A new energy vehicle electric steering wheel drive abnormal operation state analysis method

Abstract

This application relates to the field of electric steering technology, and in particular to a method for analyzing abnormal operating states of an electric steering system in new energy vehicles. The method includes: acquiring a vehicle driving state information set; based on the vehicle driving state information set, analyzing the dynamic response characteristics of the electric steering system under the coupling effect of kinetic energy recovery and drive assist, obtaining a steering system operating condition information set; based on the steering system operating condition information set, analyzing the dynamic evolution behavior of the steering system after experiencing the coupling effect of multiple factors such as continuous energy recovery, bus voltage fluctuations, and torque interference, obtaining a steering system abnormal operating condition information set; based on the steering system abnormal operating condition information set, performing visual dynamic presentation and safety risk warning, and outputting an electric steering system abnormal operation monitoring log. This improves the reliability of steering in new energy vehicles and contributes to the safe and efficient development of the industry.

Description

Technical Field

[0001] This application relates to the field of electric power steering technology, and in particular to a method for analyzing abnormal operating states of electric power steering drives in new energy vehicles. Background Technology

[0002] In existing technologies, the status monitoring and fault diagnosis of electric power steering systems in new energy vehicles usually focus on threshold judgment of a single signal or fault code triggering based on simple logic rules. Some methods attempt to analyze the performance of the power steering system under steady-state conditions or identify typical independent faults such as overcurrent and overheating.

[0003] However, with the deep coupling of kinetic energy recovery and electric power steering in new energy vehicles, the operating environment of the steering gear has become more complex. Especially under complex dynamic conditions such as frequent start-stop, high-speed energy recovery and instantaneous switching of drive assist, the driving behavior of the steering gear is affected by the coupling effect of multiple factors such as continuously changing energy recovery intensity, bus voltage fluctuations caused by it, and random torque interference from the road surface. Existing technical solutions often cannot effectively characterize and analyze the dynamic evolution process of the electric steering gear driving state under such multi-factor, strongly coupled, and transient conditions. This results in insufficient early perception of potential abnormal operating states, delayed warnings, and difficulty in accurately diagnosing atypical and gradual anomalies caused by complex coupling effects. Summary of the Invention

[0004] This application provides a method for analyzing abnormal operating states of electric power steering systems in new energy vehicles to solve the aforementioned problems. The method includes: acquiring a vehicle driving state information set; analyzing the dynamic response characteristics of the electric power steering system under the coupling effect of kinetic energy recovery and drive assist based on the vehicle driving state information set to obtain a power steering system operating condition information set; analyzing the dynamic evolution behavior of the power steering system after experiencing the coupling effect of multiple factors such as continuous energy recovery, bus voltage fluctuation, and torque interference based on the power steering system operating condition information set to obtain a power steering system abnormal operating condition information set; and performing visual dynamic presentation and safety risk warning based on the power steering system abnormal operating condition information set, and outputting an electric power steering system abnormal operation monitoring log.

[0005] The above technical solution can comprehensively capture abnormal states of electric steering systems under the coupling of multiple factors, achieving accurate monitoring and timely early warning, thus ensuring driving safety. Its visual presentation and monitoring logs provide data support for fault diagnosis, maintenance, and system optimization, overcoming the limitations of existing single-factor monitoring, improving the reliability of steering in new energy vehicles, and contributing to the safe and efficient development of the industry.

[0006] Optionally, the step of analyzing the dynamic response characteristics of the electric steering gear under the coupling effect of kinetic energy recovery and drive assist based on the vehicle driving state information set to obtain a steering gear operating condition information set includes: the vehicle driving state information set includes a driving intention information set, an energy recovery information set, and a steering gear state information set; based on the driving intention information set, analyzing the steering wheel angle change information and steering assist demand characteristics to obtain steering gear demand operating condition information; based on the steering gear demand operating condition information, combined with the energy recovery information set, analyzing the dynamic superposition and offsetting effects of kinetic energy recovery intensity and duration on drive assist under the influence of kinetic energy recovery to obtain a steering kinetic energy drive response information set; based on the steering kinetic energy drive response information set, combined with the steering gear state information set, analyzing the dynamic following and deviation characteristics between the steering gear motor response parameters and kinetic energy response characteristics to obtain the steering gear operating condition information set.

[0007] Optionally, the process of constructing the steering gear demand condition information includes: based on the driving intention information set, analyzing the impact of different steering wheel angle intervals on basic power assist demand to obtain a basic steering information set; based on the basic steering information set, and combined with the steering wheel angle change information, analyzing the compensation trend of steering gear power assist demand with steering changes during dynamic steering to obtain dynamic power assist compensation information; and based on the dynamic power assist compensation information, analyzing the adjustment information of steering assist magnitude and response speed under different vehicle speed conditions to obtain the steering gear demand condition information.

[0008] Optionally, the process of constructing the steering kinetic energy drive response information set includes: based on the energy recovery information set, analyzing the correspondence between the kinetic energy recovery intensity and the duration of a single recovery to obtain recovery load profile information characterizing the recovery force and duration; based on the recovery load profile information, combined with the steering assist demand characteristics, analyzing the steady-state compensation and dynamic following process of the drive assist output to overcome the recovery anti-drag torque under different recovery load profiles to obtain recovery-assistance coupling dynamic information; based on the recovery-assistance coupling dynamic information, analyzing the overall superposition enhancement or cancellation weakening relationship between kinetic energy recovery and vehicle drive assist to obtain the steering kinetic energy drive response information set characterizing the effect of both.

[0009] Optionally, the step of analyzing the dynamic following and deviation characteristics between the steering gear motor response parameters and the kinetic energy response characteristics based on the steering gear kinetic energy drive response information set and the steering gear state information set to obtain the steering gear operating condition information set includes: based on the steering gear kinetic energy drive response information set and the steering gear state information set, comparing and analyzing the dynamic following process of the electric steering gear response parameters to the kinetic energy response characteristics within a single kinetic energy recovery-assist coupling cycle to obtain periodic following characteristic information characterizing the instantaneous matching relationship between the two; based on the periodic following characteristic information, analyzing the persistence and evolution trend of the deviation between the steering gear motor response parameters and the kinetic energy response characteristics after experiencing continuous kinetic energy recovery-assist coupling to obtain steady-state offset evolution information; based on the steady-state offset evolution information and the periodic following characteristic information, analyzing the local mismatch risk and the overall performance degradation risk caused by the accumulation of steady-state offset in the dynamic following process to obtain the steering gear operating condition information set.

[0010] Optionally, based on the steering gear operating condition information set, the analysis of the dynamic evolution behavior of the steering gear after experiencing the coupling effects of continuous energy recovery, bus voltage fluctuations, and torque disturbances, to obtain a steering gear drive abnormal operating condition information set, includes: based on the steering gear operating condition information set, analyzing the cumulative effect of steering gear torque output baseline drift and response delay caused by continuous energy recovery on the steering gear drive behavior, to obtain energy recovery cumulative effect information; based on the energy recovery cumulative effect information, analyzing the torque pulsation and controllability disturbance characteristics of bus voltage fluctuations on steering gear drive stability under the background of baseline drift and the cumulative delay effect, to obtain voltage fluctuation disturbance information; based on the voltage fluctuation disturbance information, analyzing the coupling amplification effect between random road torque disturbances and the baseline drift and the torque pulsation during the electric steering and kinetic energy recovery coupling stage, to obtain multi-factor coupling disturbance information; based on the multi-factor coupling disturbance information, analyzing the overall deviation mode and abnormal characteristics of the steering gear drive behavior relative to the normal power assist demand response, to obtain the steering gear drive abnormal operating condition information set.

[0011] Optionally, the process of constructing the energy recovery cumulative effect information includes: based on the steering gear operating condition information set, analyzing the mean change characteristics of the basic compensation torque output by the steering gear motor to overcome the reverse drag torque during a continuous kinetic energy recovery cycle, to obtain torque output baseline drift information characterizing the long-term deviation of compensation demand; based on the torque output baseline drift information, analyzing the evolution law of the steering gear motor's response control command start time and the time to reach the target torque with the increase of the number of recovery load applications under the background of baseline drift, to obtain response delay cumulative evolution information; based on the response delay cumulative evolution information, analyzing the correlation growth relationship between baseline drift information and delay accumulation during the continuous energy recovery process, to obtain the energy recovery cumulative effect information.

