Vehicle motor torque control method and device, storage medium and vehicle

Abstract

The embodiment of the application relates to the field of electric drive control technology of new energy vehicles, and discloses a vehicle motor torque control method and device, a storage medium and a vehicle. The method comprises the following steps: collecting motor speed, motor output torque and vehicle speed data in a real-time manner under an actual running state of the vehicle and performing filtering correction processing; based on the data, a gear wear quantification parameter is calculated; the gear wear quantification parameter is input into a model reference adaptive controller (MRAC) controller, and a reference model and an adaptive control parameter in the controller are corrected; combined with the corrected reference model and the adaptive control parameter, combined with a real-time speed tracking error, a closed-loop compensation torque is calculated, and a total compensation torque is obtained; and the total compensation torque is superimposed on a basic driving torque issued by the vehicle, and a motor execution torque adapted to the current gear wear state is obtained. Through the control method, the effect of continuously detecting and compensating for transmission characteristic deviation caused by wear can be realized.

Description

Technical Field

[0001] This invention relates to the field of electric drive control technology for new energy vehicles, specifically to a method, device, storage medium, and vehicle for controlling vehicle motor torque. Background Technology

[0002] With the continuous development of the new energy vehicle industry and the increasing market penetration rate, the number of high-mileage new energy passenger vehicles is rising year by year. During long-term use, core transmission components such as the motor output shaft, reducer gears, and rolling bearings are subjected to long-term alternating loads, inevitably leading to deterioration phenomena such as tooth surface wear, lubrication degradation, and continuous expansion of gear backlash. The equivalent stiffness and damping of the transmission system drift, and the clearance during torque commutation increases significantly, ultimately inducing periodic torsional vibration and severely affecting the overall driving smoothness of the vehicle. This is particularly pronounced under conditions such as sudden acceleration, release of the accelerator, and shifting between forward and reverse gears where torque crosses zero, resulting in significant transmission shock and vehicle jerking.

[0003] Currently, torque sway reduction strategies for electric drive controllers are typically calibrated based on the condition of new vehicles with unworn gears. Control parameters remain fixed throughout the vehicle's lifespan, failing to adapt to changes in transmission system characteristics after high mileage operation. This leads to increasingly prominent transmission shock issues in aging vehicles. Relying on traditional after-sales methods to replace reducer gears and other components results in high replacement costs per unit and does not fundamentally prevent recurrence of the problem. In existing technologies, the main solution related to gear wear monitoring is the external sensor wear monitoring scheme. This type of scheme adds a vibration sensor to the outside of the reducer housing, collecting vibration data and combining it with time-domain and frequency-domain feature analysis to achieve online monitoring of gear wear conditions.

[0004] However, the aforementioned existing technical solutions require additional sensors and wiring harnesses, resulting in high vehicle assembly and modification costs, limited layout space, and the ability to only output gear wear fault diagnosis and early warning results without further effective compensation or intervention. The functions are limited and the effects are not obvious. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this application is to provide a vehicle motor torque control method, device, storage medium and vehicle, which aims to solve the technical problems of high modification cost and ineffectiveness of gear wear fault diagnosis and early warning results obtained by using external sensors for gear wear monitoring in the prior art.

[0006] In a first aspect, embodiments of this application provide a vehicle motor torque control method, the method comprising: Real-time data collection of motor speed, motor output torque, and vehicle speed under actual vehicle operating conditions, followed by filtering and correction processing; Based on the processed motor speed, motor output torque and vehicle speed data, gear wear quantification parameters are calculated. The obtained gear wear quantification parameters are input into the Model Reference Adaptive Controller (MRAC) to correct the reference model inside the controller. The desired speed is determined based on the modified reference model, and the real-time speed tracking error is calculated. Based on the gear wear quantification parameters and the real-time speed tracking error, the adaptive control parameters are updated online using the wear-sensing adaptive law, and the closed-loop compensation torque is calculated. Based on the closed-loop compensation torque, the total compensation torque is obtained; The total compensation torque is added to the base drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and then output for execution. Repeat the entire process described above in a loop.

[0007] In this embodiment, by real-time acquisition of motor speed, motor output torque, and vehicle speed signals, no additional sensor hardware is required. Wear quantification parameters are used to continuously correct the reference model and adaptive control parameters within the MRAC controller, ensuring the controller always matches the actual dynamic characteristics of the transmission system under the current gear wear state. By cyclically executing the above process, full lifecycle monitoring of the reducer gears, from healthy to heavily worn, is achieved without phased calibration. Continuous compensation for transmission characteristic deviations caused by wear is provided at the software level, significantly improving the driving smoothness of high-mileage vehicles and reducing costs.

[0008] In one possible embodiment, obtaining the total compensation torque based on the closed-loop compensation torque includes: In the composite control mode of feedforward control plus MRAC control, based on the gear wear quantification parameters, the feedforward pre-compensation torque is generated by looking up a table through a preset gear meshing mechanism lookup model. The feedforward pre-compensation torque and the closed-loop compensation torque are combined to calculate the total compensation torque.

[0009] In this embodiment, a composite control architecture combining feedforward and MRAC is constructed. The feedforward branch quickly generates pre-compensation torque based on wear quantification parameters, which can offset the main periodic deterministic disturbances caused by gear wear within milliseconds, significantly improving the system's dynamic response speed to wear disturbances. Simultaneously, the MRAC closed-loop branch accurately eliminates the residual errors caused by feedforward modeling and random operating condition disturbances. The superposition of the dual-path compensation torque balances real-time control with steady-state accuracy, ensuring that the output of the total compensation torque is both fast and precise, effectively suppressing transmission jitter and shifting shocks caused by gear wear.

[0010] In one possible embodiment, the real-time acquisition of motor speed, motor output torque, and vehicle speed data under the actual operating conditions of the vehicle, followed by filtering and correction processing, includes: Real-time acquisition of three raw signals: motor speed, motor output torque, and vehicle speed data; The three original signals are subjected to first-order digital low-pass filtering to remove high-frequency sampling noise from the sensors, resulting in the processed motor speed, motor output torque, and vehicle speed data.

[0011] In this embodiment, by performing first-order digital low-pass filtering on the three original signals, the electrical noise and quantization noise introduced by the sensor during high-frequency sampling can be effectively eliminated, while retaining the effective low-frequency components in the signal that reflect the gear wear state.

[0012] In one possible embodiment, the calculation of gear wear quantification parameters based on the processed motor speed, motor output torque, and vehicle speed data includes: For the motor speed, the time-domain wear characteristics and wear characteristic frequencies are calculated through time-domain calculation and fast Fourier transform frequency-domain calculation; The baseline residual characteristics are calculated by performing residual calculations between the motor speed and the baseline speed under healthy gear conditions. An adaptive filtering method is used to construct a road disturbance reference signal from the vehicle speed data. The low-frequency road excitation component is adaptively separated and filtered out from the corrected motor speed, while the gear meshing component is retained as the adaptive filter wear feature. The speed fluctuation amplitude is determined based on the motor speed. The time-domain wear characteristics, wear characteristic frequencies, benchmark residual characteristics, and adaptive filtered wear characteristics are normalized and weighted and fused to obtain a comprehensive wear characteristic value. Based on the comprehensive wear characteristic value, the wear level is determined by a pre-configured hysteresis threshold determination strategy; The output includes the gear wear quantification parameters, which include the wear level, the wear characteristic frequency, and the rotational speed fluctuation amplitude.

