Actuating an electric drive comprising two sub-machines, and a steering actuator of a steer-by-wire system

By employing observers to feedback control signals between sub-controllers in steer-by-wire systems, the method addresses 'torque fighting' and reduces energy losses, enhancing the efficiency and reliability of the steering actuator in steer-by-wire systems.

WO2024245478A1PCT designated stage expired Publication Date: 2024-12-05SCHAEFFLER TECHNOLOGIES AG & CO KG
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/DE2024/100366
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-06-01
Filing Date
2024-04-24
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

In steer-by-wire systems with a master-master concept, significant differences in control signals between redundant electrical machines lead to 'torque fighting,' resulting in unnecessary power consumption and energy losses, as minor control deviations accumulate due to integrator components working against each other.

Method used

The method involves using observers, specifically Luenberger observers, where each sub-controller supplies its control signal to the observer of the other sub-controller, allowing for the determination of non-measurable variables without sensors, and feedback of control signals to reduce deviations between the sub-machines, thereby minimizing 'torque fighting' and optimizing power usage.

Benefits of technology

This approach effectively reduces deviations between control signals of sub-machines, minimizing energy losses and maintaining efficient control of the steering actuator, ensuring reliable and cost-effective operation of the steer-by-wire system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure DE2024100366_05122024_PF_FP_ABST
    Figure DE2024100366_05122024_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to a method for actuating an electric drive (1) comprising at least two sub-machines (12a, 12b), wherein each of the sub-machines (12a, 12b) is regulated by a sub-regulator (10a, 10b) paired with the respective sub-machine (12a, 12b), and each of the sub-regulators (10a, 10b) has a monitoring unit (11a, 11b), each of the monitoring units (11a, 11b) being supplied with an actuation signal (21a, 21b) of the respective other sub-regulator (10a, 10b).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Control of an electric drive with two sub-machines and a steering actuator of a steer-by-wire system

[0002] The invention relates to a method for controlling an electric drive with at least two sub-machines. Furthermore, the invention relates to an electric drive with two sub-machines. Another subject of the invention is a steering actuator for a steer-by-wire system of a vehicle, in particular on the front axle.

[0003] Electric drives with at least two sub-machines are typically used in applications that place particularly high demands on operational reliability and availability, such as steer-by-wire systems for motor vehicles. Such steer-by-wire systems are steering systems with an electric actuator to which steering commands are transmitted exclusively electronically.

[0004] Since the proper functioning of the steering system is safety-critical, such steering systems are designed redundantly. Especially with steer-by-wire systems, the elimination of a mechanical connection between the steering wheel and the axle to be steered requires a redundant design of the steering system.

[0005] Generally, two concepts are known for a redundant design of a steering control system. On the one hand, there is the possibility of a master-slave concept, in which the steering system is generally controlled by a master control path. A redundant slave control path is only activated if the master control path fails. This allows the steering to continue via the slave control path despite a failure of the master control path.

[0006] On the other hand, the so-called master-master concept involves multiple equally important control paths for controlling the steering actuator. The electric drive of such a steering actuator can have multiple electric motors or multiple separate winding systems, with the electric motors or separate winding systems being controlled equally.

[0007] In such an electric drive with a master-master concept, in which the two winding systems or electrical machines drive the same mechanical axis, even a slight deviation in the control behavior of the two control devices typically results in significant differences in the set currents and torques acting on the common axis. This can result in opposing torques (so-called "torque fighting") that compensate for each other and make no contribution to the mechanical movement. Thus, the electrical machines require an unnecessarily high current consumption. This is especially true if the control devices of the respective separate winding circuits or electrical machines have an integrator component.In this case, the minor control deviations are accumulated by the integrating behavior up to a possible control signal limitation, whereby the redundant control devices work against each other to enable the desired steering maneuver.

