Method and system capable of accurately estimating angle value in TAS fault

Abstract

The invention relates to the technical field of automobile electronic control, and discloses a method and system capable of accurately estimating an angle value in TAS fault. The method comprises the following steps: acquiring an angle data set of a TAS sensor and a motor position sensor; performing fault state analysis processing on the angle data set to generate a TAS fault identifier and a fault detection moment; based on the fault detection time and the cached angle data set, reference value determination processing is carried out, and a TAS reference value and a motor angle reference value are generated; calling a predefined transmission ratio relation to carry out conversion processing on a difference value between the current motor angle value and the motor angle reference value, and generating an estimated value of a TAS angle; performing time drift compensation correction processing on the estimated value of the TAS angle to generate a corrected estimated value of the TAS angle; and generating a control strategy set of the electric power steering system according to the corrected TAS angle estimated value. According to the method, accurate angle estimation can be continuously provided when the TAS sensor fails, and functional safety and control continuity of a steering system are guaranteed.

Description

Technical Field

[0001] This invention relates to the field of automotive electronic control technology, specifically to a method and system for accurately estimating angle values ​​in the event of a TAS (Transmission Actuation System) failure. Background Technology

[0002] In existing EPS systems, the TAS sensor is used to detect the driver's steering torque and angle, and is a key sensor for EPS control. When the TAS fails, existing technologies typically use a backup sensor or directly use the motor position sensor to estimate the TAS angle. A typical existing solution is to immediately switch to using the motor position sensor signal when a TAS fault is detected, and calculate the equivalent TAS angle value through a fixed transmission ratio. This method has a time delay in fault detection (usually 10-30ms), during which the TAS value may be unreliable. Directly using the TAS value at the moment of fault detection as the estimation benchmark leads to accumulated errors, does not consider the impact of the fault detection delay on the angle estimation accuracy, resulting in inaccurate steering control; moreover, during the transient process of a fault, system stability is affected. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for accurately estimating angle values ​​in the event of a TAS failure, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a method and system for accurately estimating angle values ​​during TAS (Transient Aperture Screw) failure, the method comprising: Collect angle data sets from TAS sensor and motor position sensor, the angle data sets including TAS angle sequences and motor angle sequences; The angle data set is subjected to fault status analysis and processing to generate TAS fault identifiers and fault detection times; Based on the fault detection time and the cached angle data set, a reference value determination process is performed to generate the TAS reference value and the motor angle reference value. The difference between the current motor angle value and the motor angle reference value is converted by calling the predefined transmission ratio relationship to generate an estimated value of the TAS angle; The estimated TAS angle is corrected by time drift compensation to generate a corrected estimated TAS angle. A set of electric power steering system control strategies is generated based on the corrected TAS angle estimation value. The set of electric power steering system control strategies includes a power assist torque adjustment scheme and a fault warning scheme.

[0005] Preferably, the set of angle data collected from the TAS sensor and the motor position sensor includes: The TAS angle signal and the motor angle signal are acquired synchronously at a fixed sampling frequency, and the TAS angle signal and the motor angle signal are stored in a circular data buffer. The capacity of the circular data buffer ensures that the data storage time covers the maximum delay time of fault detection. The stored TAS angle signal and motor angle signal are timestamped to form a time-aligned angle data sequence.

[0006] Preferably, the step of performing fault state analysis processing on the angle data set to generate TAS fault identifiers and fault detection times includes: The amplitude change and signal continuity of the TAS angle signal are continuously monitored. When the amplitude exceeds the preset effective range or the duration of signal loss reaches the fault judgment threshold, the fault indicator is activated. The system time when the fault identifier is activated is recorded as the fault detection time, and the fault detection time is associated with the timestamp of the angle data sequence; The step of associating the fault detection time with the timestamp of the angle data sequence includes: Obtain the system clock value at the time of fault detection and convert it into a time base reference that is the same as the timestamp of the angle data sequence; Search the circular data buffer for the timestamp closest to the fault detection time. If an exact matching timestamp exists, directly associate the angle data corresponding to that timestamp. If no exact matching timestamp exists, select the two closest timestamps before and after the fault detection time, and calculate the time offset between the fault detection time and the two timestamps. Based on the magnitude of the time offset, an association strategy is determined: when the time offset is less than or equal to the allowable error threshold, the angle data of the closer timestamp is associated; when the time offset is greater than the allowable error threshold, the time synchronization calibration process is triggered, the time base reference is realigned, and the association is performed again.

[0007] Preferably, the step of determining the reference value based on the fault detection time and the cached angle data set to generate the TAS reference value and the motor angle reference value includes: Calculate the baseline reference time point based on the fault detection time and the maximum fault detection delay time; Retrieve the historical TAS angle value and historical motor angle value corresponding to the reference time point from the circular data buffer; If there is no direct data for the reference time point, an interpolation algorithm is used to calculate the TAS reference value and the motor angle reference value from the data of adjacent time points; The process of calculating the TAS reference value and motor angle reference value from data at adjacent time points using an interpolation algorithm includes: Identify two adjacent historical data points before and after the reference time point, and obtain their timestamps and corresponding angle values ​​respectively; Calculate the ratio of the time interval between the reference time point and its adjacent time points; Based on the principle of linear interpolation, the angle values ​​of adjacent time points are weighted using the time interval ratio to obtain the interpolated angle value of the reference time point. Interpolation calculations were performed on the TAS angle and the motor angle respectively to obtain the TAS reference value and the motor angle reference value; Verify the reasonableness of the interpolation calculation results: Check whether the interpolated angle value is within the interval formed by the adjacent angle values. If it exceeds the interval, use the secondary interpolation algorithm to recalculate.

