Vehicle torque error evaluation method and vehicle
By determining and evaluating actual torque errors in vehicles and improving evaluation accuracy with Jacobella inspection, the problem of relying on experience and lack of evaluation systems in the prior art is solved, and higher torque control accuracy and vehicle safety are achieved.
Patent Information
- Application Number
- CN202411894320.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-15
AI Technical Summary
The existing vehicle torque safety monitoring methods rely on experience and subjective judgment, resulting in differences in the implementation and effect of monitoring strategies in different teams or projects, and lacking a suitable evaluation system, making objective and quantitative evaluation impossible, affecting safety and performance.
By determining multiple actual torque errors of the vehicle during the preset time period and performing Jacobella inspection, we will determine whether the inspection return value is a preset constant. If so, calculate the evaluation index to determine the performance evaluation result. If not, directly determine the performance evaluation result, and output the performance evaluation result and correction suggestions.
It improves the accuracy of torque control, effectively reduces torque errors, improves the safety and performance of the vehicle, and improves and optimizes monitoring strategies through an objective and quantitative evaluation system.
Smart Images
Figure CN119830444A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle technology, and in particular to a vehicle torque error evaluation method and a vehicle. Background Art
[0002] At present, vehicle torque safety is crucial to the normal driving of the vehicle and the safety of personnel. Vehicle torque safety requires real-time monitoring of the vehicle's unexpected acceleration and deceleration status, as well as monitoring whether the vehicle's actual torque output intention is consistent with expectations. However, existing monitoring methods may not be able to fully meet this requirement, especially in complex and changeable driving environments. The response speed and accuracy of the monitoring system may be challenged.
[0003] The existing torque safety monitoring strategy relies heavily on the experience and subjective judgment of developers, which may lead to differences in the implementation and effect of monitoring strategies in different teams or projects, which is not conducive to the improvement of overall safety. In addition, due to the lack of a suitable evaluation system, the evaluation of safety monitoring results often remains at the qualitative level and cannot be objectively and quantitatively evaluated. This is not conducive to the continuous improvement and optimization of monitoring strategies. Summary of the invention
[0004] In view of this, an object of the present application is to provide a vehicle torque error evaluation method and a vehicle to overcome at least one of the above-mentioned defects.
[0005] In a first aspect, an embodiment of the present application provides a vehicle torque error evaluation method, the method comprising: determining multiple actual torque errors of a vehicle within a preset time period, each actual torque error being the difference between the actual torque of the vehicle and an expected torque; performing a Jacobella test on the multiple actual torque errors to determine a test return value; judging whether the test return value of the Jacobella test is a preset constant; if the test return value of the Jacobella test is the preset constant, calculating evaluation indexes of multiple actual torque errors, and determining a performance evaluation result of the vehicle's actual torque error based on the evaluation index; if the test return value of the Jacobella test is not the preset constant, directly determining the performance evaluation result of the vehicle's actual torque error; and outputting the performance evaluation result and correction suggestions for reducing the vehicle's actual torque error.
[0006] In an optional embodiment of the present application, the actual torque error is determined in the following manner: obtaining vehicle operation data within a preset time period; inputting the vehicle operation data into a torque monitoring model to obtain multiple actual torque errors, wherein the torque monitoring model is used to monitor the torque output of the engine electronic control system through a hierarchical monitoring method.
[0007] In an optional embodiment of the present application, the method of performing a Jacobella test on multiple actual torque errors to determine a test return value includes: determining the multiple actual torque errors as multiple torque error samples; testing the multiple torque error samples using a Jacobella test function to determine the skewness and kurtosis of the multiple torque error samples; and determining the test return value of the Jacobella test based on the statistics of the skewness and the kurtosis.
[0008] In an optional embodiment of the present application, if the test return value of the Jacobella test is not the preset constant, the performance evaluation result of the actual torque error of the vehicle is directly determined, including: if the test return value of the Jacobella test is not the preset constant, it means that the multiple torque error samples do not obey the normal distribution, and the performance evaluation result is determined to be the first evaluation result.
