to U and U _P An angular offset therebetween. The rotor angle is in principle dependent on the coil current I and cannot be easily calculated without knowing the coil current I.

Furthermore, the angle between the output voltage U and the current I is obtained by the connected load, and is used for a pure-ohm consumerIdeal induction voltage (idle voltage) U of the motor _P Obtained as the product of motor constant, excitation and angular velocity. In permanently excited machinesA constant excitation and thus an ideal induced voltage proportional to the angular velocity is obtained by the permanent magnets used. Therefore, the vector diagram in fig. 5b is for the angular offset +.>The method comprises the following steps:

when using a linearly operating voltage regulator 40a, for example as shown in fig. 6a, and actuating a regulating element 42a for the voltage regulator 40a, which is embodied, for example, in the form of a power transistor and operates in a linear region (triode region), the output voltage U of the electric motor 30 can be regulated to be almost constant (relative to the battery voltage). Furthermore, the use of rectifiers 34a,35a at the output of generator 30 and an electrical storage S in the form of a battery B connected downstream approximately leads to a pure ohmic load, even if small capacitances may occur in the on-board electrical system. Whereby the angular offset between the output voltage U and the current I Correspondingly near 0, wherein the addend from the above formula +.>Also tends to 0 and thus vanishes.

Idling voltage U _P In principle with the rotational speed n of the electric motor 30 _Gen Proportional to the ratio. Thus, if it is assumed that the amplitude of the output voltage U is substantially constant and thatNear zero and therefore the second addend disappears, the above formula reduces to the following relationship:

wherein the constant const is substantially composed of a constant output voltage U and constant and is therefore independent of the rotational speed n _Gen Idle voltage U of (2) _P The components are obtained.

If it is selectedThe formulation of (a) depends on the edge time t _Gen Independent of the speed n _Gen Then get +.>And t _Gen The following relationship:

wherein const' comprises, in addition to the above constant factors, a control element for the rotational speed n _Gen Calculating the edge time t (in revolutions per minute (rpm)) _Gen Constant factor (in seconds).

In the relevant time range of a typical internal combustion engine from idling to about 15000rpm, this relationship can be approximately described by a straight line equation with a negative slope, and thus a higher computational efficiency can be achieved in the application. As already indicated at the outset, the illustrated value ranges are merely of an explanatory nature and should not limit the invention.

In this configuration of the battery regulation or in a corresponding manner of regulating the battery voltage, so that the corresponding regulating element 42 is operated in the linear region, the angular offset can be estimated with sufficient accuracy with a first approximation even without knowledge of the current flow I This allows a very reliable determination of the phase voltage U _U ，U _V ，U _W Is relative to the actual rotational angular orientation alpha of the rotor 32 ₁ Angular offset between +.>

Thus, the phase-dependent voltage U of the rotor 32 _U ，U _V ，U _W Determined rotational angular orientation alpha _phase Can correspondingly be determined by a corresponding rotational speed n _Gen Angular offset of (2)To be corrected. The actual rotational angle position α of the crankshaft 17 of the internal combustion engine or the rotational angle position α of the rotor 32 can thus be determined accordingly ₁ . In the case where there is a fixed coupling between the shaft of the rotor 32 and the crankshaft 17, the rotational angle position and the rotational angle orientation have a fixed relationship with each other. Therefore, the general applicability of α=α is not limited ₁ But once a current flows, α ₁ Just-in-phase signal U _U ，U _V ，U _W ，I _U ，I _V ，I _W Is no longer visible.

By phase-dependent signals U _U ，U _V ，U _W ，I _U ，I _V ，I _W Correspondingly determining an uncorrected rotational angular position alpha _phase Determining rotor angle as described aboveThe actual angular position α can be determined with a particularly good approximation by ₁ ：/>

However, only when the corresponding phase signal U is determined _U ，U _V ，U _W ，I _U ，I _V ，I _W Without the voltage of the electrical storage device S being regulated by the switching intervention of the charging regulator 40 in the time frame of (a) the previous determination of the rotor angle for high accuracy can be used without problemsOr rotor 32 rotation angle position alpha ₁ Or an assumption made of the rotational speed n. Thus, this assumption can be applied in sections at most to the respective switching states of the charging regulator 40. Furthermore, separate sets of machine parameters for determining the rotor angle have to be stored for the respective switching states or determined during the operation of the electric machine (see fig. 10a-10d and fig. 12). Furthermore, in determining the rotor angle ∈ ->The transient transition states in the form of dynamic starting processes between the switching states should also be taken into account (see fig. 11). This will be discussed further below. Thus, by means of a further configuration of the method, a continuous determination of the rotor angle ∈can be reliably ensured>As described in more detail in fig. 7-12.