[0008] Preferably, the step of calling a predefined transmission ratio relationship to convert the difference between the current motor angle value and the motor angle reference value to generate an estimated value of the TAS angle includes: Calculate the real-time angle deviation between the current motor angle value and the motor angle reference value; The real-time angle deviation is converted into an equivalent change in TAS angle based on a predefined transmission ratio relationship. The equivalent change in the TAS angle is added to the TAS reference value to generate a preliminary estimate of the TAS angle. The step of converting the real-time angle deviation into an equivalent change in TAS angle based on a predefined transmission ratio relationship includes: Query the predefined transmission ratio relationship table to obtain the transmission ratio coefficient under the current working condition. The transmission ratio coefficient represents the proportional relationship between the change in motor angle and the change in TAS angle. Multiplying the real-time angle deviation by the transmission ratio coefficient yields the preliminary equivalent change in TAS angle. The directional consistency of the preliminary TAS angle equivalent change is verified: the sign of the equivalent change is adjusted according to the correspondence between the motor rotation direction and the TAS angle change direction. The equivalent change is corrected by applying a nonlinear compensation factor, which is dynamically adjusted based on the historical wear characteristics of the transmission system to compensate for the error caused by the transmission clearance.

[0009] Preferably, the step of performing time drift compensation correction processing on the estimated value of the TAS angle to generate a corrected estimated value of the TAS angle includes: Estimate the system processing delay time from fault detection to the current moment, and calculate the angle drift compensation amount based on the system processing delay time; The angle drift compensation is applied to the initial TAS angle estimate, and phase correction processing is performed to generate the corrected TAS angle estimate.

[0010] Preferably, the step of generating a set of electric power steering system control strategies based on the corrected TAS angle estimation value includes: Input the corrected TAS angle estimate into the assist torque lookup table to obtain the target assist torque value and generate an assist torque adjustment command; The TAS fault identifier is used to generate a graded fault warning signal, and the warning level is dynamically adjusted according to the fault duration.

[0011] Preferably, the method further includes: After the TAS sensor resumes normal operation, the actual TAS angle value sequence is collected; The corrected TAS angle estimate during the fault period is compared with the actual TAS angle value sequence by deviation analysis to generate an estimation error statistic. Adjust the predefined transmission ratio relationship or the maximum fault detection delay time based on the estimated error statistics; The step of performing deviation analysis on the corrected TAS angle estimate and the actual TAS angle value sequence during the fault period to generate an estimation error statistic includes: Align the time series during the fault period by matching the corrected TAS angle estimation sequence with the actual TAS angle value sequence by timestamp; Calculate the angular deviation value at each matching time point to form a deviation value sequence; Statistical analysis is performed on the deviation value sequence to calculate the mean, standard deviation, and maximum absolute deviation of the deviation values; Generate a histogram of deviation distribution and calculate the kurtosis and skewness indices of the deviation values ​​to describe the shape of the deviation distribution; Based on the mean and standard deviation, the confidence interval of the estimation error is calculated, which serves as the core output of the estimation error statistic.

[0012] Preferably, adjusting the predefined transmission ratio relationship or the maximum fault detection delay time based on the estimated error statistics includes: Calculate the mean and standard deviation of the estimation error. If the mean exceeds the error tolerance threshold, then use the recursive least squares method to update the transmission ratio parameters. If the standard deviation indicates that the estimation is too volatile, increase the maximum delay time for fault detection to improve stability.

[0013] Preferably, the present invention also includes a system capable of accurately estimating angle values ​​in the event of a TAS failure, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, when the processor executes the computer program, it implements the steps of the method described above for accurately estimating angle values ​​in the event of a TAS failure.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. Improve estimation accuracy: By using a reliable baseline value before the fault occurs, errors caused by fault detection delays are avoided.

[0015] 2. Enhanced system reliability: It can still provide accurate angle information in the event of a TAS failure, ensuring the normal operation of the EPS system.

[0016] 3. Improved driving safety: Accurate angle estimation avoids abrupt changes in steering control, thus improving vehicle safety.

[0017] 4. Reduce system costs: No additional hardware sensors are required; fault-tolerant control is achieved through software algorithms.

