WO2016096209A1

WO2016096209A1 - Method for operating a dc voltage converter

Info

Publication number: WO2016096209A1
Application number: PCT/EP2015/074837
Authority: WO
Inventors: Istvan Biro
Original assignee: Robert Bosch Gmbh
Priority date: 2014-12-15
Filing date: 2015-10-27
Publication date: 2016-06-23
Also published as: DE102014225827A1; CN107005156A

Abstract

The invention relates to a method and an arrangement (10) for operating a DC voltage converter. An algorithm is used which represents a linear MISO block.

Description

Description title

Method for operating a DC-DC converter

The invention relates to a method for operating a

DC-DC converter and an arrangement for carrying out the

Process.

State of the art

DC-DC converters, also referred to as DC / DC converters, are electrical circuits that convert a DC input voltage to a higher, lower, or inverted voltage level DC voltage. In this case, the conversion usually takes place by means of a periodically operating switch and an energy store or more

Energy storage.

In motor vehicles DC converters are used in particular

Used multi-voltage networks, such as these are used, for example, in hybrid vehicles. There are currently intense in the automotive industry

Efforts to develop hybrid propulsion systems

Motor vehicles undertaken. Typically in automotive multivoltage networks there are 12 V networks for conventional consumers and 48 V networks for the newer generation consumers. Hence the need to switch between the two networks. For this purpose, inverters (buck-boost converters) have proven to be suitable.

Disclosure of the invention Against this background, a procedure with the characteristics of the

Claim 1 and an arrangement according to claim 7 presented.

Embodiments result from the dependent claims and the description.

The method and arrangement presented herein provide a software solution that meets all requirements, exhibits sufficient dynamic performance without the use of hardware, and only requires limited computational capacity.

The presented algorithm that implements the procedure has been

originally designed as a fast reactive software component to control a DC to DC converter unit in a motor vehicle. It is necessary to ensure safe operation of both the batteries and the consumers involved, which are connected to the DC-DC converter. This leads to stringent requirements and limitations in addition to the regulation of the output voltage and the output current.

The method in turn is particularly suitable for all electrical or technical systems in which the signal propagation time is at least a factor of 10 smaller than the sampling time and in which the change of the input and

Output variables have a direct physical connection to each other.

The developed algorithm is suitable in a special way, the

Output voltage and / or the output current to a given target value to limit the output current limit, the minimum level of

To control input voltage and the maximum level of the input current using a single linear regulator operating in both the voltage and current control modes without the need for cascaded or switched mode regulators.

The algorithm performs the following tasks simultaneously:

1. Regulating the output voltage to meet a target value, with

Current limit, 2. regulating the output current to meet a target value, within the minimum and maximum levels of the output voltage,

3. limiting the output current to a variable level, capable of switching to the current regulation mode to maintain a multivalue within the minimum and maximum levels of the output voltage,

4. limiting the input current to a variable level by reducing the output voltage,

5. Protecting the electrical storage device by preventing the input voltage from falling below a variable minimum level, thereby reducing output power.

6. Limiting the power on the input or output side at a variable level by reducing the output voltage.

The presented method is characterized in that only one controller is used by the algorithm representing a voltage mode controller. The target value of the controller is modified with the processed and cascaded values of the error signals. The final result of the solution is a complex but very simple code. This is a discrete linear MISO (Multiple Input Single Output) block. All limits can be dynamically modified by the application.

This architecture can be useful if the control signal in each

Working mode is the same. The biggest advantage is that the controller complies with the linear transfer characteristic between the working limits as the system goes to different states, taking into account the working environment. Therefore, the transitions of the system are linear and continuous, which improves the stability and robustness of the control. Furthermore, the lack of cascade connections ensures better dynamic behavior for the overall system. The key element is the conversion and binding of the error signals of the inputs. The method can be used with most types of controllers.

The method described is, for example, in conjunction with a

DC-DC converter used, in which the switching elements are controlled by a time-controlled PWM generator. No hardware current controller is used. The transfer function Uout / Uin must be valid for the converter. The duration of the signals, each other

are faster by orders of magnitude than the sampling time.

Due to the relatively long sampling time, the system can not be divided into differently timed processes.

It should be noted that the automotive application environment places high demands on the voltage regulator algorithms of the converter. The main requirements of the controller comply with the requirements of the

Invers converter in both directions. A time varying nonlinear higher order power converter system is presented, which is available in all

Operating points stable.

The time variance is due to the dependence of the transfer function on the load resistance, the nonlinearity in the resistors, the inductances, the capacitances and the power transistors as well as the high variance in the input and output voltages.

Furthermore, a high dynamic for a coupled power electronics is provided in all operating points. Voltage and current limits are maintained at the input and output. The performance is held.