[0012] Optionally, the process of constructing the voltage fluctuation disturbance information includes: based on the energy recovery cumulative effect information, analyzing the amplitude and frequency characteristics of voltage fluctuations during the dynamic change of bus voltage caused by kinetic energy recovery, and obtaining fluctuation profile information characterizing the voltage disturbance source characteristics; based on the fluctuation profile information, combined with the torque output baseline drift information, analyzing the law of periodic oscillation of electric steering output torque caused by voltage fluctuations at the electric steering operating point corresponding to the baseline drift, and obtaining torque pulsation generation information; based on the torque pulsation generation information, combined with the response delay cumulative evolution information, analyzing the interference characteristics of torque pulsation on the controller's precise adjustment of motor output in the context of an already existing delay in electric steering response, and obtaining the voltage fluctuation disturbance information characterizing the decrease in drive stability.

[0013] Optionally, the process of constructing the multi-factor coupled interference information includes: based on the voltage fluctuation disturbance information, analyzing the initial disturbance response characteristics when encountering random road torque interference during the electric steering assist process, to obtain random interference characteristic information; based on the random interference characteristic information, combined with the torque output baseline drift information, analyzing the interaction between the random road torque interference and the existing electric steering torque output baseline, to obtain first-level coupled amplification information; based on the first-level coupled amplification information, combined with the torque pulsation generation information, analyzing the interference after the first-level coupled amplification, and the superposition and amplification of the periodic torque pulsation caused by the bus voltage fluctuation at the force level and the fusion of disturbance characteristics in the time sequence, to obtain second-level coupled amplification information; based on the second-level coupled amplification information, analyzing the degree of stability deterioration and the attenuation characteristics of the electric steering drive behavior after the two-level coupled amplification, to obtain the multi-factor coupled interference information.

[0014] Optionally, the step of visualizing and dynamically presenting the abnormal operating condition information set of the steering gear drive, providing safety risk warnings, and outputting an abnormal operation monitoring log for the electric steering gear drive includes: analyzing the deviation distribution, duration, and rate of change between the output torque and demand torque of the electric steering gear based on the multi-factor coupled interference information to obtain abnormal torque behavior characteristic information; analyzing the evolutionary pattern of the abnormal torque behavior in a complete driving cycle, including concentrated occurrence, periodic recurrence, or gradual increase in intensity, based on the abnormal torque behavior characteristic information and the abnormal operating condition information set of the steering gear drive, to obtain abnormal operation mode information; classifying and evaluating the degree of decrease in drive stability and the potential impact on steering control corresponding to different abnormal operation modes to obtain safety risk level information; configuring dynamic visualization warning schemes matching different risk levels based on the safety risk level information, generating the safety risk warning containing timestamps, abnormal operation modes, risk levels, and key related parameters, and outputting the abnormal operation monitoring log for the electric steering gear drive. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of an application scenario provided in an embodiment of this application; Figure 2 This is a flowchart illustrating a method for analyzing abnormal operating states of an electric power steering system in a new energy vehicle, as provided in an embodiment of this application. Detailed Implementation

[0017] 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, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0019] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0020] During driving with electric power steering and regenerative braking coupled, existing technologies struggle to accurately analyze the dynamic evolution of the steering gear's driving state under complex conditions where regenerative braking and steering are deeply coupled. This results in weak early detection of abnormal operating states, delayed warnings, and difficulty in effectively diagnosing atypical faults caused by the coupling of multiple factors, thus creating potential safety hazards.

[0021] Based on this, this application provides a method for analyzing the abnormal operating status of electric steering gear in new energy vehicles. For electric steering gear, it realizes comprehensive abnormal monitoring and accurate early warning under the coupling of multiple factors, effectively ensuring driving safety. Through visualized data and monitoring logs, it provides support for fault diagnosis and system optimization, breaks through the limitations of single-dimensional monitoring, significantly improves the reliability of steering in new energy vehicles, and empowers the safe and efficient development of the industry.

[0022] Figure 1 This application provides an illustration of an application scenario. In the driving process where electric steering and kinetic energy recovery are coupled, the method provided in this application is applied to improve the reliability of steering in new energy vehicles and contribute to the safe and efficient development of the industry.

[0023] Specifically, the method provided in this application can be applied to any server. The server interacts with the vehicle computer to control the sensors, obtains the vehicle driving status information set provided by the vehicle computer to control the sensors, accurately captures the dynamic response law of the steering gear, comprehensively presents the operating status, generates a safety risk warning for the driver, and outputs an abnormal operation monitoring log of the electric steering gear to the vehicle computer, ensuring stable steering control and enhancing driving safety and comfort.

[0024] For specific implementation details, please refer to the following examples.

[0025] Figure 2 This is a flowchart illustrating a method for analyzing abnormal operating states of an electric steering gear drive in a new energy vehicle, provided as an embodiment of this application. The method of this embodiment can be applied to servers in the above scenarios. Figure 2 As shown, the method includes: S201. Obtain the vehicle driving state information set. Based on the vehicle driving state information set, analyze the dynamic response characteristics of the electric steering gear under the coupling effect of kinetic energy recovery and drive assist, and obtain the steering gear operating condition information set.

[0026] The vehicle driving status information set can be a collection of various operating parameters during the driving process of a new energy vehicle, with the data source being the onboard computer controlling sensors. The electric steering gear is a core component in a new energy vehicle that provides steering assistance, ensuring steering ease and flexibility. The coupling effect of kinetic energy recovery and steering assist can be the process where kinetic energy recovery and steering assist operations occur simultaneously and influence each other during vehicle operation. Dynamic response characteristics can be the characteristics of how the electric steering gear's assist magnitude, response speed, and other performance parameters change under external forces under the coupling effect. The steering gear operating condition information set can be a collection of information on the electric steering gear's various operating states, such as normal and critical, under different driving scenarios.

[0027] Specifically, in the current context of the rapid development of the new energy vehicle industry, existing methods are unable to analyze the complex dynamic response during the coupling and switching of kinetic energy recovery and drive assist in new energy vehicles. This leads to the masking of the true load characteristics of the steering gear, the inability to effectively detect potential performance mismatches and early anomalies, and the inability to provide an accurate operating condition baseline for subsequent in-depth analysis. This increases the risk of safety hazards caused by steering response lag or abrupt changes.

[0028] S202. Based on the steering gear operating condition information set, analyze the dynamic evolution behavior of the steering gear after experiencing the coupling effect of multiple factors such as continuous energy recovery, bus voltage fluctuation and torque interference, and obtain the steering gear drive abnormal operating condition information set.

[0029] Continuous energy recovery refers to the operating state of a vehicle continuously recovering kinetic energy over a relatively long period. Bus voltage fluctuation refers to the phenomenon where the high-voltage bus voltage of a new energy vehicle deviates from its rated value. Torque interference refers to the interference effect of factors such as road surface and mechanical wear on the output torque of the electric steering system. Multi-factor coupling refers to the process where continuous energy recovery, bus voltage fluctuation, and torque interference coexist and influence each other. Dynamic evolution behavior refers to the regular characteristics of the electric steering system's operating state changing over time under the multi-factor coupling effect. Steering system drive abnormal condition information set refers to a collection of information reflecting the type and timing of electric steering system drive abnormalities.

[0030] Specifically, during the coupling and switching process of kinetic energy recovery and drive assist in new energy vehicles, multiple factors such as continuous energy recovery, voltage fluctuations and torque interference are intertwined for a long time. Existing diagnostic methods lack the ability to track the gradual evolution of system performance and cannot distinguish between normal fluctuations and abnormal trends. As a result, the implicit performance degradation and soft faults of the steering gear cannot be predicted in advance, which may lead to missing the best maintenance time and even causing functional degradation or failure.

[0031] S203. Based on the abnormal operating condition information set of the steering gear drive, perform visual dynamic presentation and safety risk warning, and output the abnormal operation monitoring log of the electric steering gear drive.

[0032] Visualized dynamic presentation can be a way to display abnormal steering gear data in an intuitive form using data visualization technology. Safety risk warnings can be functions that issue risk alerts based on abnormal data and preset thresholds through methods such as sound and light, and information push notifications. Electric steering gear drive abnormal operation monitoring logs can be documents that record abnormal steering gear monitoring data and analysis results, providing support for troubleshooting.

[0033] Specifically, during the coupling and switching of kinetic energy recovery and drive assistance in new energy vehicles, massive amounts of abstract abnormal feature data are generated. Without intuitive presentation and an immediate early warning mechanism, the analysis results are difficult to understand and apply quickly, leading to delayed maintenance decisions, untimely risk handling, and the inability to trace and reuse abnormal information due to the lack of standardized records. This makes it impossible to realize the value of the precise analysis in the early stage and to form a complete management loop.

