Method and device for self-checking sbw steering actuator failure and steering control system

Abstract

The application discloses an SBW steer-by-wire actuator fault self-checking method, device and a steering control system, and belongs to the technical field of vehicle steering control. The method comprises the following steps: according to the influence degree of each influence factor on the response signal of the excitation signal, matching the corresponding influence factor or the combination thereof for each excitation signal, establishing and storing a first adaptive model; collecting the response signal features of the excitation signal under various fault conditions, and constructing a fault reference database; detecting each influence factor in real time and comparing the influence factor with the first adaptive model, when the influence factor or the combination thereof meets the model triggering condition, injecting the excitation signal into the motor winding according to the model; collecting the motor response signal, extracting the signal features, comparing the signal features with the fault reference database features, judging and outputting the fault type; the scheme can select the excitation signal to inject into the motor and collect the feedback data without affecting the driving safety, can identify the fault type through data comparison and analysis, and can improve the reliability of the fault self-checking result.

Description

Technical Field

[0001] This application belongs to the field of vehicle steering control technology, and relates to a self-diagnosis method, device and steering control system for SBW steer-by-wire actuator faults. Background Technology

[0002] SBW (Steer-by-Wire) is the fifth generation of automotive steering systems. It completely eliminates the mechanical hard connection between the steering wheel and the front wheels, relying on electrical signals, dual actuators, and redundant electronic control to transmit steering commands and simulate road feel, thus meeting the needs of autonomous driving.

[0003] SBW comprises two core actuators: the HWA (Hand Feel) actuator and the RWA (Ride Feel) wheel control actuator. The HWA actuator, mounted on the steering column, integrates a road feel motor, three redundant angle / torque sensors, and an independent ECU. It collects the driver's steering intentions and simulates return-to-center damping, road vibration, and limit feel to provide the driver with virtual road feedback. The RWA wheel control actuator replaces the traditional EPS steering gear, featuring a dual-winding drive motor, rack and pinion position sensor, and redundant power ECU. It receives commands to drive the rack and pinion gears to rotate the front wheels, while simultaneously collecting the actual front wheel angle and tire lateral force and transmitting this feedback to the HWA for synthesized steering feel.

[0004] From the working principles and hardware structure of the two core actuators in the SBW (Steering Wheel Steering) system, it is clear that the stability and reliability of the road feel motor and drive motor, as the core power components of the actuators, directly determine the safety and reliability of the automotive steering system. During long-term operation under complex conditions, the motor is prone to potential faults such as bearing wear, rotor eccentricity, winding insulation aging, and transmission device wear. These faults are characterized by their high degree of concealment and gradual degradation. In the early stages of a fault, the motor shows no obvious abnormal noise, vibration, or power attenuation, making it difficult to identify minor deterioration using conventional testing methods. However, as the fault progresses to the middle and later stages, it can easily lead to serious problems such as motor jamming, power interruption, and winding short circuits and burnout. This not only significantly increases maintenance costs but also creates driving safety hazards. Therefore, accurate identification of early, subtle motor faults and prediction of their health status are crucial to ensuring the reliable operation of the steering system.

[0005] Currently, the health of steering systems is mostly tested under specific conditions, such as acquiring the responses of the road feel motor and drive motor to relevant excitation signals when the vehicle is stationary. In practice, the inventors have found that the test results obtained through these methods do not accurately reflect the true health status of the motors. For example, the feedback signals obtained under different test conditions for the same excitation signal are not the same. Possible reasons include that different test conditions (such as the vehicle being in a bumpy state) can affect motor performance, or that different test conditions can affect the excitation signal (such as when the vehicle is in a complex electromagnetic environment, the output excitation signal is affected by the external environment). This reduces the accuracy and reliability of fault detection results.

[0006] In summary, improving the accuracy and reliability of fault detection for SBW steer-by-wire actuators is an urgent problem to be solved. Summary of the Invention

[0007] To address the issue of low reliability and accuracy of fault detection data for SBW (Steering-by-Wire) actuators in practical applications, this application provides a self-diagnosis method for SBW actuator faults. This method monitors environmental and driving status data during vehicle operation, autonomously selects the optimal timing for outputting the excitation signal without compromising driving safety, and acquires motor feedback data. The method then determines the type of motor fault through data comparison and analysis, thereby improving the accuracy and reliability of fault diagnosis. To implement the aforementioned self-diagnosis method, this application also proposes a self-diagnosis device for SBW actuator faults. Finally, this application presents a vehicle steering control system equipped with the aforementioned self-diagnosis device.