[0008] DE 102020207 196 A1 already describes a method for operating a steering system of a motor vehicle. The method is designed to minimize the deviation between two master control paths by feeding the target manipulated variables of one control path back to the other control path and averaging the two target manipulated variables in the respective control paths. Such a system requires a sensor for each measured variable to provide the necessary input signals to the redundant control paths, which are used to determine the output-side, fed-back, and averaged target manipulated variables.

[0009] Against this background, the task is to provide an improved and cost-effective method for controlling a steering actuator within a steer-by-wire system with a master-master concept and at the same time to minimize the deviations between the redundant control paths.

[0010] The object is achieved by a method for controlling an electric drive with at least two sub-machines, wherein each of the sub-machines is controlled by a sub-controller assigned to the respective sub-machine, wherein each of the sub-controllers has an observer, wherein the observers are each supplied with a control signal from the respective other sub-controller.

[0011] In the method according to the invention, the sub-controllers feed their control signal to the observer of the respective other sub-controller. Based on known input variables and output variables of an observed reference system, the observer can determine unmeasurable or unmeasured variables, thus eliminating the need for sensors to measure these variables when controlling the electric drive. Such observers are particularly useful for measurements that are technically or economically difficult to implement. The sub-controllers each provide a control signal.

[0012] This control signal can be converted into a control variable in an actuator, which then acts on a controlled system. The control variable describes the variable set by the sub-machines, which directly or indirectly influences the variable to be controlled. In particular, the controlled variable can be the steering angle of the steering axle of a motor vehicle or of a single wheel of a motor vehicle. The control variable can, for example, be a torque present at the electric machine. A sub-machine can be a separate electric motor or a winding system within a motor with multiple winding systems. Thus, an electric drive can have multiple winding systems or multiple separate electric motors, which can work against each other in the event of control deviations between the control signals and generate what is known as "torque fighting".The method according to the invention preferably comprises two sub-machines, wherein these are designed, in particular, as separate winding systems of a common electrical machine. According to the invention, the control signal of one sub-controller is transmitted to the observer of the respective other sub-controller. Thus, the control signal of one sub-controller can be taken into account in the determination of the other control signal by the observer of the other sub-controller. This has a particularly positive effect on the deviation between the manipulated variables of the multiple sub-machines. The observer structure according to the invention thus reduces the deviation between the control signals of the multiple sub-machines within an economical framework.

[0013] Preferably, the observers of the sub-controllers are designed as Luenberger observers. Optionally, the observers of the sub-controllers can each simulate an integrator component.

[0014] According to an advantageous embodiment of the invention, the control signal of the respective associated sub-controller is fed to each observer. The observer can directly compare the control signal of the sub-controller associated with the observer with the other control signals fed back from the other sub-controllers. The reference system can thus itself detect a deviation between the control signals and act accordingly. Advantageously, the deviation between the control signals of the different sub-controllers can be further reduced. Additionally or alternatively, the control variable of the respective actuator can be fed back to the respective associated observer. Furthermore, the control signal of another observer can be fed to the observer, so that the respective observers each know their own control signal and the control signal of the other sub-controllers.Each sub-controller feeds its control signals to the other observers, whereby, provided that the respective sub-controller also feeds the control signal back to the respective observer, the deviations of the respective control signals can be taken into account for the observers.

[0015] According to a further advantageous embodiment of the invention, it is provided that the respective control signals of the respective other sub-controllers and / or the respective sub-controllers are fed proportionally to the respective other observer and / or the respective observers. Instead of feeding the entire control signal back to the observers, it is conceivable that only a portion of the control signals is fed back. Preferably, a specifically predefined portion of the control signal is fed back to the observers. In this case, the control signal of the respective other sub-controller can be fed back proportionally to the respective observer and, additionally or alternatively, the control signal of the respective sub-controller can also be fed back proportionally to the respective observer.