[0034] The method provided in this embodiment can comprehensively capture abnormal states of electric steering systems under the coupling of multiple factors, achieve accurate monitoring and timely early warning, ensure driving safety, and provide data support for fault diagnosis, maintenance and system optimization through its visualization and monitoring logs. It solves the limitations of existing single-factor monitoring, improves the reliability of steering in new energy vehicles, and helps the industry develop safely and efficiently.

[0035] In some embodiments, the vehicle driving state information set includes a driving intention information set, an energy recovery information set, and a steering gear state information set. Based on the driving intention information set, the steering wheel angle change information and steering assist demand characteristics are analyzed to obtain steering gear demand condition information. Based on the steering gear demand condition information, combined with the energy recovery information set, the dynamic superposition and offset effects of kinetic energy recovery intensity and duration on drive assist under the influence of kinetic energy recovery are analyzed to obtain a steering kinetic energy drive response information set. Based on the steering kinetic energy drive response information set, combined with the steering gear state information set, the dynamic following and deviation characteristics between the steering gear motor response parameters and kinetic energy response characteristics are analyzed to obtain a steering gear operating condition information set.

[0036] The driving intent information set can be the driver's steering operation requirements. The energy recovery information set can be the vehicle's kinetic energy recovery process status. The steering gear status information set can be the electric steering gear's own operating status. The steering wheel angle change information can be the amount and trend of steering wheel angle change per unit time. The steering assist demand characteristics can be to clarify the requirements for the magnitude and compensation range of steering assist under different operating conditions, guiding the assist output. The steering gear demand operating condition information can be a dataset that comprehensively considers driving intent and driving conditions to determine the steering assist supply standard. The kinetic energy recovery intensity can be an indicator of the vehicle's kinetic energy recovery strength, affecting the steering assist coupling effect. The duration can refer to the duration of a single kinetic energy recovery. The drive assist can be the auxiliary torque output by the steering gear. The dynamic superposition and cancellation effect can be the mutual enhancement or weakening effect between the kinetic energy recovery counter-drag torque and the drive assist. The steering kinetic energy drive response information set can record the characteristics of steering assist adjustment and response delay under coupling action, reflecting the coupling response state. The steering gear motor response parameters can be feedback data such as the motor's output torque and response time after receiving commands. Kinetic energy response characteristics can be defined as the patterns of anti-drag torque changes and peak timing during kinetic energy recovery, which are matched with the motor response parameters. Dynamic following performance can be defined as the degree of real-time matching between the steering gear motor response parameters and the kinetic energy response characteristics. Deviation characteristics can be defined as the degree of deviation and variation patterns between the motor response parameters and the kinetic energy response characteristics.

[0037] Specifically, in the operation of new energy vehicles, kinetic energy recovery and power steering are continuously coupled. Without analyzing dynamic response characteristics, the superposition / cancellation effect of these two processes cannot be captured, making it difficult to detect mismatches between motor response and kinetic energy characteristics. This leads to distorted judgment of steering gear operating conditions, overlooking potential faults and causing problems such as steering lag and insufficient power steering, seriously threatening driving safety in scenarios like sharp turns and high-speed lane changes. To address these issues, the approach begins with the synchronous acquisition and preprocessing of multi-source data: real-time acquisition of steering wheel angle sensor signals via the CAN bus, followed by real-time differential signal processing to extract steering wheel angle change information such as "angle change rate reaches 30 degrees per second." Simultaneously, combined with vehicle speed and a preset power steering map, specific steering power demand characteristics (e.g., demand torque of 5 Nm) are analyzed. Simultaneously, energy recovery information is obtained from the vehicle controller, including specific recovery intensity (e.g., "-25% of maximum braking torque") and single-time duration (e.g., "2 seconds"). Subsequently, a coupling analysis model based on system dynamics principles is used as a key technical means to combine the analyzed demand characteristics with the recovery parameters. As input, the simulation calculates the vector superposition relationship between the two on the drive shaft, and specifically quantifies and analyzes the dynamic superposition and cancellation effects. For example, the model will output a coupling response curve such as "under the current recovery intensity, in order to achieve a net assist effect of 5 Nm, the motor actually needs to output about 6.2 Nm to compensate for the cancellation part," forming a set of steering motor energy drive response information. Finally, using the time-domain comparative analysis method, the actual steering motor response parameters obtained from the steering controller are aligned with the theoretical response curve with high precision and the difference is calculated, thereby accurately extracting dynamic tracking and deviation characteristic indicators such as "actual response delay of about 15 milliseconds" and "mean steady-state tracking error of 0.3 Nm," completing a deep characterization of the system behavior characteristics under coupled conditions.

[0038] The method provided in this embodiment accurately captures the dynamic response pattern of the steering gear and fully presents the operating status. This not only solves the defect of existing methods that ignore system coupling and improves the accuracy of operating condition judgment, but also identifies mismatch risks in advance, laying a solid foundation for subsequent abnormal warnings, ensuring stable steering and control, and enhancing driving safety and comfort.

[0039] In some embodiments, based on the driving intention information set, the influence of different steering wheel angle ranges on the basic steering assist demand is analyzed to obtain the basic steering information set; based on the basic steering information set, combined with the steering wheel angle change information, the compensation trend of steering assist demand with steering changes during dynamic steering is analyzed to obtain dynamic assist compensation information; based on the dynamic assist compensation information, the adjustment information of steering assist magnitude and response speed under different vehicle speed conditions is analyzed to obtain steering demand condition information.

[0040] Different steering wheel angle ranges can be categorized based on the numerical range of steering wheel angles, used to distinguish different steering operation amplitudes. Basic power steering assist requirement refers to the amount of steering assist required to meet the driver's basic steering operations without the influence of dynamic compensation factors. Basic steering information set is a data set showing the matching relationship between different steering wheel angle ranges and their corresponding basic power steering assist requirements. Steering wheel angle change information refers to the rate and trend of steering wheel angle change over time. Dynamic steering process refers to the non-uniform, non-fixed-amplitude steering operation performed by the driver during driving. Steering assist requirement refers to the total amount of assistance provided by the steering gear to assist the driver in completing steering operations. Dynamic power steering assist compensation trend refers to the variation of steering assist compensation amount with factors such as steering angle and vehicle speed during dynamic steering. Dynamic power steering assist compensation information is a data set showing the specific values ​​and variation patterns of power steering assist compensation amount during dynamic steering. Different vehicle speed conditions refer to different speed ranges during vehicle operation. Steering assist magnitude refers to the actual power steering assist output value. Steering assist response speed refers to the time it takes for the steering gear to output the target assist after receiving a steering command. The adjustment information can be specific parameters that adapt and optimize the steering assist and response speed for different vehicle speed conditions.

[0041] Specifically, during the dynamic steering process of a vehicle, if the differences in steering wheel angle range, the rate of change of angle, and the influence of vehicle speed are ignored, and only a fixed power assist output is used, it will lead to excessive power assist at small angles and insufficient power assist at large angles. The response will be lagging during dynamic steering, with heavy steering at low speeds and sensitive steering at high speeds. This will cause inconsistent steering control, driver fatigue, and even increase the risk of collision. Furthermore, it will not be able to provide an accurate benchmark for subsequent kinetic energy recovery and drive assist coupling analysis, affecting the accuracy of abnormal operating condition identification. To address the aforementioned issues: First, a segmented static calibration method is employed. Based on extensive real-vehicle test data, the steering wheel angle is divided into multiple logical intervals with different assist characteristics. For example, the angle with an absolute value between 0° and 15° is defined as the "fine-tuning interval," associated with a smaller base assist gain to ensure stability and road feel during straight-line driving; 15° to 30° is defined as the "normal steering interval," associated with a medium base assist gain; and greater than 30° is defined as the "emergency / large angle interval," associated with the maximum base assist gain, thus constructing a basic steering information set. Next, a dynamic trend analysis method is introduced. By calculating the rate of change of the steering wheel angle (e.g., degrees / second), the base assist is compensated in real time: if the system detects a rapid increase in the angle (e.g., a rate of change exceeding 200° / second), a positive additional compensation torque is immediately superimposed on the base assist based on a preset dynamic compensation coefficient to counteract the inertia of the steering system and ensure responsiveness; conversely, during rapid return to normal steering, a positive additional compensation torque is applied. When the time is right, a negative compensation torque is superimposed to assist in returning to center, thereby generating dynamic power assist compensation information. Finally, a multi-condition adaptive fusion method is applied to fuse the above information with the real-time vehicle speed (such as from wheel speed sensors) for decision-making: the control system has a pre-set MAP map of power assist characteristics for different vehicle speed ranges (e.g., 0-20km / h low speed range, 20-80km / h medium speed range, >80km / h high speed range). Based on the current vehicle speed, the table is consulted to make a final adjustment to the "original demand" that integrates the basic and dynamic compensation. For example, at a low speed of 10km / h, a "high gain, fast response" mode may be adopted to amplify the calculated demand torque by a certain proportion and shorten the controller response cycle; while at a high speed of 100km / h, the "low gain, high damping" mode is switched to attenuate the demand torque and introduce a filtering algorithm to make the response smoother, thereby finally outputting "steering gear demand condition information" that is accurately adapted to the current instantaneous driving scenario and includes the target torque and response speed requirements.