[0009] In an optional embodiment of the present application, the performance evaluation result is determined in the following manner: if it is detected that the test return value of the Jacobella test is a preset constant, it means that the multiple torque error samples obey the normal distribution, and the average value of multiple actual torque errors is determined, and the average value is the evaluation index; it is judged whether the average value is equal to the preset constant; if the average value is equal to the preset constant, the performance evaluation result is determined to be the second evaluation result; if the average value is not equal to the preset constant, it is judged whether the average value is greater than the preset constant; if the average value is greater than the preset constant, the performance evaluation result is determined to be the third evaluation result; if the average value is less than the preset constant, the performance evaluation result is determined to be the fourth evaluation result.
[0010] In an optional embodiment of the present application, the output of the performance evaluation result and the correction suggestion for reducing the actual torque error of the vehicle include: if the performance evaluation result is the first evaluation result, it is determined that the multiple torque error samples do not conform to the normal distribution, and the correction suggestion is to check and process the outliers in the data, transform the data, and improve the skewness and kurtosis of the data; if the performance evaluation result is the second evaluation result, it is determined that the normal distribution of the multiple torque error samples is a symmetrical distribution type, and no correction suggestion is made; if the performance evaluation result is the third evaluation result, it is determined that the normal distribution of the multiple torque error samples is a left-biased distribution type, and the correction suggestion is to adjust the model threshold, add features, and improve the algorithm; if the performance evaluation result is the fourth evaluation result, it is determined that the normal distribution of the multiple torque error samples is a right-biased distribution type, and the correction suggestion is to adjust the model parameters, increase negative sample training, and introduce a classification algorithm.
[0011] In a second aspect, an embodiment of the present application further provides a vehicle, comprising the vehicle torque error evaluation method as described above.
[0012] In a third aspect, an embodiment of the present application further provides a vehicle torque error evaluation device, the device comprising: an actual torque error determination module, for determining multiple actual torque errors of a vehicle within a preset time period, each actual torque error being the difference between the actual torque of the vehicle and the expected torque; a test return value determination module, for subjecting multiple actual torque errors to a Jacobella test to determine a test return value; a judgment module, for judging whether the test return value of the Jacobella test is a preset constant; a performance evaluation result first determination module, for calculating evaluation indexes of multiple actual torque errors if the test return value of the Jacobella test is the preset constant, and determining a performance evaluation result of the vehicle's actual torque error based on the evaluation index; a performance evaluation result second determination module, for directly determining the performance evaluation result of the vehicle's actual torque error if the test return value of the Jacobella test is not the preset constant; and an output module, for outputting the performance evaluation result and correction suggestions for reducing the vehicle's actual torque error.
[0013] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the method described above are performed.
[0014] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described above are executed.
[0015] The vehicle torque error evaluation method and vehicle provided in the embodiment of the present application determine multiple actual torque errors of the vehicle within a preset time period, each actual torque error being the difference between the actual torque of the vehicle and the expected torque; subjecting the multiple actual torque errors to a Jacobella test to determine a test return value; determining whether the test return value of the Jacobella test is a preset constant; if the test return value of the Jacobella test is a preset constant, calculating evaluation indexes of the multiple actual torque errors, and determining a performance evaluation result of the actual torque error of the vehicle based on the evaluation index; if the test return value of the Jacobella test is not a preset constant, directly determining the performance evaluation result of the actual torque error of the vehicle; outputting the performance evaluation result and correction suggestions for reducing the actual torque error of the vehicle. Through the present application, the accuracy of torque control is improved based on the performance evaluation result to enhance the safety and performance of the vehicle.