[0013] In this embodiment, a four-layer hierarchical algorithm—time-domain computation, fast Fourier transform frequency-domain computation, model residual comparison, and adaptive filtering—is used in synergy to achieve a quantitative and comprehensive assessment of gear wear status from four dimensions: time-domain impact intensity, frequency-domain characteristic frequency, dynamic model deviation, and interference filtering. Specifically, the adaptive filtering method constructs a road disturbance reference signal using vehicle speed data, adaptively separating and filtering out low-frequency road excitation components from the speed signal, fundamentally eliminating misjudgments of wear caused by external operating conditions such as road bumps and rapid acceleration / deceleration.

[0014] In one possible embodiment, the time-domain wear characteristics and frequency-domain wear characteristics are calculated from the corrected motor speed through time-domain calculation and fast Fourier transform frequency-domain calculation, including: By applying a sliding window of fixed length to the processed motor speed, the speed sequence is extracted within the sliding window. The root mean square value, kurtosis value, peak-to-peak value, variance, and impact pulse index of the speed sequence are calculated to obtain the time-domain wear characteristics. A fast Fourier transform is performed on the motor speed signal within the sliding window to obtain the speed spectrum information. The theoretical meshing fundamental frequency of the gear is calculated by combining the real-time motor speed, reducer speed ratio, and gear tooth count. The amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands are extracted to determine the wear characteristic frequency of the gear.

[0015] In this embodiment, a fixed-length sliding window is applied to the motor speed. Within the window, the speed sequence is extracted, and multi-dimensional time-domain statistics such as root mean square (RMS), kurtosis, peak-to-peak value, variance, and impact pulse index are calculated. This allows for a comprehensive characterization of the impact intensity changes caused by gear wear in the time domain. Specifically, the RMS reflects the overall speed fluctuation level, the kurtosis is highly sensitive to the impact pulses caused by gear wear (the more severe the wear, the larger the kurtosis), and the peak-to-peak value and variance supplement the characterization of wear impact characteristics from the perspectives of amplitude range and dispersion, respectively. Simultaneously, the speed spectrum information is obtained through Fast Fourier Transform (FFT). Combined with real-time motor speed, reducer ratio, and gear tooth count, the theoretical meshing fundamental frequency of the gear is calculated. The amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands are extracted, refining the wear determination from the macroscopic intensity level in the time domain to the precise positioning level in the frequency domain, thus achieving accurate capture of wear characteristic frequencies.

[0016] In one possible embodiment, the step of inputting the obtained gear wear quantification parameters into the Model Reference Adaptive Controller (MRAC) to correct the reference model within the controller includes: The MRAC controller has pre-stored standard transmission stiffness and standard damping values ​​corresponding to healthy gear states. Based on the wear level in the gear wear quantification parameters, the standard transmission stiffness value and the standard damping value are adjusted and corrected by a preset linear decreasing mapping relationship; wherein, the higher the wear level, the greater the adjustment, so that the corrected reference model matches the actual dynamic characteristics under the current gear wear state; wherein, the corrected reference model outputs an ideal expected speed, which is used to calculate the tracking error with the actual output motor speed.

[0017] In this embodiment, the standard transmission stiffness and damping values ​​inside the MRAC controller are linearly reduced according to the wear level, allowing the dynamic parameters of the reference model to adaptively adjust in real time according to the degree of gear wear. The higher the wear level, the greater the reduction in transmission stiffness and damping, ensuring that the corrected reference model always accurately represents the ideal, disturbance-free response of the transmission system under the current wear condition. Since the parameters of the reference model match the dynamic characteristics of the actual transmission system after wear, the ideal expected speed output by the reference model is closer to the actual working conditions, making the subsequent speed tracking error calculation more accurate, thereby improving the compensation accuracy of the closed-loop compensation torque and avoiding compensation deviations and torque fluctuations caused by mismatch between the reference model and the actual system.

[0018] In one possible embodiment, the step of determining the desired rotational speed based on the modified reference model and calculating the real-time rotational speed tracking error; and updating the adaptive control parameters online using a wear-sensing adaptive law based on the gear wear quantization parameters and the real-time rotational speed tracking error to calculate the closed-loop compensation torque, includes: Based on the wear level in the gear wear quantification parameters, a corresponding base gain coefficient is determined through a preset level gain mapping relationship; wherein, the higher the wear level, the larger the base gain coefficient. Based on the wear characteristic frequency in the gear wear quantification parameters, the gain of the corresponding frequency band centered on the wear characteristic frequency is adjusted using a preset frequency segmentation gain function to determine the frequency adjustment gain coefficient. The base gain coefficient and the frequency adjustment gain coefficient are combined to obtain the target gain coefficient; Based on the target gain coefficient, the real-time speed tracking error, and the model output of the corrected reference model, the adaptive control parameters are updated online using an adaptive update rule constructed based on Lyapunov stability theory. The update amount of the controller parameters is calculated, and a closed-loop compensation torque matching the current wear disturbance is generated.

[0019] In this embodiment, by simultaneously embedding the wear level and wear characteristic frequency into an adaptive update rule based on Lyapunov stability theory, dual-dimensional dynamic tuning of the adaptive gain is achieved. On one hand, the base gain coefficient is determined based on the wear level; the more severe the wear, the greater the gain, providing stronger compensation to cope with larger wear disturbances. On the other hand, the gain of the corresponding frequency band is finely adjusted using a frequency-segmented gain function based on the wear characteristic frequency, specifically suppressing periodic speed fluctuations caused by gear meshing harmonics. The target gain coefficient after dual gain synthesis comprehensively considers the characteristics of both wear degree and wear characteristic frequency, ensuring that the generated closed-loop compensation torque accurately matches the current wear disturbance in both amplitude and frequency characteristics. Simultaneously, the adaptive update rule based on Lyapunov stability theory guarantees the global asymptotic stability of the closed-loop system, ensuring the safety and reliability of the compensation process.

[0020] In one possible embodiment, determining the wear level based on the comprehensive wear characteristic value using a hysteresis threshold determination strategy includes: The upgrade threshold and downgrade threshold of the wear level are pre-configured; wherein the upgrade threshold is higher than the corresponding downgrade threshold, forming a hysteresis interval; When the control cycle of the comprehensive wear characteristic value continuously reaching a preset number satisfies the upgrade threshold condition, the wear level is switched to the next higher level; when the control cycle of the comprehensive wear characteristic value continuously reaching a preset number satisfies the downgrade threshold condition, the wear level is switched to the next lower level.

[0021] In this embodiment, a hysteresis threshold determination strategy is used to determine the wear level. An upgrade threshold higher than the corresponding downgrade threshold forms a hysteresis interval, effectively preventing frequent jumps in wear level when the comprehensive wear characteristic value fluctuates around the threshold. Simultaneously, a level switch is triggered only when the control cycle requiring the comprehensive wear characteristic value to continuously reach a preset number of times meets the threshold condition. This further constrains the impact of instantaneous operating disturbances on level determination, ensuring the stability and reliability of wear level changes.

[0022] In one possible embodiment, after outputting the wear level, wear characteristic frequency, and speed fluctuation amplitude as the gear wear quantification parameters, the method further includes: When the wear level indicates severe wear, a wear fault warning signal is sent to the upper-level vehicle controller to prompt the user to perform inspection and maintenance.

[0023] In this embodiment, by actively sending a wear fault warning signal to the upper-level vehicle controller when the wear level reaches severe wear, the user can be promptly reminded to carry out inspection and maintenance, thus avoiding serious transmission system failures and safety hazards caused by excessive gear wear.