[0042] The method provided in this embodiment accurately captures the dynamic changes in power steering demand under different driving scenarios. Through hierarchical analysis and trend fitting, comprehensive steering gear demand condition information is constructed, providing a reliable benchmark for subsequent coupling effect analysis and motor response matching judgment. This effectively avoids power steering mismatch problems, improves the smoothness and stability of steering control, reduces driving risks, and ensures driving safety. At the same time, it lays a solid data foundation for monitoring and early warning of abnormal states of electric steering gear.

[0043] In some embodiments, based on the energy recovery information set, the corresponding information between the kinetic energy recovery intensity and the duration of a single recovery is analyzed to obtain recovery load profile information characterizing the recovery force and duration. Based on the recovery load profile information, combined with the steering assist demand characteristics, the steady-state compensation and dynamic following process of the drive assist output to overcome the recovery reverse drag torque under different recovery load profiles is analyzed to obtain recovery-assistance coupling dynamic information. Based on the recovery-assistance coupling dynamic information, the superposition enhancement or cancellation weakening relationship between kinetic energy recovery and vehicle drive assist is analyzed to obtain a directional kinetic energy drive response information set characterizing the effect of both.

[0044] The regenerative load profile information can be the characteristic information of the relationship between the magnitude of the regenerative force and the duration of its action during the kinetic energy recovery process. The regenerative-assist coupling dynamic information can be the information of the steady-state compensation behavior and dynamic following process of the drive assist output to overcome the regenerative anti-drag torque under different regenerative loads.

[0045] Specifically, during the kinetic energy recovery process of new energy vehicles, such as braking and coasting, the recovered reverse drag torque will couple with the steering assist. If the dynamic relationship between the two is not analyzed, the power assist output is prone to deviate from the demand under different recovery intensities and durations, causing steering lag and insufficient power assist. This will reduce handling stability at high speeds or in complex road conditions, and may even lead to loss of steering control, directly threatening driving safety. Therefore, this step is the key to avoiding risks. To address the aforementioned issues, the approach begins with deep analysis and feature engineering of the raw data: Signals from the drive motor controller and battery management system are acquired in real-time via an in-vehicle network (such as a CAN bus), including drive motor torque commands (negative values ​​indicate generator operation) and DC bus current. Sliding window filtering and integral calculations are used to extract the kinetic energy recovery intensity (e.g., a moderate recovery corresponds to a generator torque of -100 Nm) and the duration of a single recovery cycle (e.g., a complete braking process lasts 5 seconds). This constructs a recovery load profile based on intensity and time. Subsequently, this profile information, along with synchronously acquired steering assist demand characteristics (e.g., the target assist torque calculated based on steering wheel angle and vehicle speed), is input into a coupled analysis model based on motor dynamics and load disturbance observation theory. This model dynamically analyzes, through simulation calculations or fitting of measured data, the amount of torque required by the assist motor to counteract the time-varying recovery drag torque under the action of the profile. The output compensation torque curve is specifically manifested as a composite control strategy that includes feedforward compensation (steady-state compensation based on contour prediction) and closed-loop feedback regulation (such as a PI controller dynamically following the actual steering angle error). This generates dynamic information of the recovery-assist coupling that finely depicts the instantaneous interaction state between the two. Finally, using time-domain and frequency-domain correlation analysis techniques (such as calculating the cross-correlation coefficient or covariance between the actual output torque of the assist motor and the ideal required torque over multiple coupling cycles), the dynamic information is statistically analyzed and pattern recognized to determine whether the two exhibit a relationship of superposition enhancement (such as the recovery torque and assist torque being accidentally aligned in direction, resulting in unexpectedly lighter steering) or cancellation and weakening (such as the recovery torque severely canceling the assist, resulting in heavy steering) within a specific driving segment. This relationship is quantified into a series of characteristic parameters (such as coupling efficiency coefficient and average deviation), and finally encapsulated into a structured set of directional motor energy drive response information for use by the upper-level diagnostic and decision-making system.

[0046] The method provided in this embodiment provides data support for steering gear control, enabling the power assist output to dynamically adapt to kinetic energy recovery conditions, reducing steering deviation and lag, improving steering smoothness and handling precision, avoiding safety hazards caused by coupling effects from the source, optimizing steering gear operating efficiency, reducing unnecessary energy consumption, and ensuring steering safety and stability in different driving scenarios.

[0047] In some embodiments, based on the steering kinetic energy drive response information set and combined with the steering gear state information set, the dynamic following process of the electric steering gear response parameters to the kinetic energy response characteristics within a single kinetic energy recovery-assist coupling cycle is compared and analyzed to obtain periodic following characteristic information characterizing the instantaneous matching relationship between the two; based on the periodic following characteristic information, the persistence and evolution trend of the deviation between the steering gear motor response parameters and the kinetic energy response characteristics after experiencing continuous kinetic energy recovery-assist coupling are analyzed to obtain steady-state offset evolution information; based on the steady-state offset evolution information and combined with the periodic following characteristic information, the local mismatch risk and the overall performance degradation risk caused by the accumulation of steady-state offset during the dynamic following process are analyzed to obtain the steering gear operating condition information set.

[0048] Kinetic energy response characteristics can be defined as the kinetic energy change pattern of a vehicle under the coupling effect of kinetic energy recovery and drive assist. Dynamic following performance can be defined as the degree of matching between the steering motor response parameters and the real-time changes in kinetic energy response characteristics. Deviation characteristics can be defined as the numerical differences and changing patterns between the steering motor response parameters and kinetic energy response characteristics. The kinetic energy recovery-assist coupling cycle can be defined as the time interval of a complete kinetic energy recovery and drive assist interaction. Periodic following characteristic information can be defined as the instantaneous matching relationship between the motor response parameters and kinetic energy response characteristics within a single kinetic energy recovery-assist coupling cycle. Steady-state offset evolution information can be defined as the continuous changing trend of the deviation between the motor response parameters and kinetic energy response characteristics after multiple consecutive kinetic energy recovery-assist coupling cycles. Local mismatch risk can be defined as the potential risk of instantaneous mismatch between the motor response parameters and kinetic energy response characteristics during dynamic following. Overall performance degradation risk can be defined as the potential risk of a decline in the overall operating performance of the steering system caused by the accumulation of steady-state offset.

[0049] Specifically, in urban roads where new energy vehicles frequently brake and turn, the electric steering system continuously experiences the coupling effect of kinetic energy recovery and power assist. If the dynamic following and deviation characteristics of the motor response parameters and kinetic energy response features are not analyzed, it will lead to instantaneous fluctuations in steering assist within a single coupling cycle, affecting handling smoothness. Furthermore, the accumulation of deviations after continuous coupling will cause a decrease in steering assist accuracy and response lag, which may induce drive abnormalities in the long term. To address the above issues: First, high-precision time series alignment technology is used to synchronize the "steering motor response parameters" (such as actual motor current or calculated real-time output torque curve) obtained from the CAN bus in real time with the "kinetic energy response features" sent from the upper-level algorithm at the millisecond level. Then, dynamic time warping and error analysis are used to compare the two curves within a single clear "kinetic energy recovery-power assist" coupling event window (such as a strong energy recovery process lasting several seconds) and calculate key dynamic indicators: for example, by finding the time difference between the rising edge of the target curve and the rising edge of the actual curve, the "response delay time" is obtained; by calculating the root mean square error between the two curves, the "following error" is obtained. These indicators collectively constitute "cycle-following characteristic information." Then, statistical process control and trend fitting methods are introduced to process the "following error" sequence analyzed over dozens or even hundreds of consecutive coupled cycles online. For example, the moving average is calculated to observe the "baseline drift" trend (e.g., the average slowly increases from the initial 5% to 8%), the standard deviation is analyzed to determine whether the fluctuation is expanding, and linear regression is used to determine whether the offset has a statistically significant growth trend, thereby generating "steady-state offset evolution information." Finally, a rule engine or lightweight machine learning classifier with dual thresholds (e.g., a local mismatch alarm is triggered when the instantaneous delay exceeds 100 milliseconds, and a performance degradation warning is triggered when the moving average trend slope is continuously positive and exceeds a certain threshold) is used to fuse and judge the above two types of information, outputting a comprehensive "steering gear operating condition information set," which clearly indicates whether the current state is normal following, occasional transient mismatch, or has entered a performance degradation channel.