[0008] Firstly, a self-diagnosis method for SBW (Steering By-Wire) actuator faults is provided, specifically implemented using the following technical solutions: A self-diagnostic method for SBW (Steering By-Wire) actuator faults includes: Based on the degree of influence of each influencing factor on the response signal associated with each excitation signal, a corresponding influencing factor or combination thereof is configured for each excitation signal, and the associated storage is used as the first adaptation model; Obtain the signal characteristics of the response signals associated with each excitation signal under various fault conditions, and store them in a fault reference database. Real-time detection acquires each influencing factor and compares it with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. The response signal of the motor is acquired and its signal features are extracted. The signal features are compared with the signal features in the fault reference database to determine and output the fault type. The influencing factors include vehicle driving status data and external environmental data; The excitation signal is configured as a high-frequency, low-amplitude electrical signal, and the response signal is configured as the high-frequency response component generated only by the excitation signal in the back electromotive force signal and the stator three-phase current response signal generated during motor operation. The signal characteristics of the response signal include the characteristic frequencies of each order, amplitude attenuation, and harmonic distortion rate within the full frequency band spectrum corresponding to the response signal; The signal features of the response signal are obtained by performing a fast Fourier transform on the high-frequency response component to complete the time-domain to frequency-domain conversion, decomposing it to obtain the full-band spectrum corresponding to the response signal, and extracting the signal features from the full-band spectrum. The fault types include motor bearing wear, rotor eccentricity, and winding insulation aging.

[0009] Through the above technical solution, an excitation signal that matches the current influencing factors can be autonomously output at a specific time during vehicle movement, and the response signal can be collected and compared for analysis. This can determine the potential fault type of the motor, increase the amount of data used for fault determination (i.e., the data collection scenario), and improve the accuracy and reliability of subsequent fault diagnosis.

[0010] Optionally, the fault self-test method further includes: Based on the first adaptation model, each excitation signal and its corresponding influence factor or combination thereof are stored as a reference data group, and each reference data group is assigned a reliability score and stored in association. If the current impact factor does not meet the triggering conditions of the incentive signal, then predict the impact factor data for a future set time based on the current trend of the impact factor data. If the predicted impact factor data meets the triggering conditions for the excitation signal, then the first excitation signal matching the predicted impact factor data is output at the current time according to the first adaptation model. If the data of the influencing factors at a future set time is detected and matches the predicted data of the influencing factors, a second excitation signal that is the same as the first excitation signal is output. Obtain the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, and extract the features of the first signal and the features of the second signal respectively; The difference between the first signal feature and the second signal feature is compared to generate a feature offset. If the feature offset is lower than the set offset, the reliability score corresponding to the current reference data group is increased by the set increment. If the feature offset is not lower than the set value, the reliability score corresponding to the current reference data group will be reduced by the set reduction amount; The reliability score of each reference data group is calculated in real time or at set intervals. If the score is lower than the set value, the reference data group is deleted from the first adaptation model.

[0011] The above technical solution can verify whether the excitation signal will affect the vehicle driving status data. On the other hand, it can verify the response data of the same excitation signal under different influencing factor conditions. If the response signal is not sensitive to changes in the influencing factor data, the reliability score of the above reference data group will be enhanced, which means that the fault type reflected by the above response signal is more accurate and reliable.

[0012] Optionally, the degree of influence of each influencing factor on the response signal associated with each excitation signal can be obtained, including: Based on theoretical derivation or through correlation analysis, at least one set of influencing factors or their combination corresponding to each excitation signal are selected and stored as the initial reference data set; An SBW control panel is set up in the environmental chamber. The corresponding test environment is generated according to the influencing factors or their combination in the initial reference data group. A set excitation signal is injected into the motor and the response signal is obtained. The signal features are extracted as the initial signal features. The frequency and / or amplitude of the excitation signal are changed to generate at least one set of control excitation signals, which are stored together with the aforementioned influencing factors or their combination as at least one set of comparative reference data. The response signal output by the motor in response to the control excitation signal is obtained in the environmental chamber based on the comparative reference data set, and the signal features are extracted as control signal features. By comparing the initial signal features with the control signal features, the feature offset is obtained, and the influence of each influencing factor or its combination on the response signal associated with each excitation signal is generated based on the magnitude of the feature offset. The magnitude of the feature offset is set to be positively correlated with the degree of influence.

[0013] Through the above technical solution, the influencing factors and their combinations that are adapted to each excitation signal are obtained through data analysis and specific experiments, so as to avoid the influence of relevant influencing factors on the diagnostic results when fault detection is carried out in the later stage of vehicle movement.

[0014] Optionally, the fault self-test method further includes: Set and store the characteristic frequency and amplitude attenuation rate threshold of the response signal corresponding to reliable motor operation; Obtain the signal characteristics of the response signals associated with each reference data group and store them in chronological order; Analyze the trends of signal characteristics changes over time and establish a degradation fitting model. Calculate and output the prediction results of the remaining reliable runtime based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal.

[0015] The above technical solution can estimate the safe service life of the motor based on the signal characteristics of the response signal, which helps the system output warnings to remind the driver to maintain the vehicle.