[0016] According to a further advantageous embodiment of the invention, at least one sensor is assigned to each sub-machine, with the sensor detecting a measurement signal and feeding it to the observer of the sub-controller of the respective sub-machine. In addition to the signal supplied by the other sub-controller and / or the respective control signal, a measurement signal from a sensor of one of the respective sub-machines can be fed to the observer. The measurement signal can provide real data for the reference system. Advantageously, the determination of a control signal by the observer can thus be accomplished based on real measurement data and the reference system.In combination with the supplied control signal of the respective other sub-controller and / or the respective own control signal, this allows on the one hand an optimal control of the sub-machines and on the other hand the deviation between the several sub-machines to be minimized or avoided.

[0017] The sensor measures, in particular, the existing torque, the speed, or the rotational position of the respective electric motor. Alternatively, the sensor can also measure the steering angle of the wheels to be steered or the position of the steering mechanism.

[0018] In an alternative embodiment of the invention, each observer is supplied with an additional signal, wherein the additional signal represents a target value defined by a user. The target value can be derived from another measurement signal. For example, a steering angle sensor on a steering wheel can measure the driver's desired steering movement.

[0019] A further subject of the invention is an electric drive with two sub-machines and two sub-controllers, wherein each of the sub-machines can be controlled by a sub-controller assigned to the respective sub-machine, wherein each of the sub-controllers has an observer, wherein the observers are configured such that a control signal from the respective other sub-controller is supplied to the observers.

[0020] A further subject of the invention is a steering actuator of a steer-by-wire system of a vehicle with an electric drive according to one of the above-described embodiments of the invention. The electric drive and the steering actuator can achieve the same technical effects and advantages as those already described in connection with the method according to the invention for controlling an electric drive.

[0021] The steering actuator is preferably arranged on an axle of a motor vehicle to be steered. The steering actuator is particularly preferably mounted on the front axle of a motor vehicle.

[0022] In an alternative embodiment, the steering actuator is used in an independent wheel steering system. A motor vehicle with multiple steerable wheels can have a steering actuator on each steerable wheel.

[0023] Alternatively or in addition to the advantageous embodiments of the steering actuator explained above, the advantageous embodiments and features explained in connection with the method for controlling an electric drive can also be used, alone or in combination, in the steering actuator and the electric drive.

[0024] Further details and advantages of the invention will be explained below with reference to the exemplary embodiment illustrated in the drawings. Herein:

[0025] Fig. 1 is a schematic representation of an embodiment of a

[0026] Steering actuator with an electric drive with two sub-controllers and two sub-machines;

[0027] Fig. 2 is a schematic block diagram of the electric drive according to

[0028] Fig. 1 ; and

[0029] Fig. 3a, 3b, 3c exemplary simulation results to explain the method according to the invention.

[0030] Fig. 1 shows an embodiment of a steering actuator 100, which comprises an electric drive 1 as part of a steer-by-wire system to which the invention can be applied. The electric drive 1 according to the embodiment is designed as a steering system for a steer-by-wire system for a steerable axle of a vehicle. For example, the steerable axle can be a front axle or a rear axle of a motor vehicle. Alternatively, it would also be conceivable to use the electric drive 1 in an individual wheel steering system. With steer-by-wire steering in a motor vehicle, the mechanical connection between a steering wheel and the axle to be steered is omitted. From a safety perspective, such a system requires redundant control to guarantee steerability.

[0031] The electric drive 1 is part of the steering actuator 100 and has two sub-machines 12a, 12b, wherein the sub-machines 12a, 12b each represent either a winding system of an electric motor with multiple winding systems or a separate electric machine connected by a common shaft. Furthermore, the sub-machines 12a, 12b can be controlled by a sub-controller 10a, 10b assigned to the respective sub-machines 12a, 12b. Consequently, there is redundancy in the control of the sub-machines 12a, 12b through at least two control paths, and redundancy in the sub-machines 12a, 12b. A control path can have a sub-controller 10a, 10b and an actuator 12a, 12b. It is entirely conceivable for the device to have more than two sub-machines 12a, 12b and two control paths.

[0032] A control signal 21a, 21b determined by the respective sub-controller 10a, 10b is transmitted to a sub-machine 12a, 12b. Based on these control signals 21a, 21b, the sub-machines 12a, 12b are controlled preferentially.