[0050] The method provided in this embodiment accurately captures the instantaneous matching state of a single coupling cycle and the deviation evolution trend of continuous cycles, which can promptly identify local mismatch risks and overall performance degradation risks, providing comprehensive data support for subsequent abnormal operating condition analysis. It effectively makes up for the shortcomings of single parameter monitoring in existing technologies, improves the accuracy and foresight of steering gear operating status monitoring, helps to avoid drive anomalies in advance, and ensures driving safety and handling stability.

[0051] In some embodiments, based on the steering gear operating condition information set, the cumulative effects of baseline drift and response delay in steering gear torque output caused by continuous energy recovery on steering gear driving behavior are analyzed to obtain energy recovery cumulative effect information; based on the energy recovery cumulative effect information, the torque pulsation and controllability disturbance characteristics of bus voltage fluctuation on steering gear driving stability under the background of baseline drift and delay cumulative effect are analyzed to obtain voltage fluctuation disturbance information; based on the voltage fluctuation disturbance information, the coupling amplification effect between random road torque interference and baseline drift and torque pulsation is analyzed in the electric steering and kinetic energy recovery coupling stage to obtain multi-factor coupling disturbance information; based on the multi-factor coupling disturbance information, the overall deviation mode and abnormal characteristics of steering gear driving behavior relative to normal power assist demand response are analyzed to obtain steering gear driving abnormal operating condition information set.

[0052] Continuous energy recovery can refer to the repeated activation of the kinetic energy recovery function during vehicle operation. Bus voltage fluctuation can refer to the phenomenon where the vehicle's high-voltage bus voltage deviates from the standard operating voltage. Torque interference can refer to random road torque interference caused by encountering bumpy roads or obstacles during vehicle operation. Multi-factor coupling can refer to the fact that continuous energy recovery, bus voltage fluctuation, and torque interference are not independent effects. Dynamic evolution behavior can refer to the process of key performance indicators such as torque output and response speed of the electric steering system gradually changing over time under the multi-factor coupling effect. Energy recovery cumulative effect information can refer to the information related to the continuous accumulation of steering torque output baseline shift and response delay caused by continuous energy recovery. Voltage fluctuation disturbance information can refer to the information related to the interference of bus voltage fluctuation on the driving stability of the steering system under the background of baseline drift and delay accumulation. Multi-factor coupled interference information can refer to the comprehensive interference information formed by the superposition and amplification of the cumulative effect of continuous energy recovery, the disturbance of bus voltage fluctuation, and random torque interference. Torque output baseline drift can refer to the phenomenon that the average value of the basic compensation torque output by the steering system to overcome the recovered reverse drag torque gradually deviates from the initial set value. The cumulative effect of response delay can be the cumulative phenomenon where the start time of the steering motor in response to control commands and the time to reach the target torque gradually increase with the number of retraction load applications. Torque pulsation can be the periodic oscillation of the steering output torque caused by bus voltage fluctuations. Controllability disturbance characteristics can be the interference characteristics where torque pulsation makes it difficult for the steering controller to accurately adjust the motor output. The coupling amplification effect can be the effect of the interaction between random road torque interference, baseline drift, and torque pulsation, which increases the intensity of the interference and expands its range of influence.

[0053] Specifically, in the operation of new energy vehicles, continuous energy recovery, bus voltage fluctuations, and random road surface torque interference often occur simultaneously in scenarios such as long downhill slopes. The coupling of these three factors can lead to steering torque baseline drift, accumulated response delays, and torque pulsation and interference amplification, resulting in heavy steering, sluggish handling, or even momentary inaccuracy, seriously threatening driving safety. Existing single-factor analysis cannot capture the coupled hazards. This step is crucial for risk avoidance. To address the above issues: First, by using moving average filtering and trend fitting techniques, the torque output baseline drift caused by the reverse drag effect of continuous kinetic energy recovery is quantified and extracted from historical torque data (e.g., The cumulative evolution of the basic compensation torque (increasing unidirectionally from 5.0 N·m to 7.8 N·m) and response delay (e.g., command response time gradually increasing from 50 ms to 95 ms) was used to construct information on the cumulative effect of energy recovery. Subsequently, for the bus voltage disturbance accompanying the kinetic energy recovery process, Fast Fourier Transform (FFT) was used for frequency domain analysis to extract its fluctuation profile (e.g., identifying a periodic fluctuation component with an amplitude of approximately 1.5 volts and a main frequency of 12 Hz). Based on the motor mathematical model, the analysis was conducted on how to trigger the motor by changing the inverter output voltage under this specific fluctuation profile and the drifted operating point. The periodic oscillation of the output torque (i.e., torque pulsation, for example, generating a periodic disturbance of ±1.2 N·m) forms voltage fluctuation disturbance information. Then, when a random road torque disturbance (e.g., an impact signal with an amplitude of 4 N·m and a duration of 150 ms) is captured in real time by a torque sensor, a pre-built coupling analysis model is used to first analyze the vector superposition effect (first-stage coupling amplification) between this instantaneous disturbance and the existing torque output baseline (e.g., 7.8 N·m in the previous example). Then, it is analyzed how this initially amplified disturbance signal interacts with the ongoing periodic torque pulsation (±1.2 N·m, 1... The 2 Hz interference generates a beat frequency effect in the time domain and sideband modulation in the frequency domain, resulting in the accumulation of interference energy in a specific frequency band and the complexity of the disturbance waveform (second-stage coupling amplification). Finally, multi-factor coupled interference information is synthesized. Finally, a machine learning-based pattern recognition algorithm (such as support vector machine or cluster analysis) is applied to perform deep feature comparison and deviation pattern learning on the actual torque curve and the ideal demand curve after the above coupling evolution. Characteristic abnormal feature sets such as "abnormal energy concentration in a specific high-frequency band" and "phase lag continuously exceeding the threshold" are extracted, thereby structurally outputting a set of abnormal working condition information of the steering gear drive.

[0054] The method provided in this embodiment accurately captures the abnormal evolution patterns caused by the interaction of the three factors, completely solving the blind spot problem of single analysis. It can identify potential driving anomalies in advance, provide comprehensive data support for safety warnings, effectively avoid the risk of steering loss of control, significantly improve the driving stability and driving safety of electric steering, and requires no additional hardware. It is compatible with various new energy vehicles and has strong practicality.

[0055] In some embodiments, based on the steering gear operating condition information set, the average variation characteristics of the basic compensation torque output by the steering gear motor to overcome the reverse drag torque are analyzed within a continuous kinetic energy recovery cycle, obtaining torque output baseline drift information that characterizes the long-term deviation of compensation demand; based on the torque output baseline drift information, the evolution law of the steering gear motor's response control command start time and the time to reach the target torque with the increase of the number of recovery load applications is analyzed under the background of baseline drift, obtaining response delay cumulative evolution information; based on the response delay cumulative evolution information, the correlation growth relationship between baseline drift information and delay accumulation in the continuous energy recovery process is analyzed, obtaining energy recovery cumulative effect information.