The motor 30 of fig. 2b is again schematically shown in an enlarged form in fig. 6 a. The electric machine 30 has a rotor 32 with field windings and a stator 33 with stator windings, wherein the rotor 32 has a shaft 17. This is therefore an excited machine, which is common in particular in the case of motor vehicles. However, in particular for motor-limited vehicles, especially for small and light motor-limited vehicles, engines with permanent magnets, i.e. permanent magnet excited motors, are mostly used. In principle, two types of electric machines can be used within the scope of the invention, wherein in particular the charging regulator LR according to the invention is independent of the use of a corresponding type of electric machine, a permanent-magnet-excited electric machine or a separately excited electric machine.

The electric machine 30 is embodied as a three-phase generator, in which three phase voltage signals are induced, which are 120 ° phase-shifted to each other. Such three-phase generators are generally used as generators in modern motor vehicles and are suitable for the use of a charging regulator according to the invention connected downstream of said generator. In principle, all motors can be used without depending on their phase numbers within the scope of the invention.

The U-shaped structure is used for the U-shaped structure,v, W represent three phases of the three-phase generator 30. Via a positive diode D configured as a first path 34a _H And a negative diode D of the second path 35a _L The rectifier element 36 of (2) has a voltage U falling across the phases _U ，U _V ，U _W Rectifying. Thus, there is a generator voltage U between poles B+ and B _G The negative electrode is grounded in the case of this generator voltage. Such a three-phase generator 30 supplies power to, for example, a battery B or other consumer in the vehicle electrical system 110.

Furthermore, the charging regulator LR is provided with a control unit 40a, which control unit 40a is controlled by the generator voltage U _G The switch 42a is fed and is actuated with a voltage regulation of the battery B, so that the paths 34a,35a of the rectifier 36 are short-circuited. In order to prevent a parallel short circuit of battery B, a further diode D is provided, which is arranged after rectifier 36 to prevent a parallel short circuit of battery B. In the open state of the switch 42a, the rectifier 36 operates normally and thus applies electrical energy to the battery B or the electrical storage S.

Another embodiment of the charge regulator LR is shown in fig. 6 b. Elements identical or similar to those of the first embodiment (see fig. 6 a) are denoted by the same reference numerals or by the same reference numerals with the addition of the other letter b. With regard to this embodiment and with regard to the other embodiments which remain behind, the basic description of the already known elements refers in principle to the corresponding description of these embodiments, and only changes with respect to the other examples are shown with regard to the corresponding description.

In this exemplary embodiment, the schematically illustrated two-phase motor 30 is based on phases U and V, in which the phase voltages U are present in each case _U And U _V . Strictly speaking, fig. 4a shows a single-phase motor with coil ends led out at both ends. The single-phase motor is composed of two coils, one end of which is led out and the other end of which is connected, and thus is structurally a single-phase motor. The embodiment is peculiar in that the control unit 40b is arranged in the engine control device 122, which control unit 40b acts on the switch 42b for charge regulation and for the first branch 34b or for the rectifier 36The second branch 35b is short-circuited. A rotation speed detecting device 45 is also disposed in the engine control device 122. The rotation speed detection device has a communication connection 46 to a signal generator 47, which signal generator 47 is connected to at least one of the phases (V) in order to determine the phase voltage U _U ，U _V Is required for determining the rotational speed n of the motor 30 _U Or FL _V . The principle determination of the rotational speed n has been described initially (in particular with reference to fig. 4 b).

Fig. 6c shows a further embodiment of the charging regulator LR. Here, too, the switch 42c is actuated again by the control unit 40c, wherein when the switch 42c is in the closed position, it is switched on and the branches 35c or 34c of the rectifier 36 (the devices required for this are not shown) are correspondingly short-circuited. In this case, this occurs phase by phase corresponding to phases U, V, W, since here each phase is associated with a diode D1 to D3. Depending on the phases, the corresponding phases are short-circuited and overcharge of battery B is prevented. Here, diode D of first branch 34c of rectifier 36 _H Preventing battery B from being shorted in the event of a short circuit of the corresponding phases U, V, W.