[0018] 5. High adaptability: The cache time and backtracking strategy can be adjusted according to different fault detection characteristics. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the working principle of a method and system for accurately estimating angle values ​​during TAS failure, as described in this invention. Figure 2 A flowchart illustrating the association between fault detection time and the timestamp of the angle data sequence; Detailed Implementation The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Please see Figure 1 This invention provides a method and system for accurately estimating angle values ​​during TAS (Transient Aperture Screw) failure, the method comprising: Based on the collection of angle data from the TAS sensor and the motor position sensor, which includes TAS angle sequences and motor angle sequences; Fault status analysis and processing are performed on the angle data set to generate TAS fault identifier and fault detection time. Based on the fault detection time and the cached angle data set, reference value determination processing is performed to generate TAS reference value and motor angle reference value. The predefined transmission ratio relationship is called to convert the difference between the current motor angle value and the motor angle reference value, generate an estimated value of TAS angle, and perform time drift compensation correction on the estimated value of TAS angle to generate a corrected estimated value of TAS angle. A set of control strategies for the electric power steering system is generated based on the corrected TAS angle estimation value. This set of control strategies includes a power steering torque adjustment scheme and a fault warning scheme.

[0021] Example 1: Please see Figure 2 When collecting angle data sets from the TAS sensor and motor position sensor, the TAS angle signal and motor angle signal are acquired synchronously at a fixed sampling frequency and stored in a circular data buffer. The capacity of this circular data buffer ensures that the data storage time covers the maximum delay time for fault detection. The stored TAS angle signal and motor angle signal are timestamped to form a time-aligned angle data sequence. Fault state analysis processing is performed on the angle data set to generate a TAS fault identifier and a fault detection time. The amplitude change and signal continuity of the TAS angle signal are continuously monitored. When the amplitude exceeds the preset effective range or the signal loss duration reaches the fault judgment threshold, the fault identifier is activated. The system time when the fault identifier is activated is recorded as the fault detection time, and the fault detection time is associated with the timestamp of the angle data sequence. Associating the fault detection time with the timestamp of the angle data sequence involves obtaining the system clock value of the fault detection time and converting it into a time base reference identical to the timestamp of the angle data sequence. The system clock then searches the circular data buffer for the timestamp closest to the fault detection time. If a perfectly matching timestamp exists, the angle data corresponding to that timestamp is directly associated. If no perfectly matching timestamp exists, the two closest timestamps before and after the fault detection time are selected, and the time offset between the fault detection time and the two timestamps is calculated. The association strategy is determined based on the magnitude of the time offset. When the time offset is less than or equal to the allowable error threshold, the angle data with the closer timestamp is associated. When the time offset is greater than the allowable error threshold, a time synchronization calibration process is triggered, the time base reference is realigned, and the association is performed again.

[0022] In practice, the fixed sampling frequency of the TAS sensor is set to 1000 Hz, and the fixed sampling frequency of the motor position sensor is also set to 1000 Hz. The system synchronously acquires the TAS angle signal and the motor angle signal at this fixed sampling frequency. The instantaneous value of the TAS angle signal at a certain moment is 1.57 radians, and the instantaneous value of the motor angle signal at the same moment is 78.5 radians. The synchronously acquired TAS angle signal and motor angle signal are stored in a circular data buffer. The preset capacity of the circular data buffer is to store data for 100 milliseconds. This data storage time covers the preset maximum delay time of 80 milliseconds for fault detection. The system timestamps the stored TAS angle signal and motor angle signal. The time base of the timestamp comes from the system's high-precision clock counter, thus forming a time-strictly aligned angle data sequence.

[0023] In some embodiments, the system performs fault state analysis processing on the time-aligned angle data sequence. During the processing, the amplitude change and signal continuity of the TAS angle signal are continuously monitored. The preset effective range of the TAS angle signal is defined as -3.14 radians to +3.14 radians. When the monitoring logic detects that the amplitude of the TAS angle signal is 5.0 radians for 10 milliseconds, which exceeds the preset effective range, and the signal loss duration reaches the fault judgment threshold of 20 milliseconds, the system immediately activates the TAS fault flag. The system records the system time when the TAS fault flag is activated as the fault detection time. The system time is derived from the clock cycle count of the central processing unit. The recorded fault detection time is associated with the timestamp of the angle data sequence.

[0024] In the specific implementation, the operation of associating the fault detection time with the timestamp of the angle data sequence begins. The system obtains the system clock value at the fault detection time, which is an integer count value. Through a preset conversion relationship, this integer count value is converted into a millisecond-level time base reference that is the same as the timestamp of the angle data sequence. The system starts a search in the circular data buffer to find the timestamp closest to the fault detection time. During the search process, it is found that there is a timestamp in the circular data buffer that is completely consistent with the millisecond-level time base reference of the fault detection time. There is an exact matching timestamp. The association logic directly associates the historical TAS angle value and historical motor angle value corresponding to the exact matching timestamp.

[0025] Understandably, in another scenario, if the system does not find a timestamp that precisely matches the fault detection time in the circular data buffer, the association logic will select the two closest timestamps before and after the fault detection time. For example, the time corresponding to the previous timestamp is T1, and the time corresponding to the next timestamp is T2. The time offsets between the fault detection time Tc and T1 and T2 are calculated as ΔT1 and ΔT2, respectively. The system determines the association strategy based on the magnitude of the time offsets. When the calculated minimum time offset ΔT1 is less than or equal to the system's set 1-millisecond allowable error threshold, the association logic associates the angle data corresponding to the T1 timestamp. When the minimum time offset ΔT2 is greater than the 1-millisecond allowable error threshold, the system triggers the time synchronization calibration process. The time synchronization calibration process realigns the time base references of the data acquisition subsystem and the control subsystem. After the time base references are realigned, the system performs the association operation between the fault detection time and the angle data sequence timestamps again.