[0033] The at least two sub-machines 12a, 12b of the electric drive 1 are preferably arranged on a steering axle. The steering actuator 100 generates the steering movement to steer the wheels. In addition to the electric drive 1, the steering actuator 100 also has a steering gear 15, which can enable a reduction or transmission as well as a conversion from rotational to translational movement.

[0034] The steering actuator 100 is part of a steer-by-wire system, which may include a steering actuator. This steering actuator converts a user's steering movement into an electrical signal, with the steering actuator being connected to the steering actuator via electrical lines. The steering actuator 100 converts this electrical signal into a steering movement at the wheels to be steered.

[0035] Fig. 2 shows an embodiment of an electric drive 1 that can be used in the steering actuator 100. The electric drive 1 comprises two sub-controllers 10a, 10b for controlling the two sub-machines 12a, 12b, which affect the controlled system 14. Two measuring signals 23a, 23b are recorded by two sensors 13a, 13b and fed to the respective sub-controllers 10a, 10b. It is conceivable for the sensors to measure the torque, the speed, or the rotational position of the respective sub-machines 12a, 12b. The torque only requires the measurement of the total applied torque. It is conceivable to determine only the output torque of the steering actuator, not the torque at the respective sub-machines 12a, 12b. The rotational position and, to a certain extent, the speed, provide information about the steering position. Alternatively, the tire position orThe angle of a tire to be steered relative to a reference axis can be determined. Furthermore, it would be conceivable to measure a movement or position of the steering axis, in particular of the rack. Furthermore, it is conceivable that the electric drive 1, unlike that shown in Fig. 2, has additional sub-machines 12a, 12b or sub-controllers 10a, 10b or observers 11a, 11b. An input signal 9 can additionally be supplied to the sub-controller 10a, 10b.

[0036] It should be noted that an electric drive 1 can also have more than two sub-machines 12a, 12b, with each sub-machine 12a, 12b requiring control by a sub-controller 10a, 10b. Thus, such a system can also have multiple observers 11a, 11b, with each of the multiple observers 11a, 11b being able to receive the control signals 21a, 21b of the respective other sub-controllers 10a, 10b.

[0037] The subcontrollers 10a, 10b have observers 11a, 11b and the respective state feedbacks of the respective determined control signals 21a, 21b, which enable the control of the respective submachines 12a, 12b. An observer 11a, 11b can determine non-measurable or unmeasured variables based on input variables and / or output variables of an observed reference system. This advantageously enables the determination of variables that are difficult or impossible to measure.

[0038] The sub-controllers 10a, 10b each provide a control signal 21a, 21b, which serves to control the respective sub-machines 12a, 12b. Deviations in the signs of the respective control signals 21a, 21b cause the sub-machines 12a, 12b to virtually work against each other. This phenomenon is known, among other things, as a "torque fight." If the sub-machines 12a, 12b work against each other, this generates energy losses, and these energy losses can generate heat, which is not converted into a steering movement. Therefore, these deviations between the control signals 21a, 21b with different signs must be avoided.

[0039] According to the invention, the control signal 21a, 21b of a sub-controller 10a, 10b is fed back to the respective other observer 11a, 11b. For example, the control signal 21a is fed to the observer 11b. This functions analogously for the observer 11a. Feeding the control signals 21a, 21b back to the respective other observers 11a, 11b significantly reduces the possible deviations in the control signals 21a, 21b. Preferably, the observers 11a, 11b consider the respective other fed-back control signal 21a, 21b in addition to their own determined control signal 21a, 21b.

[0040] Furthermore, it is conceivable that the control signal 21a, 21b determined by the observer 11a, 11b is also fed back to the respective observer 11a, 11b. Consequently, a control signal 21a determined by the observer 11a can be fed back to the observer 11a. The observer 11a can then compare the available control signals 21a, 21b with each other. This works analogously for the observer 11b.