[0024] Secondly, embodiments of this application provide a vehicle motor torque control device, the device comprising: The data acquisition module is used to collect real-time data on motor speed, motor output torque, and vehicle speed under the actual operating conditions of the vehicle and perform filtering and correction processing. The quantization module is used to calculate gear wear quantization parameters based on the processed motor speed, motor output torque and vehicle speed data. The correction module is used to input the obtained gear wear quantification parameters into the Model Reference Adaptive Controller (MRAC) to correct the reference model inside the controller. The first calculation module is used to determine the desired speed according to the modified reference model and calculate the real-time speed tracking error; based on the gear wear quantification parameters and the real-time speed tracking error, the adaptive control parameters are updated online through the wear perception adaptive law to calculate the closed-loop compensation torque. The second calculation module is used to calculate the total compensation torque based on the closed-loop compensation torque; The execution torque generation module is used to superimpose the total compensation torque onto the basic drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and output it for execution. The repeat execution module is used to repeatedly execute all the above processes in a loop.

[0025] Thirdly, embodiments of this application provide a computer-readable storage medium storing a program for controlling the torque of a vehicle motor. When the program for the control method is executed by a processor, it implements the steps of controlling the torque of the vehicle motor as described in any embodiment of the first aspect.

[0026] Fourthly, embodiments of this application provide a vehicle, including: a vehicle body and an electronic device, wherein the electronic device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When the processor executes a program stored in the memory, it implements the vehicle motor torque control method as described in any embodiment of the first aspect. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application will be described below.

[0028] Figure 1 This is a schematic flowchart of a vehicle motor torque control method disclosed in an embodiment of this application; Figure 2This is a schematic flowchart of another vehicle motor torque control method disclosed in an embodiment of this application; Figure 3 This is a schematic diagram of a process for filtering and correcting collected data, as disclosed in an embodiment of this application. Figure 4 This is a schematic diagram of a process for calculating gear wear quantification parameters disclosed in an embodiment of this application; Figure 5 This is a schematic diagram of a process for calculating time-domain wear characteristics and wear characteristic frequencies, as disclosed in an embodiment of this application. Figure 6 This is a schematic diagram of a process for calculating closed-loop compensation torque disclosed in an embodiment of this application; Figure 7 This is a schematic diagram of a process for determining wear level disclosed in an embodiment of this application; Figure 8 This is a schematic diagram of a vehicle motor torque control device disclosed in an embodiment of this application; Figure 9 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] The embodiments of this application are described below with reference to the accompanying drawings.

[0031] This application provides a vehicle motor torque control method, device, storage medium, and vehicle. The method is applied to the electric drive controller of new energy vehicles, specifically a power domain microcontroller unit (MCU). The method operates on a hierarchical architecture of signal acquisition, wear identification, composite compensation, motor output, and closed-loop feedback. It is achieved through online gear wear detection and MRAC adaptive torque compensation, and optionally, a feedforward compensation branch can be added to form a composite control architecture of feedforward plus MRAC. All algorithms can run in the onboard power domain MCU, meeting real-time control requirements.

[0032] Figure 1 This is a schematic flowchart of a vehicle motor torque control method disclosed in an embodiment of this application; see reference. Figure 1 As shown in the embodiments of this application, a vehicle motor torque control method includes the following steps S10-S70: S10. Real-time acquisition of motor speed, motor output torque and vehicle speed data under actual vehicle operating conditions, and filtering and correction processing.

[0033] In this embodiment, the electric drive controller synchronously acquires three core operating signals—motor speed, motor output torque, and vehicle speed—at a fixed control cycle, and performs filtering and correction processing on the acquired signals. Furthermore, the vehicle's existing sensors and communication system can be used to acquire motor speed, motor output torque, and vehicle speed data during actual operation, without requiring additional sensors or other hardware.

[0034] S20. Based on the processed motor speed, motor output torque and vehicle speed data, the gear wear quantification parameters are calculated.

[0035] Specifically, the preprocessed signal is stripped of external interference and gear wear characteristics through a four-layer hierarchical algorithm, and the output is a three-dimensional quantized wear parameter including wear level, wear characteristic frequency and speed fluctuation amplitude.

[0036] S30. The obtained gear wear quantification parameters are input into the Model Reference Adaptive Controller (MRAC) to correct the reference model and adaptive control parameters inside the controller.

[0037] After the wear quantification parameters are injected into the MRAC controller, the standard transmission stiffness and standard damping values ​​are adjusted and corrected by a preset linear decreasing mapping relationship according to the wear level, so that the dynamic parameters of the reference model are adaptively adjusted in real time according to the degree of gear wear; at the same time, the wear level and wear characteristic frequency are embedded into the gain adjustment term of the adaptive law to realize the dynamic tuning of the adaptive parameters.

[0038] S40. Determine the desired speed based on the modified reference model and calculate the real-time speed tracking error; based on the gear wear quantification parameters and the real-time speed tracking error, update the adaptive control parameters online using the wear-sensing adaptive law, and calculate the closed-loop compensation torque.

[0039] Specifically, based on the modified reference model and adaptive parameters, and combined with the real-time speed tracking error, the closed-loop compensation torque is calculated according to the adaptive update rule constructed based on Lyapunov stability theory. This compensation torque is proportional to the intensity of the wear disturbance in amplitude and matches the wear characteristic frequency in frequency characteristics.

[0040] S50. Based on the closed-loop compensation torque, the total compensation torque is obtained.

[0041] S60. The total compensation torque is superimposed on the basic drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and output for execution.

[0042] During vehicle operation, when the driver depresses the accelerator pedal, the vehicle controller calculates and issues the base drive torque based on the driver's driving intention. Simultaneously, this algorithm calculates the total compensation torque based on the current gear wear state. The base drive torque and the total compensation torque are then summed to obtain the motor execution torque, which is output to the motor controller. This motor execution torque is the final torque command sent to the motor controller, which then controls the motor to output the corresponding torque to drive the vehicle.

[0043] S70. Repeat the above process in a loop.

[0044] In this embodiment, during the operation of the vehicle, the electric drive controller continuously executes all processes of steps S10 to S70 in a loop, forming a closed-loop feedback control.

[0045] As vehicle mileage increases, gear wear gradually intensifies, and the wear level, wear characteristic frequency, and other parameters output by the wear detection module are dynamically updated in sync. Based on the real-time wear status, the MRAC controller automatically and smoothly corrects the reference model parameters and adaptively adjusts the torque amplitude and frequency characteristics. The entire process requires no manual calibration or hardware repair or replacement, achieving seamless, adaptive, and smooth control of the reducer gears throughout their entire lifecycle, from brand new to heavily worn. This significantly improves the driving stability of high-mileage vehicles and reduces maintenance costs, demonstrating remarkable effectiveness.

[0046] Figure 2 This is a schematic flowchart of another vehicle motor torque control method disclosed in an embodiment of this application; see reference. Figure 2 As shown in the embodiment of this application, in step S50 above, obtaining the total compensation torque based on the closed-loop compensation torque specifically includes the following steps S501-S502: S501. In the composite control mode of feedforward control plus the MRAC control, based on the gear wear quantification parameters, the feedforward pre-compensation torque is generated by looking up a table through a preset gear meshing mechanism lookup model.

[0047] In this embodiment, to improve the dynamic response speed to deterministic and periodic disturbances caused by gear wear, a feedforward compensation branch is added to the MRAC closed-loop compensation. Using the speed fluctuation amplitude and wear characteristic frequency obtained in the above steps, a lookup table model of the preset gear meshing mechanism is used. Rapidly generate feedforward pre-compensation torque .

[0048] In this model, Γ represents a lookup table model of the gear meshing mechanism, using the wear characteristic frequency fwear and the speed fluctuation amplitude Arip as input variables, and outputting the feedforward pre-compensation torque T.ff This lookup table model is based on the gear meshing dynamics mechanism. It generates two-dimensional lookup table data by offline calibration of the optimal feedforward compensation amount under different combinations of wear characteristic frequencies and speed fluctuation amplitudes. The wear characteristic frequency *fwear* reflects the dominant frequency of gear meshing harmonics, and the speed fluctuation amplitude *Arip* reflects the intensity of the wear disturbance. Through the lookup table model, the feedforward compensation torque matching the current wear state can be quickly determined, offsetting the main periodic deterministic disturbance caused by gear backlash within milliseconds. Since the wear quantification parameters are dynamically updated in real time with the gear wear state, the feedforward compensation amount is also adaptively adjusted, achieving a dynamic match where the more severe the wear, the greater the feedforward compensation amount.