[0016] Optionally, the excitation signal is configured as a sinusoidal high-frequency voltage pulse electrical signal, the signal frequency range is configured to be 1kHz to 10kHz, the signal amplitude is configured to be 5% to 15% of the rated drive voltage of the motor, and the excitation signal injection duration is configured to be 0.2s to 1s. The signal characteristics for acquiring the response signal also include: performing envelope spectrum demodulation processing on the full-band spectrum obtained after fast Fourier transform to amplify weak harmonic components.

[0017] Through the above technical solution, the high-frequency low-amplitude excitation signal will not affect the vehicle's power output and ride comfort. At the same time, by amplifying the fault harmonic components in the spectrum, the identification of weak fault characteristics of early wear can be improved.

[0018] Optionally, the correspondence between the signal characteristics and the fault type is set as follows: The offset of each characteristic frequency is used to characterize the wear degree of the motor bearing; The amplitude attenuation rate is used to characterize the degree of rotor eccentricity of the motor; Harmonic distortion rate is used to characterize the degree of aging of motor winding insulation.

[0019] On the other hand, a self-diagnostic device for SBW (Steering By-Wire) actuator faults is provided, which is implemented using the following technical solution: A self-diagnostic device for SBW (Steering-by-Wire) actuator faults, used to implement the aforementioned self-diagnostic method for SBW actuator faults, includes: The model building module is configured to configure corresponding influence factors or combinations thereof for each excitation signal based on the degree of influence of each influence factor on the response signal associated with each excitation signal, and generate and store them as the first adaptation model; The fault database construction module is configured to acquire the signal characteristics of the response signals associated with each excitation signal under various fault conditions, and store them as a fault reference database. The signal excitation control module is configured to detect and acquire each influencing factor in real time and compare it with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. The signal acquisition and feature extraction module is configured to acquire the response signal of the motor. The response signal is configured to be the high-frequency response component generated only by the excitation signal in the back electromotive force signal and the stator three-phase current response signal generated when the motor is running. The module performs a fast Fourier transform on the high-frequency response component to complete the time-domain to frequency-domain conversion, decomposes it to obtain the full-band spectrum corresponding to the response signal, and extracts the characteristic frequencies, amplitude attenuation and harmonic distortion rate of each order from the full-band spectrum as signal features. The fault identification output module is configured to compare the extracted signal features with the signal features in the fault reference database to determine and output the fault type. The influencing factors include vehicle driving status data and external environmental data; The excitation signal is configured as a high-frequency, low-amplitude electrical signal; The fault types include motor bearing wear, rotor eccentricity, and winding insulation aging.

[0020] Optionally, the apparatus further includes a model iterative optimization module, the model iterative optimization module comprising: The data storage unit is configured to store each excitation signal and its corresponding influence factor or combination thereof as a reference data group based on the first adaptation model, assign reliability scores to each reference data group and store them in association; The factor prediction unit is configured to generate future predicted data for the impact factor at a set time based on the trend of the current impact factor data when the current impact factor does not meet the triggering conditions of the excitation signal. The excitation output verification unit is configured to output a first excitation signal that matches the impact factor prediction data at the current time according to the first adaptation model when the impact factor prediction data meets the excitation signal triggering condition, and to detect the impact factor data at a future set time. If the impact factor prediction data matches the impact factor prediction data, the unit will output a second excitation signal that is the same as the first excitation signal. The feature comparison unit is configured to acquire the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, extract the first signal feature and the second signal feature respectively, and perform a difference comparison between the first signal feature and the second signal feature to generate a feature offset. The scoring update unit is configured to increase the reliability score corresponding to the current reference data group by a set increment when the feature offset is lower than a set offset, and decrease the reliability score corresponding to the current reference data group by a set decrement when the feature offset is not lower than a set value. The data optimization unit is configured to perform reliability scoring on each reference data group in real time or at set intervals. If the score is lower than the set score, the corresponding reference data group will be deleted from the first adaptation model.

[0021] Optionally, the device further includes a fault occurrence prediction module, the fault occurrence prediction module comprising: The threshold configuration unit is configured to set and store the characteristic frequency and amplitude attenuation rate threshold of the response signal corresponding to reliable motor operation. The feature time sequence storage unit is configured to acquire the signal features of the response signals associated with each reference data group and store them in chronological order. The degradation analysis and prediction unit is configured to analyze the trend of signal characteristic changes and establish a degradation fitting model. It calculates and outputs the prediction result of the remaining reliable running time of the motor based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal. The fault warning unit is configured to receive the predicted result of the remaining reliable running time of the motor and compare it with a warning threshold. If the result is lower than the warning threshold, a fault warning message is output.

[0022] A vehicle steering control system includes a steering wheel feel simulation assembly, a steering controller, a steering actuator, and a sensor assembly, wherein the steering actuator includes the SBW steer-by-wire actuator fault self-diagnosis device as described above.