[0041] The sub-controller 10a, 10b thus determines a control signal 21a, 21b. This control signal 21a, 21b can be converted in a sub-machine 12a, 12b into a physical quantity corresponding to the sum of the control signals 21a, 21b, which affects the controlled variable in the controlled system 14. The measurement signal 23a, 23b generated by the respective sensors 13a, 13b can be fed back to the observers 11a, 11b, where the observers 11a, 11b again determine a control signal 21a, 21b based on the real data.

[0042] Furthermore, it is conceivable that an input signal 9 is transmitted to the respective observers 11a, 11b, wherein this input signal 9 takes into account the steering movement desired by a user. A target steering position defined by the user can thus be transmitted to the sub-controller 10a, 10b. This can be compared with the actual value within the observer 11a, 11b. Such a target steering position can be determined, in particular, by a sensor on a steering wheel of the motor vehicle.

[0043] Observers 11a and 11b are preferably Luenberger observers. Observers 11a and 11b can also be used to simulate an integrator component.

[0044] Furthermore, it is conceivable that only portions of the control signals 21a, 21b are fed back to the respective observers 11a, 11b or to the respective other observers 21a, 21b. This has a particularly advantageous effect on the efficient processing in the observers 11a, 11b and on the energy efficiency of the system.

[0045] Figs. 3a, 3b, and 3c clearly illustrate the positive effects of the method according to the invention. For this purpose, a highly simplified model of the electric machine from Fig. 2 was simulated. The diagrams show the results from observers 11a, 11b with an integrator component. The top of the diagrams always shows the measured position of sub-machines 12a, 12b, and the bottom shows the control signal 21a, 21b determined by observers 11a, 11b and the control variable 12. Both diagrams run against the simulation time. The lower diagram depicts the torques of the control signals 21a, 21b.

[0046] Furthermore, it should be mentioned that the simulated results in the time window from 0 to approximately 0.75 seconds represent a transient response, which is of no further relevance within the scope of this document.

[0047] Fig. 3a shows the measurement signal curve of the respective sub-machines 12a, 12b at the top, and the control signals 21a, 21b of the respective sub-machines 12a, 12b and the control variable 12 at the bottom without the influence of a deviation. A deviation of the control signals 21a, 21b can arise, in particular, from different measurement signals 23a, 23b. Deviations are also possible due to different internal resistances of the electrical lines and measurement inaccuracies.

[0048] However, it is clearly evident that the measurement signals 23a, 23b of the two sub-machines 12a, 12b overlap, and therefore there are no deviations in the signals 23a, 23b. The measured positions of the sub-machines 12a, 12b are almost identical. The lower diagram shows the torque curve in the form of the control signals 21a, 21b, with the two sub-machines 12a, 12b summed together to represent the control variable 12.

[0049] Here, no feedback of the control signals 21a, 21b of the respective sub-controllers 10a, 10b to the respective other observers 11a, 11b is required.

[0050] Fig. 3b and Fig. 3c each show examples with faulty measurements, where the measured positions of sub-machines 12a, 12b constantly deviate from each other. This is clearly evident from the measurement signals 23a, 23b. The two measurement signals 23a, 23b differ, particularly in their sign, leading to the so-called "torque fight."

[0051] In contrast to Fig. 3c, for the example in Fig. 3b, the control signals 21a, 21b of the respective sub-controllers 10a, 10b are not fed back to the other observer 11a, 11b. Fig. 3b clearly shows the negative effect of combining several equal control paths with observers with integrator components that provide differing control signals 21a, 21b. The integrator component exacerbates the discrepancy between the two positions of the respective sub-machines 12a, 12b. The torque diagram shows that as the simulation time progresses, the curves of the control signals 21a, 21b of the respective sub-machines 12a, 12b diverge. This leads to increased energy loss. The integrator component accumulates the deviation between the sub-machines 12a, 12b and thus a resulting difference between the torque curves of the sub-machines 12a, 12b can extend to infinity.As the simulation time increases, the energy loss caused by the torque fight also increases.