[0026] Optionally, the accuracy of timestamp association directly affects the accuracy of baseline value determination. Therefore, during the data acquisition hardware design phase, it is necessary to ensure that the sampling trigger signals of the TAS sensor and the motor position sensor have a high degree of synchronization. At the same time, the management strategy of the circular data buffer adopts an overwrite mechanism. When the buffer is full, the new data will overwrite the oldest data, thereby continuously retaining the angle data sequence aligned to the most recent 100 milliseconds, providing the necessary data foundation for the association of fault detection moments.

[0027] Example 2: Based on the fault detection time and the cached angle data set, a baseline value determination process is performed to generate the TAS baseline value and the motor angle baseline value. A baseline reference time point is calculated based on the fault detection time and the maximum fault detection delay time. Historical TAS angle values ​​and historical motor angle values ​​corresponding to the baseline reference time point are retrieved from the circular data buffer. If no direct data is available for the baseline reference time point, an interpolation algorithm is used to calculate the TAS baseline value and motor angle baseline value from data at adjacent time points. The interpolation algorithm for calculating the TAS baseline value and motor angle baseline value from data at adjacent time points includes identifying two adjacent historical data points before and after the baseline reference time point and obtaining their timestamps and corresponding angle values. The time interval ratio between the baseline reference time point and the adjacent time points is calculated. Based on the linear interpolation principle, the time interval ratio is used to weight the angle values ​​at the adjacent time points to obtain the interpolated angle value at the baseline reference time point. Interpolation calculations are performed on the TAS angle and motor angle respectively to obtain the TAS baseline value and motor angle baseline value. The rationality of the interpolation calculation results is verified, and it is checked whether the interpolated angle value is within the interval formed by the adjacent angle values. If it exceeds the interval, a secondary interpolation algorithm is used to recalculate.

[0028] In practice, after the fault state analysis and processing generates the fault detection time, the system performs baseline value determination processing based on the fault detection time and the cached angle data set. The fault detection time is recorded as follows: Its value is 1000 milliseconds, which is the maximum preset delay time for fault detection within the system. The baseline value is 80 milliseconds. The process for determining the baseline value is first based on the fault detection time. and maximum delay time for fault detection Calculation of reference time point The calculation relationship is as follows Substitute the values ​​to obtain the baseline reference time point. It takes 920 milliseconds.

[0029] In some embodiments, the system retrieves a reference time point from a circular data buffer. The retrieval process for the corresponding historical TAS angle values ​​and historical motor angle values ​​revealed that there was no direct data point exactly 920 milliseconds in the timestamp sequence stored in the circular data buffer. Therefore, the system used an interpolation algorithm to calculate the TAS reference value and motor angle reference value from data at adjacent time points. The specific process of the interpolation algorithm to calculate the TAS reference value and motor angle reference value from data at adjacent time points involves identifying the reference time point. Between two adjacent historical data points, the timestamp of the previous historical data point The corresponding TAS angle value is 910 milliseconds. 1.50 radians corresponds to the motor angle value. The timestamp of the next historical data point is 75.0 radians. 930 milliseconds, corresponding to the TAS angle value A value of 1.70 radians corresponds to the motor angle value. It is 85.0 radians. System calculation reference time point. The time interval ratio is calculated for the TAS angle, based on the time interval ratio between adjacent time points. The calculation formula is expressed as Substituting the specific values ​​yields Based on the principle of linear interpolation, a time interval scaling factor is used. The TAS angle values ​​at adjacent time points are weighted and calculated to obtain the reference time point. TAS interpolation angle value The calculation relationship is Substituting the values, we get Radius, this value is the TAS reference value.

[0030] It is understandable that the same interpolation calculation is performed on the motor angle to obtain the reference value of the motor angle, and the time interval scaling factor of the motor angle is calculated. , Because the timestamps are the same, The value is also 0.5. The interpolation calculation relationship for the motor angle is as follows: Substituting the values, we get In radians, this value is the reference value for the motor angle. The system verifies the reasonableness of the interpolation calculation results and checks the TAS interpolated angle value. (1.60 radians) Is it within the adjacent TAS angle values? (1.50 radians) and Within the interval formed by (1.70 radians), check the motor interpolation angle value. Is (80.0 radians) within the range? (75.0 radians) and Within the interval formed by (85.0 radians), the current interpolation results are all within the corresponding interval, and the verification is successful.

[0031] Optionally, in another implementation scenario, if verification reveals that the angle value calculated by interpolation exceeds the interval formed by the angle values ​​of adjacent historical data points, the system will use a secondary interpolation algorithm to recalculate the baseline value. The secondary interpolation algorithm will use the baseline reference time point. A quadratic polynomial is constructed using at least three neighboring historical data points for fitting, resulting in smoother and more accurate TAS and motor angle reference values. The TAS and motor angle reference values ​​generated by the reference value determination process will serve as the starting reference points for subsequent angle conversions.