[0023] This application includes at least one of the following beneficial effects: By monitoring environmental data and driving status data during vehicle operation, the system can autonomously select the optimal timing for outputting excitation signals and obtain motor feedback data without affecting driving safety. Then, through data comparison and analysis, the type of motor fault can be determined, thereby improving the accuracy and reliability of fault diagnosis. The ability to automatically update the adaptation relationship between various influencing factors and excitation signals ensures that the motor's response to the excitation signal is not affected by these factors, resulting in more accurate diagnostic results in the later stages. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the fault self-testing method of this application; Figure 2 This is a schematic diagram of the model data optimization method in this application; Figure 3 This is a schematic diagram showing the connection of the functional modules of the fault self-testing device in this application.

[0025] Figure reference numerals: 100, Model building module; 200, Fault database building module; 300, Signal excitation control module; 400, Signal acquisition and feature extraction module; 500, Fault identification output module; 600, Model iteration optimization module; 700, Fault occurrence prediction module. Detailed Implementation

[0026] The embodiments of this application are described in detail below, and examples of the embodiments are shown in the accompanying drawings.

[0027] In the description of this specification, the references to "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples" refer to specific features, structures, materials, or characteristics described in connection with the described embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0028] A self-diagnostic method for SBW (Steering By-Wire) actuator faults, such as Figure 1 As shown, the main steps include the following: S100: Based on the degree of influence of each influencing factor on the response signal associated with each excitation signal, configure the corresponding influencing factor or combination thereof for each excitation signal, and store them as the first adaptation model. S200: Obtain the signal characteristics of the response signals associated with each excitation signal under various fault conditions, and store them in a fault reference database. S300: Real-time detection and acquisition of each influencing factor and comparison with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. S400: Acquire the motor's response signal and extract its signal features. Compare these signal features with the signal features in the fault reference database to determine and output the fault type.

[0029] In step S100 above, the excitation signal is configured as a high-frequency, low-amplitude electrical signal, and the response signal is configured as the high-frequency response component generated solely by the excitation signal from the back electromotive force signal and the stator three-phase current response signal generated during motor operation. Specifically, the excitation signal is configured as a sinusoidal high-frequency voltage pulse electrical signal, with a signal frequency range of 1kHz to 10kHz, preferably 8kHz, and a signal amplitude of 5% to 15% of the motor's rated drive voltage (which can be set to 5% in specific implementations). The excitation signal injection duration is configured as 0.2s to 1s, thereby ensuring that it does not affect the vehicle's power output and ride comfort.

[0030] Influencing factors include vehicle driving status data and external environmental data. Specifically, driving status data includes vehicle speed, steering angle / steering rate, vehicle load, and battery condition. External environmental data includes temperature, humidity, electromagnetic interference, and road surface adhesion. These influencing factors, or combinations thereof, can alter the characteristics of the response signal under specific conditions. For example, in a strong electromagnetic interference environment, the response signal from the motor will be distorted. The resulting detection data, i.e., the signal characteristics of the response signal, is not reliable. Including these factors in fault analysis would cause the analysis results to deviate from the actual situation, reducing the reliability of the fault diagnosis results.

[0031] In step S100 of the implementation method of this application, obtaining the degree of influence of each influencing factor on the response signal associated with each excitation signal includes: S110, based on theoretical derivation or through correlation analysis, select at least one set of influencing factors or their combination corresponding to each excitation signal and store them as the initial reference data set; S120, SBW control console is set in environmental chamber, corresponding test environment is generated according to the corresponding influence factors or combinations in the initial reference data group, set excitation signal is injected into motor and response signal is obtained, and signal features are extracted as initial signal features. S130, change the frequency and / or amplitude of the excitation signal to generate at least one set of control excitation signals, and store them together with the aforementioned influencing factors or their combination as at least one set of comparative reference data. Based on the comparative reference data set, obtain the response signal output by the motor in response to the control excitation signal in the environmental chamber, and extract the signal features as control signal features. S140, compare the initial signal features with the control signal features to obtain the feature offset, and generate the degree of influence of each influencing factor or its combination on the response signal associated with each excitation signal based on the magnitude of the feature offset; Among them, the magnitude of the feature offset is positively correlated with the degree of influence.

[0032] In step S110 above, theoretical derivation refers to eliminating the correlation between influencing factors and response signals through theoretical analysis. For example, theoretical analysis can be used to directly eliminate the influence of ambient light intensity on the response signal.

[0033] The correlation analysis process includes: acquiring response signal data from various influencing factors and excitation signals under vehicle driving conditions; obtaining the correlation between individual influencing factors and signal characteristic changes of response signals using single-factor correlation analysis algorithms, such as the Pearson / Spearman correlation coefficient method, thereby eliminating individual influencing factors that may affect the diagnostic results; and obtaining the correlation between combinations of influencing factors and signal characteristics using multi-factor combination interaction mining algorithms, such as linear multiple regression algorithms, thereby eliminating combinations of multiple influencing factors that may affect the diagnostic results, i.e., not outputting excitation signals when the influencing factors meet the above conditions.