[0049] S502. Combine the feedforward pre-compensation torque and the closed-loop compensation torque to calculate the total compensation torque.

[0050] For example, the above-mentioned feedforward compensation torque With MRAC closed-loop compensation torque The total compensation torque is formed by superposition. The torque synthesis formula is: ; The final output torque of the motor The formula is: ; In the formula, This refers to the original vehicle's basic drive torque issued by the vehicle controller.

[0051] Because the feedforward branch quickly offsets the periodic wear disturbance of the main body, the MRAC closed-loop branch accurately eliminates the remaining random disturbances and model errors, thereby improving the dynamic response speed while ensuring the steady-state control accuracy.

[0052] Figure 3 This is a schematic diagram of a process for filtering and correcting collected data, as disclosed in an embodiment of this application; see reference. Figure 3 As shown in the embodiment of this application, in step S10 above, the real-time acquisition of motor speed, motor output torque, and vehicle speed data under the actual operating state of the vehicle and the subsequent filtering and correction processing specifically includes the following steps S101-S102: S101. Real-time acquisition of three raw signals: motor speed, motor output torque, and vehicle speed data.

[0053] For example, within each control cycle, the electric drive controller calculates the real-time motor speed using a motor rotor position sensor, obtains the motor output torque using a motor current sensor and the internal flux linkage estimation module of the motor controller, and acquires the real-time vehicle speed via the vehicle's CAN network. The acquisition of these three signals is completed synchronously within the same control cycle, ensuring time alignment between them. In this embodiment, the control cycle is set to 1 millisecond, meaning that the three raw signals are synchronously acquired every 1ms.

[0054] The three signals used in this embodiment are all standard sensor signals already present in the vehicle. No additional dedicated sensors or hardware devices are required, which realizes wear condition identification with zero hardware modification cost. This facilitates the deployment of these signals in existing mass-produced models and OTA remote upgrades via over-the-air (OTA) technology.

[0055] S102. Perform first-order digital low-pass filtering on the three original signals respectively to remove high-frequency sampling noise from the sensors, and obtain the processed motor speed, motor output torque and vehicle speed data.

[0056] For example, after the vehicle is powered on, the electric drive controller synchronously collects three core signals at a fixed control cycle of 1ms. To eliminate high-frequency sampling noise from the sensors, a first-order digital low-pass filter is used to preprocess the three original signals. The filtering formula is as follows: ; In the formula, The original sampled signal, The effective signal after filtering, the filter coefficients It balances noise filtering effectiveness with real-time signal performance.

[0057] It should be noted that the above filtering coefficients can be adjusted within the range of 0.01 to 0.2 according to actual application requirements to achieve the best balance between noise suppression and signal fidelity.

[0058] In this embodiment, by performing first-order digital low-pass filtering on the three original signals respectively, the electrical noise and quantization noise introduced by the sensor during high-frequency sampling can be effectively eliminated, while retaining the effective low-frequency components in the signal that reflect the gear wear state. In this embodiment, after obtaining the motor speed, motor output torque and vehicle speed data after the above filtering process, four-layer algorithm operations are performed in parallel based on the data to achieve interference removal and wear quantification.

[0059] Figure 4 This is a schematic diagram of a process for calculating gear wear quantification parameters disclosed in an embodiment of this application; see reference. Figure 4As shown in the embodiment of this application, in step S20 above, the calculation of gear wear quantification parameters based on the corrected motor speed, motor output torque, and vehicle speed data specifically includes the following steps S201-S207: S201. For the motor speed, the time-domain wear characteristics and wear characteristic frequencies are calculated by time-domain calculation and fast Fourier transform frequency-domain calculation.

[0060] Figure 5 This is a schematic diagram of a process for calculating time-domain wear characteristics and wear characteristic frequencies disclosed in an embodiment of this application; see reference. Figure 5 As shown in the embodiment of this application, in step S201 above, the calculation of the time-domain wear characteristics and wear characteristic frequencies of the corrected motor speed through time-domain calculation and fast Fourier transform frequency-domain calculation specifically includes the following steps S2011-S2012: S2011. Apply a sliding window of fixed length to the corrected motor speed, extract the speed sequence within the sliding window, calculate the root mean square value, kurtosis value, peak-to-peak value, variance and impact pulse index of the speed sequence, and obtain the time-domain wear characteristics.

[0061] S2012. Perform a fast Fourier transform on the motor speed signal within the sliding window to obtain the speed spectrum information. Combine the real-time motor speed, reducer speed ratio, and gear tooth count to calculate the theoretical meshing fundamental frequency of the gear. Extract the amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands to determine the wear characteristic frequency of the gear.

[0062] The speed ratio of the aforementioned reducer and the number of gear teeth are known constants.

[0063] For example, during the time-domain computation, a fixed-length sliding window is applied to the preprocessed motor speed signal. The speed sequence is extracted within the sliding window, and multi-dimensional time-domain statistical features of the speed sequence within the window are calculated iteratively online. These features include root mean square value, kurtosis value, peak-to-peak value, variance, and impact pulse index. This characterizes the impact intensity change caused by gear wear from a time-domain perspective, yielding time-domain wear characteristics. Simultaneously, the motor speed signal within the sliding window is padded with integer powers from 0 to 2 and then subjected to a Fast Fourier Transform (FFT) to obtain the real-time spectrum information of the speed signal. Combined with the real-time motor speed, reducer ratio, and number of gear teeth, the theoretical meshing fundamental frequency of the gear is calculated. The amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands are extracted to determine the wear characteristic frequencies of the gear.

[0064] S202. The residual characteristics are calculated by performing residual calculations between the motor speed and the reference speed under healthy gear conditions.

[0065] Based on a pre-calibrated standard transmission dynamics model under healthy gear conditions, and using the motor output torque as the model input, the reference speed response under ideal conditions is calculated. The filtered measured motor speed and the reference speed are compared using residual calculations to obtain the residual signal. Statistical analysis is performed on the residual signal to calculate the residual variance and periodicity characteristics. When the gear is in a healthy / undamaged state, the residual between the measured speed and the reference speed is close to zero-mean random noise. When gear wear occurs, tooth surface impact and backlash effects cause significant periodic pulse components in the residual signal, increasing the residual variance and exhibiting periodic characteristics related to the gear meshing frequency. By extracting the residual characteristics, early, minor wear can be detected from the perspective of dynamic model deviation, improving the sensitivity of wear detection.

[0066] S203. An adaptive filtering method is used to construct a road disturbance reference signal based on the vehicle speed data. The low-frequency road excitation component is adaptively separated and filtered out from the corrected motor speed, while the gear meshing component is retained as an adaptive filtering wear feature.

[0067] For example, a least mean square (LMS) adaptive filter is used. The road disturbance reference signal, constructed from the preprocessed vehicle CAN speed signal, is used as the filter's reference input, and the filtered motor speed signal is used as the filter's main input. The LMS adaptive filter automatically learns and estimates the influence component of road disturbance on motor speed by iteratively updating the filter weight coefficients, and adaptively separates and filters it from the main input signal.