[0052] Nevertheless, the counteraction of the sub-machines 12a, 12b ensures that the controlled variable remains in a desired position. The controlled variable is shown in Fig. 3a, 3b, and 3c in the upper diagrams as a bold solid line and has no reference symbol.

[0053] Fig. 3c shows how the deviations are avoided according to the invention when the control signal 21a, 21b of a respective sub-controller 10a, 10b is fed back to the other observer 11a, 11b. The feedback of the control signals 21a, 21b to the observers 11a, 11b enables the sub-controllers 10, 10b to control the sub-machines 12a, 12b according to the deviations. Consequently, no diverging torque curves arise between the two sub-machines 12a, 12b, and thus only negligible energy losses occur. The two sub-machines 12a, 12b are close to each other in their torque curves.

[0054] Furthermore, in connection with Figures 3b and 3c, it should be noted that the type of lines in the diagrams does not allow any direct conclusions to be drawn about the affiliation to a sub-machine 12a, 12b. The dashed line in Fig. 3b has the reference symbol 23a, whereas the control signal 21a in the lower diagram has a dotted line. In the present example, these two lines 23a, 21a belong to sub-machine 12a, despite the different line types. This applies accordingly to lines 23b and 21b, which belong to sub-machine 12b. Furthermore, these explanations apply analogously to Fig. 3c. List of reference symbols

[0055] 1 - electric drive

[0056] 9 - Input signal 10a, 10b - Partial controller 11a, 11b - Observer 12 - Control variable

[0057] 12a, 12b - submachine 13a, 13b - sensor

[0058] 14 - Controlled system

[0059] 15 - Steering gear

[0060] 21a, 21b - Control signal 23a, 23b - Measuring signal 100 - Steering actuator

Claims

Patent claims 1. Method for controlling an electric drive (1) with at least two sub-machines (12a, 12b), wherein each of the sub-machines (12a, 12b) is controlled by a sub-controller (10a, 10b) assigned to the respective sub-machine (12a, 12b), wherein each of the sub-controllers (10a, 10b) has an observer (11a, 11b), wherein the observers (11a, 11b) are each supplied with a control signal (21a, 21b) of the respective other sub-controller (10a, 10b).

2. Method according to one of the preceding claims, characterized in that the control signal (21a, 21b) of the respective associated sub-controller (10a, 10b) is supplied to the observers (11a, 11b).

3. Method according to one of the preceding claims, characterized in that the respective control signals (21a, 21b) of the respective other sub-controllers (10a, 10b) and / or the respective sub-controllers (10a, 10b) are fed proportionally to the respective other observer (11a, 11b) and / or the respective observer (11a, 11b).

4. Method according to one of the preceding claims, characterized in that at least one sensor (13a, 13b) is assigned to each sub-machine (12a, 12b), wherein a measurement signal (23a, 23b) is detected by the sensor (13a, 13b) and fed to the observer (11a, 11b) of the sub-controller (10a, 10b) of the respective sub-machine (12a, 12b).

5. Electric drive (1) with two sub-machines (12a, 12b) and two sub-controllers (10a, 10b), wherein each of the sub-machines (12a, 12b) can be controlled by a sub-controller (10a, 10b) assigned to the respective sub-machine (12a, 12b), wherein each of the sub-controllers (10a, 10b) has an observer (11a, 11b), wherein the observers (11a, 11b) are configured such that a control signal (21a, 21b) of the respective other sub-controller (10a, 10b) is supplied to the observers (11a, 11b).

6. Steering actuator (100) of a steer-by-wire system of a vehicle with an electric drive (1) according to claim 5.

Citation Information

Patent Citations

  • Method and device for operating a steering system of a motor vehicle

    DE102020207196A1

  • Method for controlling a steer-by-wire system

    US20030114969A1

  • Rotary electric machine control device and electric power steering apparatus using the same

    US20200317258A1

  • Electric steering control apparatus and control method thereof

    US20220379953A1