[0034] In this embodiment, the signal characteristics of the response signal include the characteristic frequencies, amplitude attenuation, and harmonic distortion rate of each order within the full-band spectrum corresponding to the response signal. Obtaining the signal characteristics of the response signal includes: performing a Fast Fourier Transform (FFT) on the high-frequency response components to complete the time-domain to frequency-domain conversion, decomposing the full-band spectrum corresponding to the response signal, and extracting signal characteristics from the full-band spectrum. Optimized, to improve the identification of weak early wear-related fault characteristics, in this embodiment, obtaining the signal characteristics of the response signal further includes: performing envelope spectrum demodulation processing on the full-band spectrum obtained after the FFT to amplify the weak harmonic components.

[0035] The above steps S120-S140, through experiments, further clarify the degree of influence of each influencing factor on the response signal associated with each excitation signal based on the data analysis results of step S110, making the first fitting model more reliable.

[0036] In step S200, the types of faults include motor bearing wear, rotor eccentricity, and winding insulation aging. Correspondingly, in this embodiment, the correspondence between signal characteristics and fault types is set as follows: the offset of each characteristic frequency is used to characterize the degree of motor bearing wear, the amplitude attenuation rate is used to characterize the degree of motor rotor eccentricity, and the harmonic distortion rate is used to characterize the degree of motor winding insulation aging. In practice, the fault type can also be determined based on the combination of signal characteristics; the specific determination process will not be elaborated here.

[0037] In practical applications, as vehicle usage time increases, various functional units within the steering system will experience varying degrees of aging or damage. The initial adaptation model preset at the vehicle's factory cannot fully adapt to the diagnosis of later faults. In the embodiments of this application, combined with... Figure 2 As shown, the fault self-diagnosis method also includes a model data optimization step: A100, based on the first adaptation model, stores each excitation signal and its corresponding influencing factor or combination thereof as a reference data set, assigning a reliability score to each reference data set and storing them accordingly. For example, the excitation signal parameters are: frequency 8kHz, amplitude 0.6V (assuming the motor's rated drive voltage is 12V, taking 5% of the rated drive voltage as the excitation signal's voltage amplitude), and injection duration 0.5S. The influencing factor parameters are: vehicle speed 0km / h~15km / h, vehicle traveling at a constant speed without steering (for simplicity, external environmental data and other vehicle driving status data are ignored). The above excitation signal and influencing factor data form a reference data set. Each reference data set is assigned a corresponding reliability score in the initial state, such as 100,000 points.

[0038] A200: Obtain the current impact factor data and compare it with the impact factors or combinations thereof stored in the first fitting model. A2101. If the current impact factor does not meet the triggering conditions of the excitation signal, i.e., no impact factor or combination thereof matching the current impact factor data is found, then the predicted impact factor data for a future set time is generated based on the trend of the current impact factor data. It should be noted that the above impact factor data includes all impact factors, and the trend of impact factor data refers to the individual trend of each impact factor, such as the trend of vehicle speed change, the trend of ambient temperature change, etc.

[0039] A2102, if the current influencing factor meets the triggering condition of the excitation signal, then the excitation signal corresponding to the current influencing factor is output according to the first adaptation model and the signal characteristics of the response signal are obtained.

[0040] A211, if the impact factor prediction data meets the triggering conditions for the excitation signal, then the first excitation signal matching the impact factor prediction data will be output at the current time according to the first adaptation model.

[0041] A212, if the predicted impact factor data does not meet the triggering conditions of the excitation signal, then proceed to step A200 to continue obtaining new impact factor data for comparison.

[0042] After the set time period, when the time reaches the aforementioned future preset moment: A220, detects and obtains the current impact factor data and compares it with the predicted impact factor data: A2210, if the current impact factor data does not match the predicted impact factor data, return to step A200.

[0043] A2220, if the current impact factor data matches the impact factor prediction data, then output the second excitation signal, which is the same as the first excitation signal; A2221, acquire the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, and extract the first signal features and the second signal features respectively; A2222, compare the difference between the first signal feature and the second signal feature to generate a feature offset; A230, compare the feature offset with a predetermined offset: A2301, if the feature offset is lower than the set offset, the reliability score corresponding to the current reference data group will be increased by the set increment. A2302 If the feature offset is not lower than the set value, the reliability score corresponding to the current reference data group will be reduced by the set reduction amount, for example, by deducting 0.1 points.

[0044] A300 assigns reliability scores to each reference data group in real time or at set intervals. If the score is lower than the set value, the reference data group is removed from the first adaptation model.

[0045] The above scheme can verify the response data of the same excitation signal under different influencing factor conditions. If the response signal is not sensitive to the changes in the influencing factor data in the current reference data set, the reliability score of the above reference data set is enhanced; otherwise, the score is reduced, and finally the built-in data relationship of the first adaptation model is updated and optimized.

[0046] In an optimized embodiment of this application, the fault self-diagnosis method further includes: S510 sets and stores the characteristic frequency and amplitude attenuation rate threshold of the response signal corresponding to reliable motor operation; S520: Obtain the signal characteristics of the response signals associated with each reference data group and store them in chronological order; S530 analyzes the trend of signal characteristic changes over time and establishes a degradation fitting model. Based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal, it calculates and outputs the prediction result of the remaining reliable running time.