[0032] Example 3: The process involves calling a predefined transmission ratio relationship to convert the difference between the current motor angle value and the motor angle reference value to generate an estimated TAS angle. It calculates the real-time angle deviation between the current motor angle value and the motor angle reference value, converts the real-time angle deviation into an equivalent change in TAS angle based on the predefined transmission ratio relationship, and adds the equivalent change in TAS angle to the TAS reference value to generate a preliminary estimated TAS angle value. Converting the real-time angle deviation into an equivalent change in TAS angle based on the predefined transmission ratio relationship involves querying a predefined transmission ratio relationship table to obtain the transmission ratio coefficient under the current operating condition. This transmission ratio coefficient represents the proportional relationship between the change in motor angle and the change in TAS angle. The real-time angle deviation is multiplied by the transmission ratio coefficient to obtain the preliminary equivalent change in TAS angle. The preliminary equivalent change in TAS angle is then checked for directional consistency. The sign of the equivalent change is adjusted according to the correspondence between the motor rotation direction and the TAS angle change direction. Finally, a nonlinear compensation factor is applied to correct the equivalent change. This nonlinear compensation factor is dynamically adjusted based on the historical wear characteristics of the transmission system to compensate for errors caused by transmission backlash.

[0033] In practical implementation, after the reference value determination process generates the motor angle reference value, the system calls the predefined transmission ratio relationship to perform conversion processing on the difference between the current motor angle value and the motor angle reference value, and sets the motor angle reference value obtained from Example 2. The value is 80.0 radians, and in a certain control cycle after the fault occurred, the system collected the current motor angle value. The value is 95.0 radians. The conversion process first calculates the current motor angle value. relative to motor angle reference value Real-time angle deviation The calculation relationship is as follows: Substituting the values, we obtain the real-time angle deviation. radian.

[0034] In some embodiments, the system converts the real-time angle deviation into an equivalent change in the TAS angle based on a predefined transmission ratio relationship. This process requires querying a predefined transmission ratio relationship table. The predefined transmission ratio relationship table stores transmission ratio coefficients for different steering column positions or vehicle speeds. Assuming the transmission ratio coefficient obtained under the current operating conditions is... The transmission ratio coefficient is 0.02. This indicates the proportional relationship between the change in motor angle and the change in TAS angle. The system will display the real-time angle deviation. Multiply by the transmission ratio coefficient The preliminary equivalent change in TAS angle was obtained. The calculation relationship is Substituting the values, we obtain the preliminary equivalent change in the TAS angle. radian.

[0035] It is understandable that the system performs a directional consistency check on the initial equivalent change in the TAS angle. This directional consistency check requires adjusting the sign of the equivalent change based on the correspondence between the motor rotation direction and the TAS angle change direction. In one correspondence, an increase in the motor angle is defined as representing a rightward turn of the steering wheel, and the TAS angle value also increases accordingly. If, at this time, the motor angle sensor detects that the motor rotation direction is forward (corresponding to an angle increase), then the initial equivalent change in the TAS angle... If the calculation result is positive and the direction is consistent, then the sign does not need to be adjusted and should be maintained. Radius. In another case, if the motor rotation direction is detected as reversed (corresponding to a decrease in angle), the initial TAS angle equivalent change... If the calculated value is positive but the direction is inconsistent, then it is necessary to... The sign is inverted. The system applies a nonlinear compensation factor to correct the equivalent change. Dynamic adjustment based on the historical wear characteristics of the transmission system aims to compensate for the nonlinear error caused by transmission backlash. The nonlinear compensation factor... The values ​​are provided by another independent wear estimation algorithm, such as The corrected equivalent change in TAS angle. From the formula The calculation yields the result; substituting the values, we get... Radius. In the formula, This represents the equivalent change in the corrected TAS angle. This represents the nonlinear compensation factor. This represents the initial equivalent change in the TAS angle.

[0036] Optionally, the final equivalent change in the TAS angle can be determined. Compared with TAS benchmark value Add them together to generate a preliminary estimate of the TAS angle. Set the TAS reference value obtained from Example 2. The value is 1.60 radians, and the calculation relationship is as follows: Substituting the values, we obtain a preliminary estimate of the TAS angle. Radius. At this point, by calling the predefined transmission ratio relationship for conversion processing, a preliminary TAS angle estimate value calculated based on the motor angle change was generated.

[0037] Example 4: The estimated TAS angle is corrected by time drift compensation to generate a corrected TAS angle estimate. The system processing delay time from fault detection to the current moment is estimated, and the angle drift compensation amount is calculated based on the system processing delay time. The angle drift compensation amount is applied to the initial TAS angle estimate for phase correction to generate the corrected TAS angle estimate. A set of electric power steering system control strategies is generated based on the corrected TAS angle estimate. The corrected TAS angle estimate is input into the power steering torque lookup table to obtain the target power steering torque value and generate a power steering torque adjustment command. A graded fault warning signal is generated by combining the TAS fault identifier, and the warning level is dynamically adjusted according to the fault duration.