[0068] For applications involving low-computing-power and low-cost MCUs, the LMS filter can be selectively downgraded to order 32, reducing the computational load with a slight loss of interference filtering accuracy. Through adaptive filtering, external interference such as road surface disturbances is completely eliminated from the signal separation perspective, fundamentally preventing wear misjudgments caused by external operating conditions such as road bumps and rapid acceleration / deceleration.

[0069] S204. Determine the speed fluctuation amplitude based on the motor speed.

[0070] Specifically, the peak-to-peak value of the motor speed signal is calculated within a sliding window, which is the difference between the maximum and minimum speed values ​​within the window, and is used as the speed fluctuation amplitude Arip. As gear wear increases transmission clearance, the speed impact amplitude during torque commutation increases, and the speed fluctuation amplitude increases accordingly. The speed fluctuation amplitude Arip, together with the wear characteristic frequency fwear, serves as the input variable for the feedforward lookup table model to generate the feedforward pre-compensation torque.

[0071] S205. Normalize and weight the time-domain wear features, wear feature frequencies, model residual features, and adaptive filter wear features to obtain a comprehensive wear feature value.

[0072] Specifically, after the above four-layer algorithm operation, four types of feature data are obtained: time-domain wear features, frequency-domain wear features, frequency and meshing frequency amplitude changes, dynamic model residual features, and high-frequency wear features extracted by adaptive filtering. The four types of feature data are first normalized, mapping each feature quantity to the standard interval [0, 1] to eliminate dimensional and numerical range differences between different feature quantities. Then, they are weighted and fused according to preset weight coefficients to obtain the comprehensive wear feature value S_wear, whose value range is normalized to the interval [0, 1]. The weight coefficients can be set based on the statistical analysis results of bench calibration experiments or real vehicle road spectrum data, reflecting the sensitivity and reliability of each feature quantity to the wear state.

[0073] S206. Based on the comprehensive wear characteristic value, the wear level is determined by a pre-configured hysteresis threshold determination strategy.

[0074] S207. Output the gear wear quantification parameters, including the wear level, the wear characteristic frequency, and the speed fluctuation amplitude.

[0075] In this embodiment, after determining the wear level, the system integrates time-domain, frequency-domain, residual, and filtered feature data to output three core quantization parameters: wear level Wgrade, wear characteristic frequency fwear, and speed fluctuation amplitude Arip. The wear level Wgrade is used for subsequent reference model parameter correction and adaptive gain tuning; the wear characteristic frequency fwear is used for feedforward lookup and frequency-segmented gain adjustment; and the speed fluctuation amplitude Arip is used for feedforward lookup. These three parameters together constitute a complete quantitative description of the gear wear state, providing accurate wear state input for the MRAC adaptive torque compensation module.

[0076] In one embodiment of this application, the MRAC controller has pre-stored standard transmission stiffness and standard damping values ​​corresponding to healthy gear states. In step S30 above, the obtained gear wear quantification parameters are input to the Model Reference Adaptive Controller (MRAC) to correct the reference model and adaptive control parameters inside the controller, including the following step A10: Step A10: Based on the wear level in the gear wear quantification parameters, the standard transmission stiffness value and the standard damping value are adjusted and corrected by a preset linear decreasing mapping relationship.

[0077] The higher the wear level, the greater the reduction, thereby making the corrected reference model match the actual dynamic characteristics under the current gear wear state; the corrected reference model outputs an ideal expected speed, which is used to calculate the tracking error between the actual output motor speed and the ideal expected speed.

[0078] In this embodiment, the reference model is the electro-drive mechanical model corresponding to a pre-calibrated healthy, wear-free gear, with built-in standard transmission stiffness. Damping Parameters. Unlike traditional fixed reference models, this embodiment uses real-time input wear levels. The online adaptive correction model's dynamic parameters are adjusted as wear increases, ensuring the reference model accurately represents the ideal, disturbance-free response of the transmission system under the current wear condition. The parameter correction formula is as follows: ; In the formula, The calibration correction factor for the whole vehicle is set to 0.04-0.15 to ensure smooth parameter correction without abrupt changes.

[0079] In this embodiment, the higher the wear level, the greater the reduction, thus making the corrected reference model match the actual dynamic characteristics under the current gear wear state. Specifically, the corrected reference model outputs an ideal expected speed, which is used to calculate the tracking error between the actual output motor speed and the desired speed. The ideal expected speed output by the wear-corrected reference model is calculated in real time. The actual output speed of the controlled object (drive motor + reducer assembly with worn gears) The tracking error is calculated using the following formula: .

[0080] In this embodiment, the standard transmission stiffness and damping values ​​inside the MRAC controller are linearly reduced according to the wear level, allowing the dynamic parameters of the reference model to adaptively adjust in real time according to the degree of gear wear. The higher the wear level, the greater the reduction in transmission stiffness and damping, ensuring that the corrected reference model always accurately represents the ideal, disturbance-free response of the transmission system under the current wear state. Because the parameters of the reference model match the dynamic characteristics of the actual transmission system after wear, the ideal expected speed output by the reference model is closer to the actual operating conditions, making the subsequent speed tracking error calculation more accurate, thereby improving the compensation accuracy of the closed-loop compensation torque and avoiding compensation deviations and torque fluctuations caused by the mismatch between the reference model and the actual system. This wear-aware reference model correction mechanism fundamentally solves the problem of severe mismatch between the traditional fixed reference model and the actual system after wear in high-mileage vehicles, ensuring that the reference model always provides an accurate ideal response benchmark throughout its entire life cycle.

[0081] Figure 6 This is a schematic diagram of a process for calculating closed-loop compensation torque disclosed in an embodiment of this application; see reference. Figure 6As shown in the embodiment of this application, the above-mentioned step S40, which involves determining the desired speed based on the modified reference model and calculating the real-time speed tracking error, and updating the adaptive control parameters online using the wear-sensing adaptive law based on the gear wear quantification parameters and the real-time speed tracking error to calculate the closed-loop compensation torque, specifically includes the following steps S401-S404: S401. Based on the wear level in the gear wear quantification parameters, determine the corresponding basic gain coefficient through a preset level gain mapping relationship; wherein, the higher the wear level, the larger the basic gain coefficient.

[0082] Specifically, the base gain coefficient is determined based on the wear level through a preset level gain mapping relationship. The higher the wear level, the greater the gear backlash, and the more severe the speed fluctuations and transmission shocks caused by wear, requiring a larger compensation gain to suppress the disturbances. Therefore, the base gain coefficient monotonically increases with the wear level. For example, wear level 0 corresponds to a gain coefficient of 0.01, and wear level 1 corresponds to a gain coefficient of 0.03.

[0083] S402. Based on the wear characteristic frequency in the gear wear quantification parameters, the gain of the corresponding frequency band centered on the wear characteristic frequency is adjusted using a preset frequency segmentation gain function to determine the frequency adjustment gain coefficient.

[0084] For example, the frequency adjustment gain coefficient is determined based on the wear characteristic frequency using a preset frequency segmented gain function. This function, centered on the wear characteristic frequency, refines and enhances the gain of the frequency band corresponding to the gear meshing harmonics, while appropriately attenuating the gain of other frequency bands. Through frequency segmented gain adjustment, the adaptive mechanism has a stronger targeted suppression capability for periodic speed fluctuations caused by wear. The design of the frequency segmented gain function ensures that the dominant frequency component of the adaptive compensation torque matches the wear characteristic frequency, thereby achieving precise and directional suppression of wear disturbances in the frequency domain.

[0085] S403. Combine the basic gain coefficient with the frequency adjustment gain coefficient to obtain the target gain coefficient.

[0086] For example, the synthesis method can be weighted summation or product combination. The resulting target gain coefficient comprehensively considers the macroscopic impact of wear degree and the microscopic frequency band characteristics of wear characteristic frequency, so that the output of the subsequent adaptive law is accurately matched with the current wear disturbance in both amplitude and frequency characteristics.