Transistors may also be used in the upper path 34c and diodes in the lower path 35c for this purpose. In this case, the current flow I is regulated by a short circuit via the upper path 34c, while the lower path 35c avoids a short circuit of the battery B (corresponding devices not shown).

Another embodiment of the charge regulator LR is depicted in fig. 6 d. The second path 35d of the rectifier 36 has in each case a switch 42d in the form of a transistor for each phase U, V, W, the switch 42d being shown in the form of a MOSFET transistor as a transistor with a corresponding reverse diode. The transistors have both a rectifying function in the lower path 35d of the rectifier and a shorting function of the respective phase associated with the respective transistor, respectively. The rectifier 36 can thereby be short-circuited by a corresponding actuation of the corresponding transistor 42d by the control unit 40d, and thus the current flow I into the battery B can be inhibited. Here again through diode D in the first path 34D _H Preventing short circuit of battery B.

Depicted in FIG. 6eAnother embodiment of the charging regulator LR is described. The first path 34e is here equipped with a transistor T _H And the second path 35e is equipped with a transistor T _L These transistors are associated with respective phases U, V, W. Corresponding transistor T _H ，T _L Can be acted upon in each case by the control unit 40e in such a way that the phase voltage U can be produced in each case _U ，U _V ，U _W In turn, may short circuit the respective paths 34e,35e for charge regulation of battery B.

In the present case, the control unit 40e is arranged separately from the engine control device 122, wherein the two are connected to each other by means of a data connection 125e for exchanging data or for controlling the control unit 40e by the engine control device 122 or controlling the engine control device 122 by the control unit 40 e. In the case of charge regulation, the respective transistor T is controlled in the respective path 35e,34e _H ，T _L Making them conductive. To protect battery B, corresponding transistors T of the other path, respectively _H ，T _L It should be switched to the cut-off direction, respectively, so that the battery B is prevented from being shorted.

Another embodiment of the charge regulator LR is shown in fig. 6 f. Here, this embodiment differs from the embodiment shown in fig. 6d only in that both the engine control device 122 and the control unit 40f are structurally accommodated in a common housing, which provides a synergistic advantage for controlling the internal combustion engine 112 or the electric machine 30, respectively.

In principle, it is readily understood that the computing unit 40 or the engine control device 122 can be accommodated separately or together in a common housing.

Fig. 7a and 7b show the regulation of the operating voltage Us of the electrical storage S according to the first exemplary embodiment (fig. 7 a) and according to an alternative second exemplary embodiment (fig. 7 b). In the figure, the phase voltage U is plotted on the left vertical axis _U，V，W One, the operating voltage U of the electrical storage S is plotted on the right vertical axis _S And time is plotted in arbitrary units on the horizontal axis. Furthermore, the operating voltage U of the electrical storage is shown by a dashed line _S Upper threshold U of (2) _Soll1 And a lower threshold U _Soll2 In the followingWhen these thresholds are reached or lower and/or exceeded, a corresponding voltage regulation by the voltage regulator LR or 40 is initiated (see fig. 6a to 6 f).

In the graph, the phase voltage U _U，V，W As a solid line shows the operating voltage U of the electrical storage S _S Shown as a dash-dot line. The description of the diagrams from fig. 7 is similar to the description of the diagrams from fig. 8a,8b and 9, and reference is therefore generally made herein to these illustrations as well. In principle, it is readily understood that the phase voltages U shown here are merely exemplary _U The voltages selected as voltages of a single-phase motor or of an exemplary phase of a multiphase machine, wherein the illustration of the method according to the invention can also be carried out on other phases of a multiphase motor, and the analyses of the respective phases can also be combined with one another.

As can be seen in FIG. 7a, there is a rising edge Fl _U And falling edge Fl _D After a first occurrence of a half-wave of the phase voltage of the electrical storage S, the operating voltage U of the electrical storage S _S Exceeding the upper threshold U _Soll1 . In addition, after the first half wave, further edges Fl can be detected _U At the beginning of said other edge by a phase voltage U _U Characteristic value W of (2) _Uu To identify. Due to the basis of the characteristic value W _Uu Reliably identifies the rising edge Fl of the phase voltage _U Thus, the value W can be reached _Uu The regulation intervention is then carried out by means of the charge regulator 40 of the electrical storage S, whereby the phase voltage U is limited in particular _U And thereby inhibit at least from this phase U _U To charge the electrical storage S. Until reaching and falling edge Fl _D Associated next feature value W _UD The control intervention of the control 40 is triggered again, since the operating voltage U of the electrical storage S _S Again to below the upper threshold U _Soll1 . Other threshold U _Soll2 Illustrating the operating voltage U of the electrical storage S _S In which the regulator intervention is resumed and the electrical storage S is charged again.