[0047] The above technical solution can estimate the safe service life of the motor based on the signal characteristics of the response signal, which helps the system output warnings to remind the driver to maintain the vehicle.

[0048] To implement the aforementioned SBW (Steering By-Wire) actuator fault self-testing method, this application also proposes an SBW actuator fault self-testing device, such as... Figure 3 As shown, it includes: a model building module 100, a fault database building module 200, a signal excitation control module 300, a signal acquisition and feature extraction module 400, and a fault identification output module 500.

[0049] The model building module 100 is configured to configure corresponding influencing factors or combinations thereof for each excitation signal based on the degree of influence of each influencing factor on the response signal associated with each excitation signal, and generate and store it as a first-fit model. The fault database construction module 200 is configured to acquire the signal characteristics of the response signals associated with each excitation signal under various fault conditions, and store them as a fault reference database. The above-mentioned model building module 100 and fault database construction module 200 can be loaded into a dedicated server provided by a cloud computing provider for efficient data processing and analysis operations.

[0050] The signal excitation control module 300 is configured to detect and acquire various influencing factors in real time and compare them with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. In this embodiment, a high-frequency, low-amplitude excitation signal is injected into the drive motor winding through the inverter circuit of the vehicle motor controller. The high-frequency, low-amplitude excitation signal forms a micro-vibration excitation inside the motor.

[0051] The signal acquisition and feature extraction module 400 is configured to acquire the motor's response signal. The response signal is configured to consist of the high-frequency response component generated solely by the excitation signal from the back electromotive force signal and the stator three-phase current response signal generated during motor operation. A fast Fourier transform is performed on the high-frequency response component to complete the time-domain to frequency-domain conversion, decomposing it to obtain the full-band spectrum corresponding to the response signal. Characteristic frequencies, amplitude attenuation, and harmonic distortion rates are extracted from the full-band spectrum as signal features. In specific implementation, the back electromotive force signal and the stator three-phase current response signal generated during motor operation are acquired synchronously, the fundamental drive component is filtered out, and the high-frequency response component generated solely by the high-frequency excitation signal is extracted.

[0052] The fault identification output module 500 is configured to compare the extracted signal features with the signal features in the fault reference database to determine and output the fault type. In practical applications, the fault identification output module 500 can reuse the data processing unit in the ECU to perform data comparison and analysis.

[0053] Influencing factors include vehicle driving status data and external environment data. The excitation signal is configured as a high-frequency, low-amplitude electrical signal. Fault types include motor bearing wear, rotor eccentricity, and winding insulation aging.

[0054] In detail, in the embodiments of this application, the fault self-testing device further includes a model iteration optimization module 600, which includes: a data storage unit, a factor prediction unit, an excitation output verification unit, a feature comparison unit, a scoring update unit, and a data optimization unit.

[0055] The data storage unit is configured to store each excitation signal and its corresponding influence factor or combination thereof as a reference data set based on the first adaptation model, assign reliability scores to each reference data set and store them in association. The factor prediction unit is configured to predict and generate influence factor prediction data for a future set time based on the changing trend of the current influence factor data when the current influence factor does not meet the excitation signal triggering condition. The excitation output verification unit is configured to output a first excitation signal that matches the influence factor prediction data at the current time according to the first adaptation model when the influence factor prediction data meets the excitation signal triggering condition, and to detect the influence factor data for a future set time. If it matches the influence factor prediction data, a second excitation signal identical to the first excitation signal is output. The feature comparison unit is configured to acquire the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, extract the first signal features and the second signal features respectively, and perform a difference comparison between the first signal features and the second signal features to generate a feature offset. The scoring update unit is configured to increase the reliability score corresponding to the current reference data group by a set increment when the feature offset is lower than a set offset, and decrease the reliability score corresponding to the current reference data group by a set decrement when the feature offset is not lower than a set value.

[0056] The data optimization unit is configured to assign reliability scores to each reference data group in real time or at set intervals. If the score is lower than the set score, the corresponding reference data group will be deleted from the first adaptation model, and the data optimization will be performed on the first adaptation model.

[0057] In a further optimized version, the fault self-testing device of this application also includes a fault occurrence prediction module 700, which includes: a threshold configuration unit, a feature time sequence storage unit, a degradation analysis prediction unit, and a fault early warning unit.

[0058] The threshold configuration unit is configured to set and store the characteristic frequency and amplitude attenuation rate thresholds of the response signal corresponding to reliable motor operation. The feature timing storage unit is configured to acquire the signal characteristics of the response signals associated with each reference data group and store them in chronological order. The degradation analysis and prediction unit is configured to analyze the trend of signal characteristic changes over time and establish a degradation fitting model, calculate and output the predicted result of the remaining reliable operating time of the motor based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal. The fault early warning unit is configured to receive the predicted result of the remaining reliable operating time of the motor and compare it with an early warning threshold; if it is lower than the early warning threshold, a fault early warning message is output.