[0056] A continuous kinetic energy recovery cycle can be multiple consecutive kinetic energy recovery processes during vehicle operation, with each cycle corresponding to a complete "recovery load application - steering gear compensation response" process. The anti-drag torque can be the reverse resistance torque exerted on the steering gear by the vehicle's braking system and energy recovery device during kinetic energy recovery. The base compensation torque can be the reference torque output by the steering gear motor to counteract the anti-drag torque. The mean variation characteristic can be the evolution of the statistical average value of the base compensation torque with the number of recovery cycles within a continuous kinetic energy recovery cycle. The torque output baseline drift information can be the characteristic information of the steering gear base compensation torque deviating from its initial reference value over a long period. The cumulative evolution information of response delay can be the evolution of the steering gear motor's response time to control commands and the time to reach the target torque, gradually increasing with the number of recovery load applications. The baseline drift information can be the core data of the steering gear torque output reference offset. The cumulative delay amount can be the cumulative increase in response delay time with the number of recovery load applications. A continuous energy recovery process can be the process of the kinetic energy recovery system continuously operating under specific driving scenarios. The correlation growth relationship can be a correlation between the synchronous increase in baseline drift magnitude and cumulative delay.

[0057] Specifically, during continuous kinetic energy recovery processes such as long downhill slopes and frequent deceleration, the steering motor needs to repeatedly output basic compensation torque to counteract the drag torque. If this process is ignored, the torque output baseline will gradually drift and the response delay will accumulate. When the two aggravate each other, the steering gear will be more prone to instability when subjected to voltage fluctuations or road interference, resulting in steering lag and abnormal torque output, which directly threatens driving safety. Therefore, it is necessary to accurately capture this cumulative effect. To address the aforementioned issues: First, based on the steering gear operating condition information set, a sliding window mean analysis method is employed to process the torque time-series data covering multiple consecutive kinetic energy recovery events (such as continuous recovery actions reaching dozens of times). A time window containing a specific number of recovery cycles is set, and the arithmetic mean of the basic compensation torque output by the steering gear motor to overcome the reverse drag torque is calculated within the window. Then, this window is slid along the time axis, and a series of window means are continuously calculated, thereby plotting the curve of the change in the compensation torque mean as the event sequence progresses. This quantifies and generates torque output baseline drift information that characterizes the long-term, unidirectional shift in compensation demand. Next, for the same event sequence, time-series analysis methods are used to accurately... By tracking the start time of the steering motor from receiving the torque control command to the actual output torque starting to rise, and the establishment time of the torque value reaching the command target value within each recovery-assist coupling cycle, the evolution trajectory of these two types of time delay parameters with the increase of recovery cycles is statistically analyzed to obtain the cumulative evolution information of response delay. Finally, using mathematical modeling methods such as Pearson correlation analysis or regression analysis, the baseline drift sequence and the cumulative delay sequence obtained above are correlated to calculate the correlation coefficient between the two or fit their functional relationship, thereby establishing a quantitative model describing how "baseline drift" and "response delay" increase and evolve synergistically during continuous energy recovery, and comprehensively generating information on the cumulative effect of energy recovery.

[0058] The method provided in this embodiment accurately quantifies the cumulative law and correlation between torque output baseline drift and response delay, providing core basic data for subsequent analysis of voltage fluctuation disturbances and multi-factor coupling interference. It effectively avoids abnormal misjudgments caused by basic state deviation, locks in the initial causes of steering gear drive stability degradation in advance, lays a solid foundation for subsequent abnormal warning and fault diagnosis, and ensures the long-term reliable operation of the steering system.

[0059] In some embodiments, based on the cumulative effect information of energy recovery, the amplitude and frequency characteristics of voltage fluctuations during the dynamic change of bus voltage caused by kinetic energy recovery are analyzed to obtain fluctuation profile information characterizing the voltage disturbance source characteristics. Based on the fluctuation profile information, combined with torque output baseline drift information, the law of periodic oscillation of electric steering output torque caused by voltage fluctuations at the electric steering operating point corresponding to the baseline drift is analyzed to obtain torque pulsation generation information. Based on the torque pulsation generation information, combined with response delay cumulative evolution information, the interference characteristics of torque pulsation on the controller's precise adjustment of motor output are analyzed in the context of an already existing delay in electric steering response to obtain voltage fluctuation disturbance information characterizing the decrease in drive stability.

[0060] Dynamic changes in bus voltage can refer to the real-time fluctuations in the voltage of the power supply bus as the regenerative load changes during vehicle kinetic energy recovery. The amplitude and frequency characteristics of voltage fluctuations can be defined as follows: amplitude refers to the maximum deviation of the bus voltage from its rated value, and frequency refers to the periodic repetition rate of the voltage fluctuation. Fluctuation profile information can be a set of information that comprehensively characterizes the voltage disturbance source by integrating characteristics such as voltage fluctuation amplitude, frequency, and duration. The electric steering operating point can be the combination of operating parameters when the steering motor stably outputs the basic compensation torque after baseline drift occurs. Torque pulsation generation information can be the regularity information of the periodic oscillation of the steering output torque caused by voltage fluctuations.

[0061] Specifically, during the driving process where electric steering and kinetic energy recovery are coupled, if multi-factor coupling interference is not analyzed, random road torque interference will be amplified by superimposed with existing torque output baseline drift and torque pulsation caused by voltage fluctuations. This will lead to a sharp deterioration in steering stability and a decrease in recovery capability, resulting in steering lag, handling deviations, and other problems. In severe cases, it can cause driving safety hazards and make it impossible to accurately identify the root cause of the drive abnormality. To address the above problems: First, signal processing technology is used to perform time-frequency domain analysis on the real-time acquired bus voltage signal (e.g., using Fast Fourier Transform or Wavelet Transform to extract energy in specific frequency bands) to quantify the amplitude of voltage fluctuations (e.g., the peak-to-peak value of the fluctuation reaches a certain value) and the dominant frequency characteristics (e.g., the main fluctuations are concentrated in a certain frequency range), thereby accurately characterizing the "fluctuation profile information" as the source of disturbance. Then, combined with the "torque output baseline drift information" characterizing the steering gear's deviation from the original working area due to long-term compensation for energy recovery reverse drag torque, the small-signal disturbance modeling method of the motor is used to analyze the unit change in voltage fluctuation (e.g., voltage drop of a certain value) at the steady-state operating point after the offset (e.g., corresponding to a specific motor current bias). Numerical values ​​influence the motor terminal voltage and electromagnetic torque constant, causing periodic oscillations in the output torque, including the magnitude and frequency of the oscillation. This is the construction of "torque pulsation generation information." Finally, "response delay cumulative evolution information," reflecting the slowed system response speed, is integrated. Through control loop interference coupling analysis, the interference details caused by the aforementioned periodic torque pulsations on the controller's (such as the PID current loop) adjustment process are specifically evaluated. For example, if the torque pulsation frequency approaches or exceeds the reduced bandwidth when the control system bandwidth has been reduced due to delay, the controller will be unable to effectively suppress it, resulting in a peak in the torque tracking error spectrum at the corresponding frequency. This allows for the quantitative determination of "voltage fluctuation disturbance information," which characterizes the decrease in drive stability.

[0062] The method provided in this embodiment accurately captures the interaction mechanism between random road surface torque disturbance and baseline drift and torque pulsation, and fully reconstructs the interference process of multi-factor coupling on steering gear drive. This not only provides key data support for subsequent abnormal working condition identification, but also predicts the trend of stability deterioration in advance, improves the comprehensiveness and accuracy of abnormal monitoring, lays a solid foundation for safety risk warning, and ensures steering control safety.

[0063] In some embodiments, based on voltage fluctuation disturbance information, the initial disturbance response characteristics when encountering random road surface torque interference during the electric steering assist process are analyzed to obtain random disturbance characteristic information; based on the random disturbance characteristic information, combined with torque output baseline drift information, the interaction between random road surface torque interference and the existing electric steering torque output baseline is analyzed to obtain first-level coupling amplification information; based on the first-level coupling amplification information, combined with torque pulsation generation information, the interference after the first-level coupling amplification is analyzed, and the periodic torque pulsation caused by bus voltage fluctuation is superimposed and amplified at the force level and the disturbance characteristics are fused in the time sequence to obtain second-level coupling amplification information; based on the second-level coupling amplification information, the degree of stability deterioration and the attenuation characteristics of the electric steering drive behavior after the two-level coupling amplification are analyzed to obtain multi-factor coupling disturbance information.