As can be seen in fig. 7a, the phase voltage U _U The rising edge of the second half-wave of (2) appears W _Uu And falling edge occurrence W _UD Between which are locatedIs sufficient to regulate the operating voltage U of the electrical storage S _S . While in fig. 7b a scene similar to the one shown in fig. 7a is shown, but where at the phase voltage U _U The regulator intervention is maintained on the second and third half-waves of (a) in order to adapt the operating voltage U of the electrical storage accordingly _S So that the operating voltage drops again below the setpoint value U _Soll1 . In addition, the phase voltage U is shown in dashed lines in fig. 7 to 9 _U Is suppressed by a corresponding adjustment intervention of the charge adjuster 40.

Fig. 8a and 8b show the operating voltage U of the electrical storage S _S Is an alternative scenario of voltage regulation. Fig. 8a and 8b show the dynamic behavior of the voltage regulator 40 or its actuation, in which the operating voltage U for the electrical storage S is set _S From the intervention of regulation by means of the value W _Uu Detecting the start of an edge, i.e. triggered by the corresponding edge, wherein once the battery voltage is at U _Soll1 And U _Soll2 Within a desired range (see fig. 8 a), the regulating intervention by the regulator 40 is re-triggered, or as shown in fig. 8b, when the operating voltage U of the electrical storage S _S Has fallen below the nominal value U again _Soll2 The charge regulation by the charge regulator 40 is then re-activated. Here, the charging regulation by means of the charging regulator 40 is also performed by detecting the value W associated with the respective edge _Uu Edge-triggered, in which a reliable determination of the rotational angle position of the rotor 32 of the electric motor 30 or the rotational speed N of the rotor is always ensured. In fig. 8a, the next edge Fl is also considered _D Is a minimum time interval T of (2) _min . It is ensured here that, when a phase voltage U is detected _U And the next falling edge Fl _D Associated value W _UD When this phase voltage has already assumed a fixed value. Thus, by selecting the time interval T accordingly _min It is ensured that the voltage edges are not determined in transients but in a virtually stable fixed state, thereby ensuring an accurate determination of the rotational angle position of the rotor 32 or of its rotational speed n. To ensure that a steady state exists, a minimum interval T is below time _min In the case of (a) as in FIG. 8b after detection of the next following edge or below the corresponding operating voltage setpoint U _Soll2 The adjustment intervention of the regulator 40 is then triggered again.

In a further alternative embodiment of the actuation method of the charging regulator 40, as shown in fig. 9, the current flow I into the electrical storage S is inhibited or activated by clocked actuation of the charging regulator 40. The clocked operation preferably takes place within a half wave, so that the characteristic value W of the edge can be passed _Uu And W is _UD To determine the rising edge Fl _U And falling edge Fl _D To enable an accurate determination of the rotational angular position of the rotor 32 or its rotational speed n. Since the PWM time period is also much smaller than the time constant of the motor, the switching moments associated with determining the corresponding values are no longer important, which is why attention to the phase signals for the switching process is no longer mandatory. As shown, the operating voltage U of the electrical storage S present here _S Is almost constant. In principle, the activation time t of the regulator, which is dependent on the current applied to the electrical storage S _On Or a dead time t during which no current is applied to the electrical storage S _Off The current application of the battery may be regulated. The relevant control variable is the so-called duty cycle, which is given as the switching-on time and switching-off time adjusted by the charging regulator 40, for example:

typical frequencies of the corresponding clocked action of the regulator 40, which can be carried out by means of typical Pulse Width Modulation (PWM), are in the range between 10kHz and 100kHz, preferably 20kHz. In principle, however, the frequency should be chosen to be sufficiently large that even for high rotational speeds n, a sufficient number of switching processes can still be available between the two voltage edges. However, the frequency is preferably chosen such that it does not contribute significantly to the noise interference effect perceived by the user.