[0059] Finally, this application discloses a vehicle steering control system, including a steering wheel feel simulation assembly, a steering controller, a steering actuator, and a sensor assembly. The steering actuator includes the aforementioned SBW steer-by-wire actuator fault self-diagnosis device, so as to diagnose and predict motor faults in the actuator in advance.

[0060] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A self-diagnostic method for SBW (Steering By-Wire) actuator faults, characterized in that, include: Based on the degree of influence of each influencing factor on the response signal associated with each excitation signal, a corresponding influencing factor or combination thereof is configured for each excitation signal, and the associated storage is used as the first adaptation model; Obtain the signal characteristics of the response signals associated with each excitation signal under various fault conditions, and store them in a fault reference database. Real-time detection acquires each influencing factor and compares it with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. The response signal of the motor is acquired and its signal features are extracted. The signal features are compared with the signal features in the fault reference database to determine and output the fault type. The influencing factors include vehicle driving status data and external environmental data; The excitation signal is configured as a high-frequency, low-amplitude electrical signal, and the response signal is configured as the high-frequency response component generated only by the excitation signal in the back electromotive force signal and the stator three-phase current response signal generated during motor operation. The signal characteristics of the response signal include the characteristic frequencies of each order, amplitude attenuation, and harmonic distortion rate within the full frequency band spectrum corresponding to the response signal; The signal features of the response signal are obtained by performing a fast Fourier transform on the high-frequency response component to complete the time-domain to frequency-domain conversion, decomposing it to obtain the full-band spectrum corresponding to the response signal, and extracting the signal features from the full-band spectrum. The fault types include motor bearing wear, rotor eccentricity, and winding insulation aging.

2. The self-diagnosis method for SBW steer-by-wire actuator faults according to claim 1, characterized in that, The fault self-test method also includes: Based on the first adaptation model, each excitation signal and its corresponding influence factor or combination thereof are stored as a reference data group, and each reference data group is assigned a reliability score and stored in association. If the current impact factor does not meet the triggering conditions of the incentive signal, then predict the impact factor data for a future set time based on the current trend of the impact factor data. If the predicted impact factor data meets the triggering conditions for the excitation signal, then the first excitation signal matching the predicted impact factor data is output at the current time according to the first adaptation model. If the data of the influencing factors at a future set time is detected and matches the predicted data of the influencing factors, a second excitation signal that is the same as the first excitation signal is output. Obtain the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, and extract the features of the first signal and the features of the second signal respectively; The difference between the first signal feature and the second signal feature is compared to generate a feature offset. If the feature offset is lower than the set offset, the reliability score corresponding to the current reference data group is increased by the set increment. If the feature offset is not lower than the set value, the reliability score corresponding to the current reference data group will be reduced by the set reduction amount; The reliability score of each reference data group is calculated in real time or at set intervals. If the score is lower than the set value, the reference data group is deleted from the first adaptation model.

3. The self-diagnosis method for SBW steer-by-wire actuator faults according to claim 2, characterized in that, To obtain the degree of influence of each influencing factor on the response signal associated with each excitation signal, including: Based on theoretical derivation or through correlation analysis, at least one set of influencing factors or their combination corresponding to each excitation signal are selected and stored as the initial reference data set; An SBW control panel is set up in the environmental chamber. The corresponding test environment is generated according to the influencing factors or their combination in the initial reference data group. A set excitation signal is injected into the motor and the response signal is obtained. The signal features are extracted as the initial signal features. The frequency and / or amplitude of the excitation signal are changed to generate at least one set of control excitation signals, which are stored together with the aforementioned influencing factors or their combination as at least one set of comparative reference data. The response signal output by the motor in response to the control excitation signal is obtained in the environmental chamber based on the comparative reference data set, and the signal features are extracted as control signal features. By comparing the initial signal features with the control signal features, the feature offset is obtained, and the influence of each influencing factor or its combination on the response signal associated with each excitation signal is generated based on the magnitude of the feature offset. The magnitude of the feature offset is set to be positively correlated with the degree of influence.

4. The SBW steer-by-wire actuator fault self-diagnosis method according to claim 2, characterized in that, The fault self-test method also includes: Set and store the characteristic frequency offset and amplitude attenuation rate threshold of the response signal corresponding to reliable motor operation; Obtain the signal characteristics of the response signals associated with each reference data group and store them in chronological order; Analyze the trends of signal characteristics changes over time and establish a degradation fitting model. Calculate and output the prediction results of the remaining reliable runtime based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal.

5. The self-diagnosis method for SBW steer-by-wire actuator faults according to claim 1, characterized in that, The excitation signal is configured as a sinusoidal high-frequency voltage pulse electrical signal, with a signal frequency range of 1kHz to 10kHz, a signal amplitude of 5% to 15% of the rated drive voltage of the motor, and an excitation signal injection duration of 0.2s to 1s. The signal characteristics for acquiring the response signal also include: performing envelope spectrum demodulation processing on the full-band spectrum obtained after fast Fourier transform to amplify weak harmonic components.