Furthermore, the code is adapted to limited capacities, runtime limitations and limited sampling rates. Thus, the controllers have to work at lower sampling rates.

The operation of the algorithm is basically based on the close and simple physical connection of the various inputs and outputs. This means that the input signals to each other with actual simple mathematical functions can be transformed.

Thus, an adaptive, discrete MISO multifunction algorithm with a single regulator for DC-DC converters is presented. The algorithm developed is characterized by the fact that it uses only a simple linear controller.

Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

It is understood that the features mentioned above and those yet to be explained below can be used not only in the particular combination indicated, but also in other combinations or in isolation, without departing from the scope of the present invention.

Brief description of the drawings

FIG. 1 shows an embodiment of the arrangement.

FIG. 2 shows an input current limiting unit.

FIG. 3 shows in a graph the frequency response as a function of the load

FIG. 4 shows a block diagram of a multi-input controller

Embodiments of the invention

The invention is schematically illustrated by means of embodiments in the drawings and will be described in detail below with reference to the drawings.

Figure 1 shows the structure of the arrangement or controller arrangement, which is generally designated by the reference numeral 10 and for carrying out the presented method is used. The illustration also shows a

DC-DC converter 12, which is driven by the arrangement 1 0.

In addition, the illustration shows a gain adjustment 1 4, a controller 1 6, in particular voltage regulator, in this case a PID controller, a

Output current limit 1 8, an input current limit 20, a

Input voltage limit 22. Furthermore, a unit 24 for

Load detection, a first block 28 that selects a minimum value, a second block 30 that selects a minimum value, and a member 32 that performs the function> 0. There is a tension as a controlled variable

U _se t 40, a supply voltage U _ve rsorg 42, a maximum output current I _a us, max 44, a maximum supply current rsorg.max 46 and a minimum supply voltage U _V ersor _g , min 48. Outputs of the DC-DC converter 1 2 are a voltage U 50 and a current I 52. Input of the DC-DC converter 12 and thus manipulated variable is the duty cycle 54th

As part of the voltage regulation by means of the voltage regulator 1 6 determined in the aforementioned converters, the duty cycle, the ratio between the input and output voltage. The control signal as control variable of the control is the duty cycle, which physically determines the ratio between the input and output voltage. Therefore, the control is independent of the reference signal. The control function is generally a PID block. The type of regulator can vary, however.

In the output current limit 1 8, the applied current control is an indirect method, based on the ohmic law and the

Voltage regulation. The algorithm estimates the current resistive load of DC-DC converter 10 and generates it and with it

actual maximum current a voltage value. If this value is less than the current target voltage, the target size will be replaced by the lower value.

(1 ) V /

Figure 2 illustrates in a block diagram the input current limit, which is designated in Figure 1 by reference numeral 20. The illustration shows an input

70 for a maximum input current parameter and an input 72 for a maximum output current parameter. Furthermore, an input 74 is provided for the duty cycle, size at the output 76 is a maximum output current. Also shown is an efficiency block 80 and a block 82 that selects a minimum value.

The following applies:

By exploiting the switching DC-DC converter, the

Limiting procedures are extended. The voltage ratio and efficiency of the converter, which is represented in block 80, determines the relationship between the input and output voltages. Therefore, the actual input current value can be easily converted to the output side. To

Comparing the two values will be less than the actual one

Current limit considered.

For input / output power limitation, the current limiter structure is extended with mathematical formulas to consider power limits.

In some cases, it is necessary if the input or the

Output voltage is in the highest working range. In these cases, the output power could exceed the nominal value of the unit and a

Overload situation without violating the current limits.

_Pj ,, P

j

The simple Min () function ensures that the lowest of all calculated values will always be valid. In the current control mode, using a simple logic switch in the module at the current limiter output, the structure can be switched to a full function current regulator mode, and the input becomes

Target current value specified.

When monitoring the input voltage level, which is implemented by block 22 in FIG. 1, it should be noted that in many cases the power source of the motor vehicle does not provide sufficient power to maintain the voltage level of the energy storage unit. Therefore, the converter unit must consider the minimum of the voltage on the input side and must ensure a minimum input voltage level by controlling the output power, taking into account the command of

Vehicle power management. This is the most complex situation for the controller. However, the conversion of the values for this case is also given. To increase the robustness of the system, an additional auxiliary controller is integrated into the control loop. This reduces the actual target voltage of the main regulator when the input voltage reaches the minimum level and forms a reverse control loop.

In the parallel work modifier, simply summing the

modifying signals of the inputs to parallel processing, a stronger limitation may be active at the same time, however, the system provides the highest possible performance without expensive determining software.

The various delimiters establish a natural order of priority for the physical behavior of the system.