Assuming that the inertia of the electrical variable of the motor 30 is greater than the pulse width modulationThe system produces a similar behaviour, in particular in determining the rotor angle θ, as shown in the case of a linear controller (as shown in fig. 6 a). The estimation of the rotor angle, which is represented by the input variables "duty cycle" and rotational speed, or the estimation of the angular position α of the rotor 32 can therefore be carried out by means of a unique characteristic curve or by means of a characteristic curve family, as in the case of a linear regulator. Thus, as already mentioned, it is possible on the one hand to set the pulse width by correspondingly selecting the operating frequency of the pulse width modulation and by the duty cycle, and on the other hand by using the characteristic value W _Uu And W is _UD Reliably detecting edges to ensure the operating voltage U of the electrical storage S _S Which is necessary for determining the secondary variable of the pole wheel voltage or the rotational angular position of the rotor 32 and its rotational speed.

In principle, it is easy to understand that the operating voltage U of the electrical storage S _S Rating U of (2) _Soll1 Or U (U) _Soll2 May depend on different operating points of the motor or engine speeds. In addition, the operating voltage U _S Rating U of (2) _{Soll1，Soll2} It may also depend on the operating point of the internal combustion engine, such as the respective load or the mixture of fuel and combustion air (mixture).

Furthermore, by highly accurate determination of the rotational angular position θ of the rotor 32, a corresponding adjustment, preferably by short-circuiting or by releasing the load of the generator (see fig. 6a to 6 f), can also be caused, so that no adjustment intervention is performed, for example in the region where a high resolution of the rotational speed n is required or the rotational angular position θ of the rotor 32 needs to be determined. Even in the region of the injection process in which the ignition of the internal combustion engine or the injection into the internal combustion engine takes place, the ignition and injection process in turn sensitively depend on the operating voltage U of the electrical accumulator _S The regulation of the electrical storage S by the electric machine 30 can also be inhibited, so that the operating voltage U is not thereby changed _S To interfere with the corresponding injection or ignition. Furthermore, in order to ensure that the high-resolution rotational speed n of the motor 30 or its rotational angular position θ can be determined as well as possible, a constant angular range with respect to the zero position of the rotor is possibleThe controller intervention takes place within the enclosure, so that the rotational angle position θ or the rotational speed N can always be determined with high accuracy.

In a further alternative embodiment of the method, the rotor angle can also be determined with a corresponding switching intervention of the voltage regulator LR(FIGS. 10-13). The formula is used generically herein

To determine rotor angle

Fig. 10a and 10b show a first output voltage U of 14 volts for the generator _G (see FIG. 10 a) and other output voltages U of 2 volts for the generator _G (see fig. 10 b), rotor angle with respect to motor speedIs a function of the corresponding characteristic curve of the model. For other output voltages U of the generator _G Corresponding characteristic curves for other machine parameters of the electric machine 30 can also be stored in a corresponding characteristic curve family O _Kenn1 ，O _Kenn2 Is a kind of medium. Depending on the switching states S1, S2 of the switching charge regulator LR, the selected characteristic curve family O can be selected accordingly _Kenn1 ，O _Kenn2 And a characteristic curve stored with the characteristic curve family, and from this, the rotor angle of the electric machine 30 is determined>

Assuming that the output voltage of the generator is approximately constant, for example, maintained at a battery voltage of approximately 14 volts, the rotor angle is thus obtained with good approximationPurely dependent on the rotational speed n of the electric machine 30, whereby the rotor angle is according to the above +.>The formula illustrated yields a corresponding characteristic curve, which is shown in fig. 10 a. The scenario illustrated in fig. 10a thus essentially reflects a first switching state S1, in which electrical energy is applied to the electrical storage S, in this case battery B.

In particular for a short-circuit switching voltage regulator LR, as shown in particular in fig. 6a to 6f, if the battery B connected to the electric machine 30 reaches a corresponding threshold value in terms of capacity or battery voltage, the voltage regulator LR switches the output of the electric generator 30 into a state resembling a short circuit. Thereby avoiding overcharge of the battery. In principle, in the event of a short circuit, the output voltage U of the generator 30 _G Depending on the topology used for the voltage regulator. In the event of a short circuit, the output voltage U _G About 0.1 to 3 volts, depending on the topology, and may also be approximately assumed constant for the topologies used separately. Whereby for an output voltage U of the generator of approximately 2 volts _G Approximately the rotor angle shown in fig. 10b is obtainedVariation of the characteristic curve with respect to the rotational speed n of the motor 30. As already mentioned at the outset, a further output voltage U for the generator _G Or corresponding characteristic curves, which also depend on other machine variables of the electric machine 30, can also be stored in corresponding characteristic curve families.