6. The self-diagnosis method for SBW steer-by-wire actuator faults according to claim 1, characterized in that, The correspondence between the signal characteristics and the fault types is set as follows: The offset of each characteristic frequency is used to characterize the wear degree of the motor bearing; The amplitude attenuation rate is used to characterize the degree of rotor eccentricity of the motor; Harmonic distortion rate is used to characterize the degree of aging of motor winding insulation.

7. A self-diagnostic device for SBW (Steering By-Wire) actuator malfunctions, characterized in that, To implement the SBW steer-by-wire actuator fault self-test method as described in any one of claims 1-6, comprising: The model building module (100) is configured to configure corresponding influence factors or combinations thereof for each excitation signal according to the degree of influence of each influence factor on the response signal associated with each excitation signal, and generate and store it as the first adaptation model; The fault database construction module (200) is configured to obtain the signal characteristics of the response signals associated with each excitation signal under various fault conditions and store them as a fault reference database. The signal excitation control module (300) is configured to detect and acquire each influencing factor in real time and compare it with the influencing factors or combinations thereof stored in the first adaptation model. When one or more influencing factors meet the triggering conditions set by the first adaptation model, an excitation signal is generated or selected according to the first adaptation model and injected into the motor winding. The signal acquisition and feature extraction module (400) is configured to acquire the response signal of the motor, wherein the response signal is configured to be the high-frequency response component generated only by the excitation signal in the back electromotive force signal and the stator three-phase current response signal generated when the motor is running; and to perform a fast Fourier transform on the high-frequency response component to complete the time-domain-frequency domain conversion, decompose it to obtain the full-band spectrum corresponding to the response signal, and extract the characteristic frequencies, amplitude attenuation and harmonic distortion rate of each order from the full-band spectrum as signal features; The fault identification output module (500) is configured to compare the extracted signal features with the signal features in the fault reference database, determine and output the fault type; The influencing factors include vehicle driving status data and external environmental data; The excitation signal is configured as a high-frequency, low-amplitude electrical signal; The fault types include motor bearing wear, rotor eccentricity, and winding insulation aging.

8. The SBW steer-by-wire actuator fault self-test device according to claim 7, characterized in that, The device further includes a model iteration optimization module (600), which includes: The data storage unit is configured to store each excitation signal and its corresponding influence factor or combination thereof as a reference data group based on the first adaptation model, assign reliability scores to each reference data group and store them in association; The factor prediction unit is configured to generate future predicted data for the impact factor at a set time based on the trend of the current impact factor data when the current impact factor does not meet the triggering conditions of the excitation signal. The excitation output verification unit is configured to output a first excitation signal that matches the impact factor prediction data at the current time according to the first adaptation model when the impact factor prediction data meets the excitation signal triggering condition, and to detect the impact factor data at a future set time. If the impact factor prediction data matches the impact factor prediction data, the unit will output a second excitation signal that is the same as the first excitation signal. The feature comparison unit is configured to acquire the first response signal and the second response signal corresponding to the first excitation signal and the second excitation signal, extract the first signal feature and the second signal feature respectively, and perform a difference comparison between the first signal feature and the second signal feature to generate a feature offset. The scoring update unit is configured to increase the reliability score corresponding to the current reference data group by a set increment when the feature offset is lower than a set offset, and decrease the reliability score corresponding to the current reference data group by a set decrement when the feature offset is not lower than a set value. The data optimization unit is configured to perform reliability scoring on each reference data group in real time or at set intervals. If the score is lower than the set score, the corresponding reference data group will be deleted from the first adaptation model.

9. The SBW steer-by-wire actuator fault self-diagnosis device according to claim 8, characterized in that, The device further includes a fault occurrence prediction module (700), the fault occurrence prediction module (700) comprising: The threshold configuration unit is configured to set and store the characteristic frequency offset and amplitude attenuation rate threshold of the response signal corresponding to reliable motor operation. The feature time sequence storage unit is configured to acquire the signal features of the response signals associated with each reference data group and store them in chronological order. The degradation analysis and prediction unit is configured to analyze the trend of signal characteristic changes and establish a degradation fitting model. It calculates and outputs the prediction result of the remaining reliable running time of the motor based on the characteristic frequency offset rate and amplitude attenuation rate of the response signal. The fault warning unit is configured to receive the predicted result of the remaining reliable running time of the motor and compare it with a warning threshold. If the result is lower than the warning threshold, a fault warning message is output.

10. A vehicle steering control system, comprising a steering wheel feel simulation assembly, a steering controller, a steering actuator, and a sensor assembly, characterized in that, The steering actuator includes a self-diagnostic device for SBW steering actuator faults as described in any one of claims 7-9.

Images