[0087] S404. Based on the target gain coefficient, the real-time speed tracking error, and the model output of the corrected reference model, the adaptive control parameters are updated online using an adaptive update rule constructed based on Lyapunov stability theory. The update amount of the controller parameters is calculated, and a closed-loop compensation torque matching the current wear disturbance is generated.

[0088] In this embodiment, a fundamental adaptive law is constructed based on Lyapunov stability theory to ensure the global asymptotic stability of the closed-loop system. This invention innovatively embeds wear level and wear characteristic frequency into the adaptive law gain adjustment term, achieving precise dynamic tuning of the adaptive parameters. The core formula of the adaptive law is as follows: θ

[0089] In the formula, These are adjustable parameters for the controller. This represents the base gain coefficient for wear level. A piecewise gain function is defined for the wear characteristic frequency. The higher the wear level, the greater the overall gain correction weight, providing stronger compensation. Simultaneously, the gain of the corresponding frequency band is finely adjusted based on the wear characteristic frequency to specifically suppress gear meshing harmonic disturbances, ultimately generating a closed-loop compensation torque in real time that precisely matches the current wear disturbance. .

[0090] The system performs weighted fusion of multi-dimensional feature data and makes comprehensive decisions to output the current wear level of the gear (e.g., healthy / light / moderate / heavy), precise wear feature frequency and speed fluctuation amplitude. The wear level is mapped to the compensation parameter using a linear limit, and the wear level jump does not exceed 1 level in a single control cycle, preventing sudden increases and decreases in compensation torque.

[0091] Figure 7 This is a schematic diagram of a process for determining wear level disclosed in an embodiment of this application; see reference. Figure 7 As shown in the embodiment of this application, in step S206 above, the step of determining the wear level based on the comprehensive wear characteristic value using a hysteresis threshold determination strategy includes the following steps S2061-S2062: S2061. Pre-configure the upgrade threshold and downgrade threshold of the wear level; wherein the upgrade threshold is higher than the corresponding downgrade threshold, forming a hysteresis interval.

[0092] S2062. When the control cycle of the comprehensive wear characteristic value continuously reaching a preset number satisfies the upgrade threshold condition, the wear level is switched to the next higher level; when the control cycle of the comprehensive wear characteristic value continuously reaching a preset number satisfies the downgrade threshold condition, the wear level is switched to the next lower level.

[0093] For example, after normalizing and weighting the four data sources—time-domain features, frequency-domain FFT features, dynamic model residual features, and LMS adaptive filtering—a comprehensive wear feature value with a range of [0,1] is obtained. To avoid frequent changes in wear levels caused by random fluctuations in characteristic values ​​due to instantaneous vehicle bumps, rapid acceleration and deceleration, torque reversal, etc., a hysteresis threshold determination strategy is adopted to classify wear levels. At the same time, the wear level is restricted to a jump of only ±1 level within a single control cycle to ensure smooth adjustment of control parameters.

[0094] Examples are given below, including the following specific situations: 1. Wear level upgrade judgment threshold (to trigger the upgrade of the level, 10 consecutive sets of sliding window data must stably meet the threshold condition). Level 0 (Health Wear Level) =0) Comprehensive wear characteristics: 0≤ <0.25 The gears have almost no backlash, and new cars may only experience slight wear during running-in. The compensation torque is close to zero, and the original factory electronic control calibration parameters are used.

[0095] Level 1 (Slight wear level) =1) Overall wear characteristics: 0.25≤ <0.45 The gears exhibit a small amount of normal aging backlash, with slight speed shocks only occurring during torque zero-crossing shifts and Tip-in / Tip-out conditions. A low-gain torque compensation strategy is adopted.

[0096] Level 2 (Moderate wear level) =2) Overall wear characteristics: 0.45≤ <0.70 The gear backlash is significantly increased, and the torque reversal jerking is prominent during normal driving. Medium compensation gain is activated to suppress transmission vibration.

[0097] Level 3 (Severe Wear Level) =3) Overall wear characteristics: 0.70≤ ≤1.00 The gear backlash exceeded the design allowable range, causing severe impact. The maximum compensation torque was activated to suppress vibration, and at the same time, a reducer wear fault warning signal was sent to the upper vehicle controller to remind the user to bring the vehicle in for inspection.

[0098] 2. Wear level downgrade determination hysteresis threshold (triggers downgrade switching of the wear level to prevent the wear level from oscillating back and forth due to operating condition disturbances). Downgrade from level 1 to level 0: Requires 10 consecutive sets of sliding windows to meet the requirements. <0.20; Downgrade from level 2 to level 1: Requires 10 consecutive sets of sliding windows to meet the requirements. <0.40; Downgrade from level 3 to level 2: Requires 10 consecutive sets of sliding windows to meet the requirements. <0.65.

[0099] After the above four-layer algorithm is used to complete the fusion of multi-source features, and by combining time domain, frequency domain, residual, and filtered feature data, three core quantization parameters are finally output: wear level (divided into healthy level 0 / mild level 1 / moderate level 2 / severe level 3). Wear characteristic frequency Speed ​​fluctuation amplitude .

[0100] In one embodiment of this application, after step S207 above, where the wear level, wear characteristic frequency, and speed fluctuation amplitude are output as the gear wear quantification parameters, the following step B10 is further included: Step B10: When the wear level indicates severe wear, a wear fault warning signal is sent to the upper-level vehicle controller to prompt the user to perform inspection and maintenance.

[0101] For example, when the wear level is determined to be level 3, i.e., severe wear level W_grade = 3, the system automatically sends a wear fault warning signal to the upper-level vehicle controller. This warning signal is transmitted to the vehicle controller via the CAN bus, and the vehicle controller prompts the user to perform timely inspection and maintenance through the instrument display or the vehicle information system. When gear wear reaches the severe level, although the adaptive compensation system is still working normally and suppressing transmission vibration, the wear of the gears is close to the end of their service life. Continued operation may lead to serious faults such as broken gear teeth and bearing failure. Therefore, timely sending a warning signal to remind the user to have the gears inspected can effectively avoid safety hazards and reduce the risk of more serious damage to transmission system components and high repair costs caused by excessive gear wear.

[0102] In this embodiment, by proactively sending a wear fault warning signal to the upper-level vehicle controller when the wear level reaches severe wear, the user can be promptly reminded to perform inspection and maintenance, avoiding serious transmission system failures and safety hazards caused by excessive gear wear. This mechanism, while achieving full lifecycle adaptive torque compensation, provides effective early warning protection for vehicle safety, enabling users to perform maintenance and repairs at appropriate times, reducing the risk of more serious component damage and repair costs caused by excessive gear wear.

[0103] Figure 8This is a schematic diagram of a vehicle motor torque control device disclosed in an embodiment of this application; see reference. Figure 8 As shown, the vehicle motor torque control device provided in this embodiment includes: The data acquisition module 801 is used to collect motor speed, motor output torque and vehicle speed data in real time under the actual operating conditions of the vehicle and perform filtering and correction processing. Quantization module 802 is used to calculate gear wear quantization parameters based on the processed motor speed, motor output torque and vehicle speed data; The correction module 803 is used to input the obtained gear wear quantification parameters into the Model Reference Adaptive Controller (MRAC) controller to correct the reference model inside the controller. The first calculation module 804 is used to determine the desired speed according to the modified reference model and calculate the real-time speed tracking error; based on the gear wear quantification parameters and the real-time speed tracking error, the adaptive control parameters are updated online through the wear perception adaptive law to calculate the closed-loop compensation torque. The second calculation module 805 is used to calculate the total compensation torque based on the closed-loop compensation torque; The torque generation module 806 is used to superimpose the total compensation torque onto the basic drive torque issued by the vehicle to obtain a motor execution torque that adapts to the current gear wear state and outputs it for execution. The repeat execution module 807 is used to repeatedly execute all the above processes in a loop.