Based on these characteristics, as soon as the motor has reached a sufficiently pronounced state of equilibrium, the rotor angle in the respective switching state S1, S2 can accordingly be derived directly from the system variables (e.g. the internal resistance of the motor 30, the coil inductance and the course of the desired idling voltage and output voltage). Fig. 10c and 10d show respective alternative representations of the rotor angle determination, wherein here, corresponding to fig. 10a and 10b, the rotor angle is shown The course of the edge time with respect to at least two edges of the phase signal of the motor 30. The advantages of this illustration are: for the corresponding output voltage U of the generator _G =14 volts (see fig. 10 c) and U _G =2 volts (see fig. 10 d), ensuring rotor angle +.>Is largely linear in the course of the characteristic curve.

However, an unbalanced effect, the so-called transient state, may occur between the switching states S1 and S2 due to the switching between the switching states S1 and S2, which unbalanced effect leads to a dynamic behavior of the system variables of the electric machine 30, whereby the rotor angleIt is also possible that the time-varying course of (a) has strong oscillations (see fig. 11). These unbalanced conditions typically develop in a time window that is characteristic of the corresponding motor 30. After this time window T, typical dynamic effects for damped oscillation systems are generally damped such that they transition to a state that is well approximated to stationary, whereby the rotor angle can be determined on the basis of the system variables of the electric machine 30, as already described above>(see fig. 11), the system variables may be stored in a corresponding family of characteristic curves.

However, as shown in fig. 11 during a time period T, this dynamic behavior is typical for the motor 30 and the corresponding operating state of the motor 30, and therefore accordingly in determining the rotor angle These dynamic starting processes D can also be considered. For this purpose, rotor angle->Is essentially divided into three parts, namely a second determination rule K2 during the other switching states S2 in the left part of the diagram, wherein the charging regulator LR inhibits the transfer of electrical energy into the memory S. As shown in fig. 11, this state is substantially stationary. In this state there is a rotor angle +.>Now, at time t=0, the switching from the other switching state S2 to the charging state S1 is carried out, followed by a second portion T represented by the rotor angular dynamics. As soon as the state of charge assumes a standstill again (see fig. 11, right side), the determination rule K1 can be based on the corresponding characteristic curve or characteristic curve family O _Kenn1 Determining rotor angle->In this state S1, a rotor angle is present +.>

In the present case, five amplitude values are exemplarily illustrated in the dynamic time range TThese amplitude values are associated with the respective times t _i Association, where i=1 to 5. In the present case, these values illustrate the maximum amplitude of the dynamic rotor angle change process. Depending on the type and extent of these amplitudes, at a given time t _i Determining the rotor angle, in particular in the time range T>The amplitude variation may be taken into account. Thus, if the amplitude is #) >Amplitude variation of (2) below threshold +.>Or the corresponding threshold band->It can be assumed that the amplitude variation is approximately constant and accordingly in determining the rotor angle +.>The amplitude variation is used only as a constant. However, if the rotor angle->Is greater than the threshold +.>The dynamic behaviour within the duration T is used to determine the rotor angle +.>These can be stored in the present case as application variables, respectively, and can be used, respectively, to determine the rotor angle after the respective switching operation>The determination of the respective application variable can be carried out in particular by means of reference measurements or a suitable simulation model. It is easy to understand that the switching process can be performed at a suitable position in the working cycle of the internal combustion engine of the drive motor 30, depending on the influence of the rotational speed gradient on the suitability of the correction term. In particular, a working cycle is considered in which a rotational speed course n with a gradient that is as flat as possible, i.e. a small gradient, is desired.

A possible variant of this application may be to specify the starting duration T as a function of the respective operating parameter (for example the rotational speed of the motor or the output voltage of the motor) and to analyze the times T of the minima and maxima during the starting process _i Sum amplitudeIn order to determine the corresponding rotor angle during the start-up process in the time range T>Interpolation may be performed between the applied values of the course curves. Interpolation methods using linear interpolation, quadratic interpolation, or exponential interpolation are provided herein. Depending on the type and extent of the dynamics and the type and extent of the corresponding interpolation method, the accuracy can be adapted as required or increased almost arbitrarily depending on the numerical costs. Corresponding correction terms can then be calculated based on the interpolation method, which correction terms are to be used for correcting the rotor angle +.>

In a further alternative, it is also possible to learn, while running, that it is possible to correct the edge moments or to correct the rotor angleIs used for the correction of the correction term. This is elucidated on the basis of fig. 12. Showing a typical phase signal U that can be detected on the motor 30 _U Is a process of changing (1). At time t ₁ Here, the state of the voltage regulator is changed from the switching state S1, in which the discharge current flows into the electric storage S, to the switching state S2, in which the current flow into the electric storage S is inhibited by shorting the phases U, V, W of the motor 30. Whereby the rotor angle is increased from a first fixed height +. >Raised to other fixed height->Furthermore, the other rotor angle ∈ ->In a first time after the switching processThe dynamic process is superimposed and has a value different from the value of the calculation rule or characteristic curve family belonging to the switching state. These different dynamic rotor angle values can be determined by extrapolation of the rotational speed signal.