[0038] In practical implementation, after generating the initial TAS angle estimate, the system performs time drift compensation correction on the initial TAS angle estimate. The initial TAS angle estimate is calculated in the previous processing step, for example, its value is 1.915 radians. The time drift compensation correction process needs to estimate the system processing delay time from the fault detection time to the current time. The system processing delay time includes the accumulation of multiple factors such as signal sampling delay, data processing calculation cycle, and communication transmission delay. By monitoring the timestamp of the system task scheduler in real time, the system processing delay time within the current control cycle is estimated. It takes 5 milliseconds.

[0039] In some embodiments, based on system processing delay time The angle drift compensation is calculated to correct the phase lag in the TAS angle estimation caused by processing delay. The system calculates an average angle change rate based on historical data of the TAS angle change rate over a period of time before the fault occurred. For example, the average rate of change of angle The angle drift compensation is 0.1 radians per second. The calculation relationship is as follows Substituting the values, we obtain the angle drift compensation amount. Radius. The system compensates for angular drift. The initial TAS angle estimate is applied to a phase correction process, which compensates for the angle drift. Add the initial TAS angle estimate to generate the corrected TAS angle estimate. The calculation relationship is Substituting the values, we obtain the corrected estimated TAS angle. radian.

[0040] It is understandable that a set of control strategies for the electric power steering system is generated based on the corrected TAS angle estimate. This set of control strategies includes a power assist torque adjustment scheme and a fault warning scheme. The system will then use the corrected TAS angle estimate... Input the assist torque lookup table, which is a preset two-dimensional mapping table. Its inputs are the TAS angle value and the vehicle speed value, and the output is the target assist torque value. Assuming the current vehicle speed... For 60 km / h, consult the assist torque lookup table to obtain the corresponding value. radians and Target assist torque value per kilometer per hour The torque is 8.5 Nm. The system generates a torque adjustment command and sends the command to the motor controller to execute the torque adjustment scheme.

[0041] Optionally, the system combines TAS fault identifiers to generate graded fault warning signals. TAS fault identifiers are activated during the fault status analysis and processing phase. The graded fault warning signals dynamically adjust the warning level according to the fault duration. For example, when the fault duration is less than 2 seconds, a level 1 warning signal is generated, which alerts the driver via a yellow indicator light on the instrument panel. When the fault duration is greater than or equal to 2 seconds, a level 2 warning signal is generated, which alerts the driver via a red indicator light on the instrument panel and an audible alert. The dynamic adjustment of the graded fault warning signals ensures that the fault warning scheme can provide corresponding warnings according to the severity of the fault.

[0042] Example 5: After the TAS sensor resumes normal operation, the actual TAS angle value sequence is collected. Deviation analysis is performed between the corrected TAS angle estimate during the fault period and the actual TAS angle value sequence to generate an estimation error statistic. Based on the estimation error statistic, the predefined transmission ratio relationship or the maximum fault detection delay time is adjusted. The deviation analysis process for generating the estimation error statistic includes aligning the time series during the fault period, matching the corrected TAS angle estimate sequence with the actual TAS angle value sequence by timestamp, calculating the angle deviation value at each matching time point to form a deviation value sequence, and performing statistical analysis on the deviation value sequence to calculate the mean, standard deviation, and maximum absolute deviation of the deviation values. A deviation distribution histogram is generated, and the kurtosis and skewness indices of the deviation values ​​are calculated to describe the deviation distribution pattern. The confidence interval of the estimation error is calculated based on the mean and standard deviation, serving as the core output of the estimation error statistic. Adjusting the predefined transmission ratio relationship or the maximum delay time for fault detection based on the estimated error statistics includes calculating the average and standard deviation of the estimated error. If the average exceeds the error tolerance threshold, the transmission ratio parameters are updated using the recursive least squares method. If the standard deviation indicates that the estimation fluctuation is too large, the maximum delay time for fault detection is increased to improve stability.

[0043] In practice, after the TAS sensor resumes normal operation, the system collects the actual TAS angle value sequence. The actual TAS angle value sequence comes from the newly valid TAS sensor signal, assuming that the fault period started from the fault detection time. When the TAS sensor returns to normal The system records and stores a sequence of corrected TAS angle estimates calculated for each control cycle within this time period. This sequence contains... Data points, for example Each data point corresponds to a timestamp and a corrected estimated TAS angle value. After the TAS sensor returns to normal, the system synchronously collects and records the actual TAS angle value sequence at the same sampling frequency. The time range covered by the actual TAS angle value sequence is aligned with the time range during the fault period.