[0064] Random road torque disturbances can be instantaneous torque disturbances caused by random factors such as uneven road surfaces or running over obstacles during electric steering. Initial disturbance response characteristics can be the initial characteristics of the electric steering system, such as torque response and speed changes, exhibited within a short period when encountering random road torque disturbances. Random disturbance characteristic information can include key attributes such as the intensity, duration, and rate of change of the random road torque disturbance. First-stage coupling amplification information can be the relevant characteristic information regarding the increased disturbance intensity caused by the interaction between the random road torque disturbance and the existing torque output baseline drift of the steering system. Second-stage coupling amplification information can be the relevant characteristic information regarding the further amplification of the disturbance caused by the superposition of the disturbance after first-stage coupling amplification with periodic torque pulsations caused by bus voltage fluctuations at the force level and the fusion of temporal characteristics. The degree of stability deterioration can be the extent to which the steering system's driving behavior deviates from the normal power assist response after two stages of coupling amplification. The attenuation characteristics of recovery capability can be the decrease in the steering system's ability to recover from a disturbed state to a normal driving state.

[0065] Specifically, in the operation of new energy vehicles, continuous energy recovery, bus voltage fluctuations and random road torque interference often occur coupled. If the coupling amplification effect of the three is ignored, it will lead to abnormal steering torque output, increased response delay, steering lag, inaccurate handling, or even sudden deviation, which seriously threatens driving safety. Therefore, it is necessary to specifically analyze this coupling interference to avoid risks. To address the aforementioned issues: The process begins with synchronous acquisition and feature extraction of multi-source signals. For example, data from motor current, torque sensor, and bus voltage signals are acquired via the CAN bus. Wheel speed pulses and vehicle accelerometer information are used to assist in identifying random road surface interference events. Time-frequency analysis techniques (such as wavelet transform) are employed to decompose sudden torque fluctuations, extracting random interference components with aperiodic, wide-bandwidth characteristics. For instance, a disturbance pulse generated by driving over a manhole cover, lasting approximately 0.1 seconds, with its spectral energy concentrated in a specific frequency band (e.g., 30-50Hz), is quantified to quantify its initial amplitude and energy, forming "random interference characteristic information." Subsequently, identification and interference propagation path modeling techniques are used to couple this feature with the current torque output baseline drift identified from historical data (e.g., a persistent steady-state offset of approximately ± several Newton-meters). By constructing a transfer function model describing the modulation relationship between the two, the amplification effect of random interference in the offset background is simulated and calculated, for example, obtaining its peak value... The "first-stage coupling amplification information" is amplified several times. Then, a periodic torque pulsation model (e.g., a sinusoidal component with an amplitude of several Newton-meters and a frequency in the same frequency as the voltage fluctuation) is introduced from the voltage-torque correlation analysis. Using multi-signal fusion and time-series alignment analysis, the interference signal after the first stage amplification is superimposed and convolved with the periodic pulsation in the time domain. The analysis focuses on the amplitude superposition (e.g., peak superposition leading to instantaneous torque exceeding the limit) and the disturbance waveform distortion and duration extension caused by frequency intermodulation when the two meet at a specific phase. This reveals the more complex "second-stage coupling amplification information". Finally, based on Lyapunov stability theory or transient response performance indicators, the model under the combined interference after two stages of amplification is simulated and evaluated. The analysis quantifies the root mean square value of torque tracking error growth, phase margin decrease, and the significant extension trend of the time required to recover from the disturbance to steady state. This results in the formation of "multi-factor coupling interference information" that comprehensively characterizes the deterioration of stability and the decay of recovery capability.

[0066] The method provided in this embodiment accurately captures the two-stage amplification mechanism of multi-factor coupling, clearly presents the law of deterioration of steering gear drive stability, fills the gap of single-factor analysis, provides reliable data support for subsequent abnormal working condition identification and risk classification, predicts potential safety hazards in advance, improves the reliability of the steering system, ensures driving control safety and stability, and optimizes the driving experience.

[0067] In some embodiments, based on multi-factor coupled interference information, the deviation distribution, duration, and rate of change between the output torque and demand torque of the electric steering system are analyzed to obtain abnormal torque behavior characteristic information. Based on the abnormal torque behavior characteristic information, combined with the abnormal steering system drive operating condition information set, the evolutionary pattern of abnormal torque behavior exhibiting concentrated occurrence, periodic recurrence, or gradual increase in intensity within a complete driving condition cycle is analyzed to obtain abnormal operation mode information. Based on the abnormal operation mode information, the degree of decrease in drive stability and the potential impact on steering control corresponding to different abnormal operation modes are graded and evaluated to obtain safety risk level information. Based on the safety risk level information, a dynamic visualization early warning scheme matching different risk levels is configured, and a safety risk early warning containing timestamps, abnormal operation modes, risk levels, and key correlation parameters is generated, and an electric steering system drive abnormal operation monitoring log is output.

[0068] Abnormal torque behavior characteristic information can be information related to the deviation between the output torque and the required torque of the electric steering system. Abnormal operating mode information can be information on the evolution pattern of abnormal torque behavior within a complete driving cycle. Safety risk level information can be a risk level identifier based on the potential impact of abnormal operating modes on steering system drive stability and steering handling. Dynamic visualization early warning schemes can be early warning formats designed for different safety risk levels, which can intuitively present the risk status.

[0069] Specifically, during driving with electric power steering and regenerative braking coupled, if the electric power steering experiences abnormal driving due to multi-factor coupling interference and is not promptly warned and recorded, it can lead to steering lag, torque pulsation, and handling deviations. This can cause skidding at high speeds, potentially inducing collisions in complex road conditions, and increasing maintenance difficulty and safety hazards due to the lack of monitoring logs, making fault tracing difficult. To address these issues, the solution begins with deep feature extraction of the aforementioned "multi-factor coupling interference information." Through time-domain and frequency-domain statistical analysis, the real-time deviation between the actual output torque and the required torque is calculated, quantifying its "deviation distribution" (e.g., statistics show that approximately 70% of the deviation values ​​are concentrated in the -3Nm to +3Nm range, but there are outliers exceeding ±8Nm), "deviation duration" (e.g., identifying steady-state offset events lasting longer than 0.5 seconds), and "deviation change rate" (e.g., capturing a transient process where the deviation increases sharply by 5Nm within 0.05 seconds). This process accurately generates "abnormal torque behavior characteristic information." Subsequently, based on this characteristic information and integrating a complete "steering gear drive abnormal working condition information set" including vehicle speed, steering wheel angle, and regenerative braking intensity, a time window-based sequence pattern recognition and clustering analysis method is used to automatically identify the evolution pattern of abnormal behavior in a complete driving cycle. For example, it is found that certain abnormal features always "concentrate on appearing" after a continuous high-intensity kinetic energy recovery phase; or they are synchronized with the "bus voltage fluctuation" of a specific frequency (such as 100Hz) of the transmission system, showing "periodic recurrence." Alternatively, the deviation amplitude may exhibit a linear "gradual increase in intensity" with vehicle mileage or accumulated recovered energy. Then, using an evaluation model based on rules and a risk matrix, different modes are "graded and evaluated": for example, short-lived events that "occur frequently" are rated as medium risk because they may cause a momentary decrease in handling; while the "gradually increasing intensity" mode is directly rated as high risk because it foreshadows potential systemic performance degradation. Finally, based on the "safety risk level information" obtained from the evaluation, a "dynamic visualization warning scheme" is configured and triggered via the vehicle's HMI rendering engine and communication bus: for example, a yellow warning is displayed only on a small icon on the dashboard when the risk is low, while a flashing red icon and a brief alert sound are used when the risk is high. Simultaneously, the data recording service packages the current "timestamp," the determined "abnormal operating mode" (such as "periodic recurrence - associated bus fluctuation"), the assessed "risk level," and all "key associated parameters" (such as real-time deviation value, current recovery intensity, and motor temperature) to generate a structured "safety risk warning" record, which is ultimately summarized in chronological order into a complete "electric steering drive abnormal operation monitoring log" file.

[0070] The method provided in this embodiment visualizes anomalies and provides tiered warnings, enabling drivers to quickly perceive risks and respond in a timely manner, thus avoiding the risk of loss of control. The generated monitoring logs provide accurate traceability for maintenance, shortening the maintenance cycle. At the same time, the accumulation of anomaly data helps to optimize the steering gear design, improve its stability under complex operating conditions, and ensure driving safety and user experience.

[0071] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A method for analyzing abnormal operation state of an electric power steering of a new energy vehicle, characterized in that, include: Obtain a set of vehicle driving state information; based on the set of vehicle driving state information, analyze the dynamic response characteristics of the electric steering gear under the coupling effect of kinetic energy recovery and drive assist, and obtain a set of steering gear operating condition information. Based on the steering gear operating condition information set, the dynamic evolution behavior of the steering gear after experiencing the coupling effect of multiple factors such as continuous energy recovery, bus voltage fluctuation and torque interference is analyzed, and the steering gear drive abnormal operating condition information set is obtained. Based on the aforementioned abnormal operating condition information set of the electric steering gear drive, a visual dynamic presentation and safety risk warning are provided, and an abnormal operation monitoring log of the electric steering gear drive is output.