[0104] The process of obtaining the total compensation torque based on the closed-loop compensation torque includes: In the composite control mode of feedforward control plus MRAC control, based on the gear wear quantification parameters, the feedforward pre-compensation torque is generated by looking up a table through a preset gear meshing mechanism lookup model. The feedforward pre-compensation torque and the closed-loop compensation torque are combined to calculate the total compensation torque.

[0105] In one embodiment of this application, the acquisition module 801 is specifically used for: Real-time acquisition of three raw signals: motor speed, motor output torque, and vehicle speed data; The three original signals are subjected to first-order digital low-pass filtering to remove high-frequency sampling noise from the sensors, resulting in the processed motor speed, motor output torque, and vehicle speed data.

[0106] In one embodiment of this application, the quantization module 802 is specifically used for: For the motor speed, the time-domain wear characteristics and wear characteristic frequencies are calculated through time-domain calculation and fast Fourier transform frequency-domain calculation; The baseline residual characteristics are calculated by performing residual calculations between the motor speed and the baseline speed under healthy gear conditions. An adaptive filtering method is used to construct a road disturbance reference signal from the vehicle speed data. The low-frequency road excitation component is adaptively separated and filtered out from the corrected motor speed, while the gear meshing component is retained as the adaptive filter wear feature. The speed fluctuation amplitude is determined based on the motor speed. The time-domain wear characteristics, wear characteristic frequencies, benchmark residual characteristics, and adaptive filtered wear characteristics are normalized and weighted and fused to obtain a comprehensive wear characteristic value. Based on the comprehensive wear characteristic value, the wear level is determined by a pre-configured hysteresis threshold determination strategy; The output includes the gear wear quantification parameters, which include the wear level, the wear characteristic frequency, and the rotational speed fluctuation amplitude.

[0107] In one embodiment of this application, the quantization module 802 specifically calculates the time-domain wear characteristics and wear characteristic frequencies of the motor speed through time-domain calculation and fast Fourier transform frequency-domain calculation, including: By applying a sliding window of fixed length to the processed motor speed, the speed sequence is extracted within the sliding window. The root mean square value, kurtosis value, peak-to-peak value, variance, and impact pulse index of the speed sequence are calculated to obtain the time-domain wear characteristics. A fast Fourier transform is performed on the motor speed signal within the sliding window to obtain the speed spectrum information. The theoretical meshing fundamental frequency of the gear is calculated by combining the real-time motor speed, reducer speed ratio, and gear tooth count. The amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands are extracted to determine the wear characteristic frequency of the gear.

[0108] In one embodiment of this application, the above-mentioned correction module 803 is specifically used for: The MRAC controller has pre-stored standard transmission stiffness and standard damping values ​​corresponding to healthy gear states. Based on the wear level in the gear wear quantification parameters, the standard transmission stiffness value and the standard damping value are adjusted and corrected by a preset linear decreasing mapping relationship; wherein, the higher the wear level, the greater the adjustment, so that the corrected reference model matches the actual dynamic characteristics under the current gear wear state; wherein, the corrected reference model outputs an ideal expected speed, which is used to calculate the tracking error with the actual output motor speed.

[0109] In one embodiment of this application, the first calculation module 804 is specifically used for: Based on the wear level in the gear wear quantification parameters, a corresponding base gain coefficient is determined through a preset level gain mapping relationship; wherein, the higher the wear level, the larger the base gain coefficient. Based on the wear characteristic frequency in the gear wear quantification parameters, the gain of the corresponding frequency band centered on the wear characteristic frequency is adjusted using a preset frequency segmentation gain function to determine the frequency adjustment gain coefficient. The base gain coefficient and the frequency adjustment gain coefficient are combined to obtain the target gain coefficient; Based on the target gain coefficient, the real-time speed tracking error, and the model output of the corrected reference model, the adaptive control parameters are updated online using an adaptive update rule constructed based on Lyapunov stability theory. The update amount of the controller parameters is calculated, and a closed-loop compensation torque matching the current wear disturbance is generated.

[0110] In one embodiment of this application, the quantization module 802 determines the wear level based on the comprehensive wear characteristic value using a hysteresis threshold determination strategy, specifically for: The upgrade threshold and downgrade threshold of the wear level are pre-configured; wherein the upgrade threshold is higher than the corresponding downgrade threshold, forming a hysteresis interval; When the comprehensive wear characteristic value continuously reaches a preset number of control cycles that meet the upgrade threshold condition, the wear level will be switched to the next higher level. When the comprehensive wear characteristic value continuously reaches a preset number of control cycles that meet the downgrade threshold condition, the wear level is switched to the next lower level.

[0111] In one embodiment of this application, the quantization module 802 is further used for: When the wear level indicates severe wear, a wear fault warning signal is sent to the upper-level vehicle controller to prompt the user to perform inspection and maintenance. In another embodiment of this application, a vehicle is also provided, including: a vehicle body and electronic equipment, see reference to Figure 9 As shown, the electronic device includes a processor 1110, a communication interface 1120, a memory 1130, and a communication bus 1140, wherein the processor 1110, the communication interface 1120, and the memory 1130 communicate with each other through the communication bus 1140. Memory 1130 is used to store computer programs; Processor 1110, when executing a program stored in memory, implements the vehicle motor torque control method described above, including: Real-time data collection of motor speed, motor output torque, and vehicle speed under actual vehicle operating conditions, followed by filtering and correction processing; Based on the processed motor speed, motor output torque and vehicle speed data, gear wear quantification parameters are calculated. The obtained gear wear quantification parameters are input into the Model Reference Adaptive Controller (MRAC) to correct the reference model and adaptive control parameters inside the controller. Based on the revised reference model and the adaptive control parameters, and combined with the real-time speed tracking error, the closed-loop compensation torque is calculated. Based on the closed-loop compensation torque, the total compensation torque is obtained; The total compensation torque is added to the base drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and then output for execution. Repeat the entire process described above in a loop.

[0112] The electronic device provided in this invention uses a processor to execute a program stored in its memory. This allows for real-time acquisition of motor speed, motor output torque, and vehicle speed signals without the need for additional sensor hardware. It utilizes wear quantification parameters to continuously correct the reference model and adaptive control parameters within the MRAC controller, ensuring the controller always matches the actual dynamic characteristics of the transmission system under the current gear wear state. By repeatedly executing the above process, it achieves full lifecycle detection of the reducer gears, from healthy to heavily worn, without the need for staged calibration. It continuously compensates for transmission characteristic deviations caused by wear at the software level, significantly improving the driving smoothness of high-mileage vehicles and reducing costs.

[0113] Specifically, the program may include program code, which includes computer-executable instructions. The memory may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device. The processor may be a central processing unit (CPU), a microcontroller unit (MCU), or an application-specific integrated circuit (ASIC).

[0114] The communication bus 1140 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, it is shown in the figure with only one thick line, but this does not indicate that there is only one bus or one type of bus.

[0115] The communication interface 1120 is used for communication between the above-mentioned electronic device and other devices.

[0116] The memory 1130 may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0117] The processor 1110 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0118] In another embodiment of this application, a computer-readable storage medium is also provided, on which a program for a vehicle motor torque control method is stored. When the program for the control method is executed by a processor, it implements the steps of the vehicle motor torque control method described in any of the foregoing method embodiments.

[0119] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0120] It should be understood that the application of this application is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims. Those skilled in the art can understand that implementing all or part of the processes of the above embodiments and making equivalent changes according to the claims of this application still fall within the scope of this application.