For this purpose, it is assumed in a first step that the rotor angle assumes its other fixed value directly after the switching processAccording to the geometrical distance between the signal edges and the fixed rotor angle +.>And->The angle difference between the signal edges is calculated, which is produced in the case of a purely fixed rotor angle behaviour. In the case of a fixed rotor angle, the first signal edge FL 'following the switching operation can be extrapolated as a function of the time interval between the signal edges preceding the switching operation or of the associated rotational speed change operation n' _UU Estimate the time interval deltat from the moment of occurrence of (a) _korr1 . An actual first signal edge FL after detection of the switching process _UU Thereafter, based on the measured time interval and the estimated time interval Δt _korr1 The difference between them determines the angle associated with this time difference and uses this angle as a measure for determining the first dynamic rotor angle after the switching process >Is used for correcting the term of the correction term. Also, it may be the second signal edge FL _Ud Determining Δt _korr2 And the associated second dynamic rotor angle +.>If the other edges following the switching process should be affected by the dynamic oscillation process of the rotor angle, the other edges can also be provided in the same wayThe edges determine the corresponding correction terms.

In order to obtain good extrapolation quality, different extrapolation methods can be used, such as linear interpolation, quadratic interpolation, exponential interpolation or spline interpolation, on the one hand. Depending on the current rotational speed course, the extrapolation method can be selected in a suitable manner. On the other hand, it is provided that the switching process and thus the determination method are carried out at a point in time when the rotational speed change process has a particularly low or known dynamics, so that rotational speed influences can be taken into account in the extrapolation in a simple manner and the result of determining the correction term is not influenced.

Instead of determining that a corresponding dynamic rotor angle is provided by adding to a fixed rotor angleAnd a correction factor by which the found rotor angle is multiplied may also be determined.

Instead of estimating the signal edge Fl 'by extrapolating the previous course of the rotational speed signal under the assumption of a fixed rotor angle' _UU ，Fl' _Ud It is also possible to use a signal change from a similar operating point of the previous revolution, which is not influenced by the switching process, and to compare the moments of occurrence of unaffected or affected signal edges and to determine therefrom the signal change for calculating the dynamic rotor angleOr correction factors of (a).

The determined correction term or correction factor and the dynamic rotor angle obtained therefrom after the switching process And in particular for determining the angular orientation a of the rotation of the rotor 32 of the motor 30. In this case, it is advantageous to determine the rotor angle +.>Machine parameters that vary over time may also be considered. This is important in particular in the case of an electric motor 30 whose characteristics change over time while it is running. Thus, in determining the rotor angle +.>The corresponding degradation effects of the motor 30 while it is running can always be taken into account.

The basis for determining the rotor angle is schematically illustrated in fig. 13A flow chart of the foregoing method. In step SU1, the motor 30 is switched on by means of the charging regulator LR, wherein a first switching state S1 or a further switching state S2 is assumed. In step SU2, the rotor angle +_ in the stationary state of motor 30 is determined by means of a corresponding determination rule K1 for the first switching state S1 and a corresponding determination rule K2 for the further switching state S2 >In a further alternative step SU3, the rotor angle is determinedThe dynamic oscillation starting process D is considered in the time range T which is typical for the respective motor 30. In a further alternative step SU4, the rotor angle is corrected +.>Corresponding corrections based on the operating parameters of the motor that vary while the motor 30 is running may also be considered. />

Claims

1. Method for determining a rotor angle (θ) of an electric machine (30) having a rotor (32) and a stator (33) having at least one phase winding (U, V, W), wherein the electric machine (30) is associated with a switched charge regulator (LR) which is designed to regulate the electric machine (30) and to apply electric energy to an electrical store (S), wherein the switched charge regulator (LR) has a first switching state (S1) in which the electric energy is applied to the electrical store (S) and a further switching state (S2) in which the electric energy is at least partially inhibited from being applied to the electrical store (S), wherein the rotor angle (θ) is determined in the first switching state (S1) by means of a first determination rule (K1) and the rotor angle (θ) is determined in the further switching state (S2) by means of a further determination rule (K2),