[0044] In some embodiments, the system performs deviation analysis on the corrected TAS angle estimate and the actual TAS angle value sequence during the fault period. The deviation analysis first aligns the time series during the fault period, precisely matching the corrected TAS angle estimate sequence and the actual TAS angle value sequence using the same timestamp. The timestamp matching is based on the system's global clock, and the resulting sequence... For data, for example, in the first Matching time points The corrected TAS angle estimate is The actual TAS angle value is The system calculates the angular deviation value at each matching time point. The calculation formula is: ,right After calculation at each time point, a result is formed containing Deviation value sequence of each deviation value An exemplary sequence of deviation values ​​is as follows: radians. The system's bias value sequence Perform statistical analysis and calculate the average value of the deviation. ,average value The overall deviation of the estimated values ​​is characterized by the standard deviation of the deviation values. Standard deviation It characterizes the degree of fluctuation in the estimated values ​​and calculates the maximum absolute deviation. Maximum absolute deviation It is a sequence of deviation values The element with the largest absolute value. The system generates a deviation distribution histogram, which divides the range of deviation values ​​into several continuous intervals and counts the number of deviation values ​​falling into each interval. The system calculates the kurtosis index of the deviation values. kurtosis index The system calculates the skewness index of the deviation values ​​to describe the steepness of the deviation distribution pattern. Skewness index Describes the asymmetry in the distribution of deviations. The system is based on the average value. and standard deviation Calculate the confidence interval for the estimation error. The confidence interval is... ,in It is a coverage factor that depends on the confidence level, for example At a 95% confidence level, this confidence interval serves as the core output of the estimation error statistic.

[0045] It is understandable that the system adjusts the predefined transmission ratio relationship or the maximum fault detection delay time based on the estimation error statistics, and calculates the average value of the estimation error. Given a value of 0.025 radians, estimate the standard deviation of the error. The error tolerance threshold is 0.05 radians. The value is 0.02 radians, and the calculation results show that the average value is... (0.025 radians) exceeds the error tolerance threshold (0.02 radians), the system uses recursive least squares method to update the transmission ratio parameters in the transmission ratio relationship. The recursive least squares method utilizes the deviation value sequence. The transmission ratio coefficients in the predefined transmission ratio table are updated online iteratively with the corresponding sequence of motor angle changes, so that the average deviation of subsequent estimates approaches zero.

[0046] Optionally, in another implementation scenario, the system presets an estimated fluctuation threshold. The standard deviation is 0.03 radians. A deviation of 0.08 radians indicates excessive estimation fluctuation and a high standard deviation. Greater than the estimated fluctuation threshold The system then increases the maximum delay time for fault detection to improve stability, for example, by reducing the original maximum delay time for fault detection. The increase from 80 milliseconds to 100 milliseconds, resulting in a longer maximum fault detection delay, means that the baseline determination process can select earlier and more stable historical angle data as the benchmark, thereby reducing estimation fluctuations caused by data disturbances instantaneously before the fault. By adjusting the predefined transmission ratio relationship or the maximum fault detection delay, the system achieves self-calibration and optimization of the estimation method during the fault period after the TAS sensor returns to normal.

[0047] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for accurately estimating angle values ​​during TAS (Transportation Aqueduct) failure, characterized in that, The method includes: Collect angle data sets from TAS sensor and motor position sensor, the angle data sets including TAS angle sequences and motor angle sequences; The angle data set is subjected to fault status analysis and processing to generate TAS fault identifiers and fault detection times; Based on the fault detection time and the cached angle data set, a reference value determination process is performed to generate the TAS reference value and the motor angle reference value. The difference between the current motor angle value and the motor angle reference value is converted by calling the predefined transmission ratio relationship to generate an estimated value of the TAS angle; The estimated TAS angle is corrected by time drift compensation to generate a corrected estimated TAS angle. A set of electric power steering system control strategies is generated based on the corrected TAS angle estimation value. The set of electric power steering system control strategies includes a power assist torque adjustment scheme and a fault warning scheme.

2. The method for accurately estimating angle values ​​during TAS failure according to claim 1, characterized in that, The set of angle data collected from the TAS sensor and the motor position sensor includes: The TAS angle signal and the motor angle signal are acquired synchronously at a fixed sampling frequency, and the TAS angle signal and the motor angle signal are stored in a circular data buffer. The capacity of the circular data buffer ensures that the data storage time covers the maximum delay time of fault detection. The stored TAS angle signal and motor angle signal are timestamped to form a time-aligned angle data sequence.

3. The method for accurately estimating angle values ​​during TAS failure according to claim 2, characterized in that, The step of performing fault state analysis processing on the angle data set to generate TAS fault identifiers and fault detection times includes: The amplitude change and signal continuity of the TAS angle signal are continuously monitored. When the amplitude exceeds the preset effective range or the duration of signal loss reaches the fault judgment threshold, the fault indicator is activated. The system time when the fault identifier is activated is recorded as the fault detection time, and the fault detection time is associated with the timestamp of the angle data sequence; The step of associating the fault detection time with the timestamp of the angle data sequence includes: Obtain the system clock value at the time of fault detection and convert it into a time base reference that is the same as the timestamp of the angle data sequence; Search the circular data buffer for the timestamp closest to the fault detection time. If an exact matching timestamp exists, directly associate the angle data corresponding to that timestamp. If no exact matching timestamp exists, select the two closest timestamps before and after the fault detection time, and calculate the time offset between the fault detection time and the two timestamps. Based on the magnitude of the time offset, an association strategy is determined: when the time offset is less than or equal to the allowable error threshold, the angle data of the closer timestamp is associated; when the time offset is greater than the allowable error threshold, the time synchronization calibration process is triggered, the time base reference is realigned, and the association is performed again.