Claims

1. A method for controlling the torque of a vehicle motor, characterized in that, The method includes: Real-time data collection of motor speed, motor output torque, and vehicle speed under actual vehicle operating conditions, followed by filtering and correction processing; Based on the processed motor speed, motor output torque and vehicle speed data, gear wear quantification parameters are calculated. The obtained gear wear quantification parameters are input into the Model Reference Adaptive Controller (MRAC) to correct the reference model inside the controller. The desired speed is determined based on the modified reference model, and the real-time speed tracking error is calculated. Based on the gear wear quantification parameters and the real-time speed tracking error, the adaptive control parameters are updated online using the wear-sensing adaptive law, and the closed-loop compensation torque is calculated. Based on the closed-loop compensation torque, the total compensation torque is obtained; The total compensation torque is added to the base drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and then output for execution. Repeat the entire process described above in a loop.

2. The method according to claim 1, characterized in that, The process of obtaining the total compensation torque based on the closed-loop compensation torque includes: In the composite control mode of feedforward control plus MRAC control, based on the gear wear quantification parameters, the feedforward pre-compensation torque is generated by looking up a table through a preset gear meshing mechanism lookup model. The feedforward pre-compensation torque and the closed-loop compensation torque are combined to calculate the total compensation torque.

3. The method according to claim 1, characterized in that, The process of collecting and filtering data on motor speed, motor output torque, and vehicle speed under actual vehicle operating conditions in real time includes: Real-time acquisition of three raw signals: motor speed, motor output torque, and vehicle speed data; The three original signals are subjected to first-order digital low-pass filtering to remove high-frequency sampling noise from the sensors, resulting in the processed motor speed, motor output torque, and vehicle speed data.

4. The method according to claim 1, characterized in that, Based on the processed motor speed, motor output torque, and vehicle speed data, the gear wear quantification parameters are calculated, including: For the motor speed, the time-domain wear characteristics and wear characteristic frequencies are calculated through time-domain calculation and fast Fourier transform frequency-domain calculation; The baseline residual characteristics are calculated by performing residual calculations between the motor speed and the baseline speed under healthy gear conditions. An adaptive filtering method is used to construct a road disturbance reference signal from the vehicle speed data. The low-frequency road excitation component is adaptively separated and filtered out from the corrected motor speed, while the gear meshing component is retained as the adaptive filter wear feature. The speed fluctuation amplitude is determined based on the motor speed. The time-domain wear characteristics, wear characteristic frequencies, benchmark residual characteristics, and adaptive filtered wear characteristics are normalized and weighted and fused to obtain a comprehensive wear characteristic value. Based on the comprehensive wear characteristic value, the wear level is determined by a pre-configured hysteresis threshold determination strategy; The output includes the gear wear quantification parameters, which include the wear level, the wear characteristic frequency, and the rotational speed fluctuation amplitude.

5. The method according to claim 4, characterized in that, The time-domain wear characteristics and wear characteristic frequencies are calculated by time-domain calculation and fast Fourier transform frequency-domain calculation of the motor speed, including: By applying a sliding window of fixed length to the processed motor speed, the speed sequence is extracted within the sliding window. The root mean square value, kurtosis value, peak-to-peak value, variance, and impact pulse index of the speed sequence are calculated to obtain the time-domain wear characteristics. A fast Fourier transform is performed on the motor speed signal within the sliding window to obtain the speed spectrum information. The theoretical meshing fundamental frequency of the gear is calculated by combining the real-time motor speed, reducer speed ratio, and gear tooth count. The amplitude changes of the meshing fundamental frequency, its harmonics, and sidebands are extracted to determine the wear characteristic frequency of the gear.

6. The method according to claim 4, characterized in that, The obtained gear wear quantification parameters are input into the Model Reference Adaptive Controller (MRAC) to correct the reference model within the controller, including: The MRAC controller has pre-stored standard transmission stiffness and standard damping values ​​corresponding to healthy gear states. Based on the wear level in the gear wear quantification parameters, the standard transmission stiffness value and the standard damping value are adjusted and corrected by a preset linear decreasing mapping relationship; wherein, the higher the wear level, the greater the adjustment, so that the corrected reference model matches the actual dynamic characteristics under the current gear wear state; wherein, the corrected reference model outputs an ideal expected speed, which is used to calculate the tracking error with the actual output motor speed.

7. The method according to claim 4, characterized in that, The process involves determining the desired rotational speed based on the modified reference model and calculating the real-time rotational speed tracking error; then, based on the gear wear quantification parameters and the real-time rotational speed tracking error, updating the adaptive control parameters online using a wear-sensing adaptive law to calculate the closed-loop compensation torque, including: Based on the wear level in the gear wear quantification parameters, a corresponding base gain coefficient is determined through a preset level gain mapping relationship; wherein, the higher the wear level, the larger the base gain coefficient. Based on the wear characteristic frequency in the gear wear quantification parameters, the gain of the corresponding frequency band centered on the wear characteristic frequency is adjusted using a preset frequency segmentation gain function to determine the frequency adjustment gain coefficient. The base gain coefficient and the frequency adjustment gain coefficient are combined to obtain the target gain coefficient; Based on the target gain coefficient, the real-time speed tracking error, and the model output of the corrected reference model, the adaptive control parameters are updated online using an adaptive update rule constructed based on Lyapunov stability theory. The update amount of the controller parameters is calculated, and a closed-loop compensation torque matching the current wear disturbance is generated.

8. The method according to claim 4, characterized in that, The determination of the wear level based on the comprehensive wear characteristic value using a hysteresis threshold determination strategy includes: The upgrade threshold and downgrade threshold of the wear level are pre-configured; wherein the upgrade threshold is higher than the corresponding downgrade threshold, forming a hysteresis interval; When the comprehensive wear characteristic value continuously reaches a preset number of control cycles that meet the upgrade threshold condition, the wear level will be switched to the next higher level. When the comprehensive wear characteristic value continuously reaches a preset number of control cycles that meet the downgrade threshold condition, the wear level is switched to the next lower level.

9. The method according to claim 4, characterized in that, After outputting the wear level, wear characteristic frequency, and speed fluctuation amplitude as the gear wear quantification parameters, the method further includes: When the wear level indicates severe wear, a wear fault warning signal is sent to the upper-level vehicle controller to prompt the user to perform inspection and maintenance.

10. A vehicle motor torque control device, characterized in that, The device includes: The data acquisition module is used to collect real-time data on motor speed, motor output torque, and vehicle speed under the actual operating conditions of the vehicle and perform filtering and correction processing. The quantization module is used to calculate gear wear quantization parameters based on the processed motor speed, motor output torque and vehicle speed data. The correction module is used to input the obtained gear wear quantification parameters into the Model Reference Adaptive Controller (MRAC) to correct the reference model inside the controller. The first calculation module is used to determine the desired speed according to the modified reference model and calculate the real-time speed tracking error; based on the gear wear quantification parameters and the real-time speed tracking error, the adaptive control parameters are updated online through the wear perception adaptive law to calculate the closed-loop compensation torque. The second calculation module is used to calculate the total compensation torque based on the closed-loop compensation torque; The execution torque generation module is used to superimpose the total compensation torque onto the basic drive torque issued by the vehicle to obtain the motor execution torque adapted to the current gear wear state and output it for execution. The repeat execution module is used to repeatedly execute all the above processes in a loop.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for vehicle motor torque control, which, when executed by a processor, implements the steps of the vehicle motor torque control described in any one of claims 1 to 9.

12. A vehicle, characterized in that, include: The vehicle body and electronic equipment, wherein the electronic equipment includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in a memory, it implements the vehicle motor torque control method according to any one of claims 1 to 9.

Images