Wherein a dynamic oscillation starting process (D) of the electric machine (30) caused by the switching on during a time change of at least one of the machine variables on which the rotor angle (θ) is based is taken into account when determining the rotor angle (θ) of the electric machine (30) after switching on the electric machine (30) in the range of the first switching state (S1) and/or during a time duration (T) after switching on the electric machine (30) in the range of the further switching states (S2).

2. The method according to claim 1, wherein the motor (30) is actuated in the first switching state (S1) such that electrical energy is applied to the electrical storage (S), wherein the motor (30) is actuated in the other switching state (S2) such that the current flow from the motor (30) to the electrical storage (S) is regulated without load by shorting at least one of the phase windings (U, V, W) or by leaving at least one of the phase windings (U, V, W) free of current.

3. Method according to claim 1 or 2, wherein the first determination rule (K1) is based on a numerical model and/or on a characteristic curve family (O _Kenn1 ) Determining the rotor angle (θ) and/or based on other numerical models and/or on other characteristic curve families (O _Kenn2 ) -determining the rotor angle (θ).

4. Method according to claim 1 or 2, wherein, in the case of the first determination rule (K1), the voltage (U) of the electrical storage (S) is taken into account in determining the rotor angle (θ) of the electrical machine (30) _Bat )。

5. Method according to claim 1 or 2, wherein the dynamic start-up process (D) during the time variation of the rotor angle (θ) is characterized such that the amplitude (θ _i ) And/or operating parameters of the electric machine (30) and using them as a measure for taking into account the dynamic starting process (D) when determining the rotor angle (θ).

6. A method according to claim 3, wherein the family of characteristic curves (O _Kenn1 ) Or the other characteristic curve family (O _Kenn2 ) Having at least an output voltage (U) of the electric machine (30) _G ) And a rotational speed (n) or at least one phase signal (U) _U ，U _V ，U _W ，I _U ，I _V ，I _W ) Is formed by a pair of edges (Fl _Uu ，Fl _Vu ，Fl _Wu ，Fl _Ud ，Fl _Vd ，Fl _Wd ) The time between (Δt) as an input variable and the rotor angle (θ) as an output variable.

7. A method according to claim 3, wherein at least one rotor angle (θ) value stored in at least one characteristic curve family of the characteristic curve families is used in a first revolution of the rotor (32) and is corrected by a rotor angle (θ) value determined based on measurements for the duration (T) of the dynamic vibration starting process (D), wherein the corrected rotor angle (θ _korr1 ，θ _korr2 ) For other revolutions of the rotor (32).

8. The method of claim 7, wherein the other rotation is a rotation of the rotor (32) within a time frame of other dynamic vibration starting processes.

9. Method according to claim 1 or 2, wherein an undisturbed change of a motor variable of the electric machine (30) is learned at the same or similar operating point of the electric machine (30) in terms of a rotor angle change (θ), the motor variable being used for determining the rotor angle (θ), no switching process (S1, S2) occurring at the operating point, wherein a further corrected rotor angle (θ) is determined by comparing the undisturbed change of the motor variable with a change of a motor variable following at least one switching process of the switching processes (S1, S2) _korr1 ，θ _korr2 )。

10. Method according to claim 1 or 2, wherein the rotational angle of the rotor (32) is determined according to at least one rotational angular orientation (a _Phase ) To inhibit the flow of current from the motor (30) to the electrical storage (S).

11. The method according to claim 10, wherein the rotor angle (θ) is used for determining a rotational angular orientation (a) of the rotor (32) _Phase )。

12. Method according to claim 1 or 2, wherein the phase signal (U _U ，U _V ，U _W ，I _U ，I _V ，I _W )。

13. The method of claim 12, wherein the electronic circuit is an engine control device (122).

14. A computing unit designed to perform the method according to any of the preceding claims by means of a corresponding integrated circuit and/or by means of a computer program stored on a memory.

15. The computing unit of claim 14, wherein the computing unit is an engine control device (122) for an internal combustion engine (12).

16. A computer readable storage medium having a computer program stored thereon, which, when executed on a computing unit, causes the computing unit to perform the method according to any of claims 1 to 13.