The natural order is:

0. maximum supply current

1 . maximum output current

2. Holding the output voltage or the output current at the target value

A lower atomic number means a higher priority. If necessary, the order can be influenced by a simple Decision logic is used to obtain a more flexible application, for example by the switch in the power control or the

Strombegrenzermodus. In the gain adaptation implemented by block 14 in Figure 1, the controlled system is heavily dependent on the load. The disadvantage of the linear regulators is that they can perform maximum dynamic performance in a relatively narrow range around a single operating point without risking instability. In response to the challenge of the time variant system, the gain scheduling method, which considers the dynamic change in control settling versus actual situation, was used to account for and compensate for the effect of the wide range of dynamically changing load.

Based on the measured data, the gain planning table was created. This method increases the robustness of the closed loop for a wide variety of external loads. With this extension, the converter can supply both an open connection and a load of 10 mOhm, similar to a short circuit. FIG. 3 explains how the transfer function changes as a function of the load. Note that in practical applications, the main controller is a single integrator part, which makes gain planning very easy. FIG. 3 shows in a graph the frequency response as a function of the load.

The size [dB] is entered at the top and the phase [deg] at the bottom. In this case, the characteristic of the open circuit of the arrangement in the case of different ohmic loads at a duty cycle of 30% is entered from an input of the duty cycle to the output to the DC-DC converter starting with 0.025 O to 100 O in seven steps.

When the system reaches a limit, the controller starts with the

Reduce output power linearly to monitor the value at the limit until it is physically possible. If the operating conditions are outside the controllable range, then the device must be in the fail-safe condition go. If the conditions are normalized then the facility will continue to work error free.

The presented process has a number of advantages. So secures the

Software control the time invariance of the controller throughout the entire

Lifetime. Due to the continuous voltage control, there are no transitions between operating ranges or operating points. The adaptive control loop achieves maximum dynamics and stability. Despite non-linear hardware elements, the process is stable and robust. The

Code size is also extremely compact.

The algorithm is easy to set, maintain and adapt to other facilities. This leads to a reduction in development time. The architecture presented may also be useful for other systems where the control signal is the same in all working modes. In addition, a transparent and flexible code is achieved.

FIG. 4 shows a block diagram of the theoretical signal flow diagram of a multi-input controller. There are provided a first input 100 for a maximum supply current, a second input 102 for a supply current, a third input 104 for a minimum

Supply voltage, a fourth input 106 for a supply voltage, a fifth input 108 for the controlled variable U _se t, a sixth input 1 10 for a maximum output current, a seventh input 1 12 for a

Output current, an eighth input 1 14 for a selection signal, whether a

Soannungs- or current control is performed, a ninth input 1 16 for a Ausgnagsspannung and a tenth input 1 18 for an estimated load.

Furthermore, an input current limit 130, which is integrated in block 18 of FIG. 1, represents an input voltage limit 132 corresponding to block 22 of FIG. 1 and an output current limit 134 corresponding to block 18 of FIG.

The illustration also shows saturation blocks 140, 142, and 144, which ensure that the outputs remain negative, and one Selection switch 146 for current limiting / regulation. Furthermore, a gain adjustment 150 corresponding to block 14 in FIG. 1 and a voltage regulation 154 corresponding to block 16 in FIG. 1 are shown.

Simply adding the errors results in parallel work. Another limit is active at the same time, resulting in a natural priority stemming from the physical behavior of the system for which the method was developed.

The natural order:

0. minimum supply voltage,

1 . maximum supply current,

2. maximum output current,

3. Hold the output voltage or output current at target value.

A lower atomic number means a higher priority.

If necessary, the order in which a simple arbitration logic is used to obtain a wider field of implication, for example the switch between the current control and the

Strombegrenzermodus.

Claims

claims

1. A method for operating a DC-DC converter (12) whose output voltage is controlled, wherein in the context of an algorithm, a controller (16), which represents a voltage mode controller and outputs as a manipulated variable a duty cycle is used, wherein the algorithm represents a linear MISO block ,

2. The method of claim 1, wherein a gain adjustment (14) is made.

3. The method of claim 1 or 2, wherein an output current limit (18) is performed.

4. The method according to any one of claims 1 to 3, wherein a

Power limitation is performed.

5. The method according to any one of claims 1 to 4, wherein the controller (16) is a PID controller is used.

6. The method according to any one of claims 1 to 5, which is used in conjunction with a DC-DC converter (12) are used, are controlled in the switching elements by a timed PWM generator.

7. Arrangement for operating a DC voltage regulator (12), the

in particular for carrying out a method according to one of claims 1 to 6, is arranged. the arrangement (10) implementing an algorithm comprising a controller (16) representing a voltage mode controller and outputting a duty cycle as a manipulated variable, the algorithm representing a linear MISO block.

8. Arrangement according to claim 8, wherein as a controller (16) a PID controller

9. Arrangement according to claim 8 or 9, which is designed for use in a motor vehicle.