2. The method of claim 1, wherein, Based on the vehicle driving state information set, the dynamic response characteristics of the electric steering gear under the coupling effect of kinetic energy recovery and drive assist are analyzed to obtain the steering gear operating condition information set, including: The vehicle driving status information set includes a driving intention information set, an energy recovery information set, and a steering gear status information set. Based on the driving intention information set, the steering wheel angle change information and steering assist demand characteristics are analyzed to obtain steering gear demand condition information. Based on the steering gear demand condition information and the energy recovery information set, the dynamic superposition and offset effects of kinetic energy recovery intensity and duration on drive assistance under the influence of kinetic energy recovery are analyzed to obtain the steering kinetic energy drive response information set. Based on the steering gear kinetic drive response information set and the steering gear state information set, the dynamic following and deviation characteristics between the steering gear motor response parameters and kinetic response characteristics are analyzed to obtain the steering gear operating condition information set.

3. The method of claim 2, wherein, The process of constructing the steering gear demand condition information includes: Based on the driving intention information set, the impact of different steering wheel angle intervals on basic power assist demand is analyzed to obtain the basic steering information set. Based on the aforementioned basic steering information set, and combined with the steering wheel angle change information, the compensation trend of steering assist demand during dynamic steering is analyzed to obtain dynamic assist compensation information. Based on the dynamic power assist compensation information, the adjustment information of steering assist magnitude and response speed under different vehicle speed conditions is analyzed to obtain the steering gear demand condition information.

4. The method of claim 3, wherein, The process of constructing the directional motor energy-driven response information set includes: Based on the energy recovery information set, the corresponding information between the kinetic energy recovery intensity and the duration of a single recovery is analyzed to obtain the recovery load profile information characterizing the recovery force and duration. Based on the regenerative load profile information and the steering assist demand characteristics, the steady-state compensation and dynamic following process of the drive assist output to overcome the regenerative drag torque under different regenerative load profiles are analyzed to obtain the regenerative-assist coupling dynamic information. Based on the aforementioned recovery-assistance coupling dynamic information, the overall superposition enhancement or cancellation weakening relationship between kinetic energy recovery and vehicle drive assist is analyzed to obtain the directional kinetic energy drive response information set characterizing the effect of both.

5. The method of claim 4, wherein, Based on the steering gear kinetic energy drive response information set and combined with the steering gear state information set, the dynamic following and deviation characteristics between the steering gear motor response parameters and kinetic energy response characteristics are analyzed to obtain the steering gear operating condition information set, including: Based on the aforementioned kinetic energy drive response information set and combined with the aforementioned steering gear state information set, the dynamic following process of the electric steering gear response parameters to the kinetic energy response characteristics within a single kinetic energy recovery-assist coupling cycle is compared and analyzed to obtain periodic following characteristic information characterizing the instantaneous matching relationship between the two. Based on the periodic following characteristic information, the persistence and evolution trend of the deviation between the steering motor response parameters and the kinetic energy response characteristics after experiencing continuous kinetic energy recovery-assist coupling are analyzed to obtain steady-state offset evolution information. Based on the steady-state offset evolution information and the periodic following characteristic information, the local mismatch risk and the overall performance degradation risk caused by the accumulation of steady-state offset in the dynamic following process are analyzed to obtain the steering gear operating condition information set.

6. The method of claim 5, wherein, Based on the steering gear operating condition information set, the dynamic evolution behavior of the steering gear after experiencing the coupled effects of continuous energy recovery, bus voltage fluctuations, and torque disturbances is analyzed to obtain a steering gear drive abnormal operating condition information set, including: Based on the steering gear operating condition information set, the cumulative effects of steering gear torque output baseline drift and response delay caused by continuous energy recovery on steering gear driving behavior are analyzed to obtain energy recovery cumulative effect information. Based on the energy recovery cumulative effect information, the torque pulsation and controllable disturbance characteristics of the bus voltage fluctuation on the steering gear drive stability under the background of baseline drift and the delay cumulative effect are analyzed to obtain voltage fluctuation disturbance information. Based on the voltage fluctuation disturbance information, the coupling amplification effect between random road torque disturbance and baseline drift and torque pulsation during the electric steering and kinetic energy recovery coupling stage is analyzed to obtain multi-factor coupling disturbance information. Based on the multi-factor coupled interference information, the overall deviation pattern and abnormal characteristics of the steering gear driving behavior relative to the normal power assist demand response are analyzed to obtain the steering gear driving abnormal condition information set.

7. The method according to claim 6, characterized in that, The process of constructing the cumulative effect information of energy recovery includes: Based on the steering gear operating condition information set, the average variation characteristics of the basic compensation torque output by the steering gear motor to overcome the reverse drag torque are analyzed within the continuous kinetic energy recovery cycle, and the torque output baseline drift information characterizing the long-term deviation of the compensation demand is obtained. Based on the torque output baseline drift information, the evolution of the start time of the steering motor response control command and the time to reach the target torque under the background of baseline drift is analyzed as the number of times the regenerative load is applied increases, and the cumulative evolution information of response delay is obtained. Based on the cumulative evolution information of the response delay, the correlation between baseline drift information and the cumulative delay amount during the continuous energy recovery process is analyzed to obtain the cumulative effect information of energy recovery.

8. The method of claim 7, wherein, The process of constructing the voltage fluctuation disturbance information includes: Based on the energy recovery cumulative effect information, the amplitude and frequency characteristics of voltage fluctuations are analyzed during the dynamic change of bus voltage caused by kinetic energy recovery, and fluctuation profile information characterizing the voltage disturbance source is obtained. Based on the fluctuation profile information and the torque output baseline drift information, the law of periodic oscillation of the electric steering output torque caused by the voltage fluctuation is analyzed at the electric steering operating point corresponding to the baseline drift, and the torque pulsation generation information is obtained. Based on the torque pulsation generation information and the response delay cumulative evolution information, the interference characteristics of the torque pulsation on the controller's precise adjustment of the motor output are analyzed in the context of an already existing delay in the electric power steering response, thus obtaining the voltage fluctuation disturbance information characterizing the decrease in drive stability.

9. The method of claim 8, wherein, The process of constructing the multi-factor coupling interference information includes: Based on the voltage fluctuation disturbance information, the initial disturbance response characteristics when encountering random road surface torque disturbance during the electric steering assist process are analyzed to obtain random disturbance characteristic information; Based on the random disturbance feature information and combined with the torque output baseline drift information, the interaction between the random road surface torque disturbance and the existing electric steering torque output baseline is analyzed to obtain the first-level coupling amplification information; Based on the first-stage coupling amplification information, combined with the torque pulsation generation information, the interference after the first-stage coupling amplification is analyzed. The interference is superimposed and amplified at the force level and the periodic torque pulsation caused by the bus voltage fluctuation is fused in the time sequence to obtain the second-stage coupling amplification information. Based on the second-stage coupling amplification information, the stability deterioration and recovery capability attenuation characteristics of the electric steering drive behavior after two-stage coupling amplification are analyzed to obtain the multi-factor coupling interference information.

10. The method of claim 9, wherein, The system, based on the abnormal operating condition information set of the electric steering gear drive, performs visual dynamic presentation and safety risk warning, and outputs an abnormal operation monitoring log of the electric steering gear drive, including: Based on the multi-factor coupled interference information, the deviation distribution, deviation duration and deviation change rate between the output torque and the required torque of the electric steering system are analyzed to obtain abnormal torque behavior characteristic information; Based on the abnormal torque behavior characteristic information and combined with the steering gear drive abnormal working condition information set, the evolution law of the abnormal torque behavior in the complete driving working condition cycle, which shows the concentrated occurrence, periodic recurrence or gradual increase in intensity, is analyzed to obtain abnormal operation mode information. Based on the abnormal operating mode information, the degree of decrease in drive stability and the potential impact on steering control corresponding to different abnormal operating modes are graded and evaluated to obtain safety risk level information. Based on the safety risk level information, a dynamic visualization early warning scheme matching different risk levels is configured, and the safety risk early warning containing timestamps, abnormal operation modes, risk levels and key related parameters is generated, and the abnormal operation monitoring log of the electric steering gear drive is output.

Images