4. The method for accurately estimating angle values ​​during TAS failure according to claim 3, characterized in that, The process of determining the reference value based on the fault detection time and the cached angle data set to generate the TAS reference value and the motor angle reference value includes: Calculate the baseline reference time point based on the fault detection time and the maximum fault detection delay time; Retrieve the historical TAS angle value and historical motor angle value corresponding to the reference time point from the circular data buffer; If there is no direct data for the reference time point, an interpolation algorithm is used to calculate the TAS reference value and the motor angle reference value from the data of adjacent time points; The process of calculating the TAS reference value and motor angle reference value from data at adjacent time points using an interpolation algorithm includes: Identify two adjacent historical data points before and after the reference time point, and obtain their timestamps and corresponding angle values ​​respectively; Calculate the ratio of the time interval between the reference time point and its adjacent time points; Based on the principle of linear interpolation, the angle values ​​of adjacent time points are weighted using the time interval ratio to obtain the interpolated angle value of the reference time point. Interpolation calculations were performed on the TAS angle and the motor angle respectively to obtain the TAS reference value and the motor angle reference value; Verify the reasonableness of the interpolation calculation results: Check whether the interpolated angle value is within the interval formed by the adjacent angle values. If it exceeds the interval, use the secondary interpolation algorithm to recalculate.

5. The method for accurately estimating angle values ​​during TAS failure according to claim 4, characterized in that, The process of calling a predefined transmission ratio relationship to convert the difference between the current motor angle value and the motor angle reference value to generate an estimated value for the TAS angle includes: Calculate the real-time angle deviation between the current motor angle value and the motor angle reference value; The real-time angle deviation is converted into an equivalent change in TAS angle based on a predefined transmission ratio relationship. The equivalent change in the TAS angle is added to the TAS reference value to generate a preliminary estimate of the TAS angle. The step of converting the real-time angle deviation into an equivalent change in TAS angle based on a predefined transmission ratio relationship includes: Query the predefined transmission ratio relationship table to obtain the transmission ratio coefficient under the current working condition. The transmission ratio coefficient represents the proportional relationship between the change in motor angle and the change in TAS angle. Multiplying the real-time angle deviation by the transmission ratio coefficient yields the preliminary equivalent change in TAS angle. The directional consistency of the preliminary TAS angle equivalent change is verified: the sign of the equivalent change is adjusted according to the correspondence between the motor rotation direction and the TAS angle change direction. The equivalent change is corrected by applying a nonlinear compensation factor, which is dynamically adjusted based on the historical wear characteristics of the transmission system to compensate for the error caused by the transmission clearance.

6. The method for accurately estimating angle values ​​during TAS failure according to claim 5, characterized in that, The process of performing time drift compensation correction on the estimated TAS angle to generate a corrected estimated TAS angle includes: Estimate the system processing delay time from fault detection to the current moment, and calculate the angle drift compensation amount based on the system processing delay time; The angle drift compensation is applied to the initial TAS angle estimate, and phase correction processing is performed to generate the corrected TAS angle estimate.

7. The method for accurately estimating angle values ​​during TAS failure according to claim 6, characterized in that, The step of generating a set of electric power steering system control strategies based on the corrected TAS angle estimation value includes: Input the corrected TAS angle estimate into the assist torque lookup table to obtain the target assist torque value and generate an assist torque adjustment command; The TAS fault identifier is used to generate a graded fault warning signal, and the warning level is dynamically adjusted according to the fault duration.

8. The method for accurately estimating angle values ​​during TAS failure according to claim 7, characterized in that, The method further includes: After the TAS sensor resumes normal operation, the actual TAS angle value sequence is collected; The corrected TAS angle estimate during the fault period is compared with the actual TAS angle value sequence by deviation analysis to generate an estimation error statistic. Adjust the predefined transmission ratio relationship or the maximum fault detection delay time based on the estimated error statistics; The step of performing deviation analysis on the corrected TAS angle estimate and the actual TAS angle value sequence during the fault period to generate an estimation error statistic includes: Align the time series during the fault period by matching the corrected TAS angle estimation sequence with the actual TAS angle value sequence by timestamp; Calculate the angular deviation value at each matching time point to form a deviation value sequence; Statistical analysis is performed on the deviation value sequence to calculate the mean, standard deviation, and maximum absolute deviation of the deviation values; Generate a histogram of deviation distribution and calculate the kurtosis and skewness indices of the deviation values ​​to describe the shape of the deviation distribution; Based on the mean and standard deviation, the confidence interval of the estimation error is calculated, which serves as the core output of the estimation error statistic.

9. The method for accurately estimating angle values ​​during TAS failure according to claim 8, characterized in that, The step of adjusting the predefined transmission ratio relationship or the maximum fault detection delay time based on the estimated error statistics includes: Calculate the mean and standard deviation of the estimation error. If the mean exceeds the error tolerance threshold, then use the recursive least squares method to update the transmission ratio parameters. If the standard deviation indicates that the estimation is too volatile, increase the maximum delay time for fault detection to improve stability.

10. A system capable of accurately estimating angle values ​​in the event of a TAS failure, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for accurately estimating angle values ​​in the event of a TAS failure as described in any one of claims 1 to 9.

Images