[0016] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0018] Figure 1 A flow chart of a vehicle torque error evaluation method provided in an embodiment of the present application;
[0019] Figure 2 A flowchart for determining a test return value provided in an embodiment of the present application;
[0020] Figure 3 A flowchart for determining performance evaluation results provided in an embodiment of the present application;
[0021] Figure 4 A schematic diagram of a normal distribution of a second evaluation result provided in an embodiment of the present application;
[0022] Figure 5 A schematic diagram of a normal distribution of a third evaluation result provided in an embodiment of the present application;
[0023] Figure 6 A schematic diagram of a normal distribution of a fourth evaluation result provided in an embodiment of the present application;
[0024] Figure 7 A schematic diagram of the structure of a vehicle torque error evaluation device provided in an embodiment of the present application;
[0025] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0026] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work belongs to the scope of protection of the present application.
[0027] First, the application scenarios to which the present application is applicable are introduced. The present application can be applied in the field of vehicles.
[0028] Research has found that the existing torque safety monitoring strategy relies heavily on the experience and subjective judgment of developers, which may lead to differences in the implementation and effectiveness of the monitoring strategy in different teams or projects, which is not conducive to improving overall safety. In addition, due to the lack of a suitable evaluation system, the evaluation of safety monitoring results often remains at a qualitative level and cannot be objectively and quantitatively evaluated, which is not conducive to the continuous improvement and optimization of the monitoring strategy. In addition, if the driver does not turn the steering wheel, the vehicle will not be able to drive in a straight line due to torque execution errors, affecting the driving experience.
[0029] Based on this, an embodiment of the present application provides a vehicle torque error evaluation method and a vehicle, which determines multiple actual torque errors of the vehicle within a preset time period, each actual torque error being the difference between the actual torque of the vehicle and the expected torque; performs a Jacobella test on the multiple actual torque errors to determine a test return value; determines whether the test return value of the Jacobella test is a preset constant; if the test return value of the Jacobella test is a preset constant, calculates evaluation indexes of multiple actual torque errors, and determines a performance evaluation result of the actual torque error of the vehicle based on the evaluation index; if the test return value of the Jacobella test is not a preset constant, directly determines the performance evaluation result of the actual torque error of the vehicle; outputs the performance evaluation result and correction suggestions for reducing the actual torque error of the vehicle. Through the present application, the accuracy of torque control is improved based on the performance evaluation result, the torque error is effectively reduced, and the safety and performance of the vehicle are improved.
[0030] See also Figure 1 , Figure 1 This is a flow chart of the vehicle torque error evaluation method provided in the embodiment of the present application. Figure 1 As shown in , the vehicle torque error evaluation method provided by the embodiment of the present application includes:
[0031] S101. Determine a plurality of actual torque errors of a vehicle within a preset time period, each actual torque error being a difference between an actual torque of the vehicle and an expected torque.
[0032] The actual torque error is determined as follows:
[0033] Obtain vehicle operation data within a preset time period;
[0034] First, the data source needs to be determined. Vehicle operation data may come from on-board sensors, diagnostic systems, or remote monitoring systems. These data include vehicle speed, accelerator pedal, torque, voltage signal, engine speed, fuel injection amount, intake volume, exhaust temperature, etc., as well as directly related torque output data.
[0035] Identify a preset time period, such as data from the past minute, hour, or a specific time period. This helps ensure data coherence and consistency, and that the data is stored in an appropriate format for subsequent processing and analysis. Typically, this data is stored in a time series format with timestamps and corresponding sensor readings.
[0036] The vehicle operation data is input into the torque monitoring model to obtain a plurality of actual torque errors, wherein the torque monitoring model is used to monitor the torque output of the engine electronic control system through a hierarchical monitoring method.
[0037] Preferably, the torque monitoring model here can be a three-layer torque monitoring model.
[0038] The three-layer torque monitoring model uses a hierarchical monitoring method to conduct real-time and comprehensive monitoring of the torque output of the engine electronic control system. The model divides the monitoring tasks into three levels, each of which is responsible for different monitoring content and tasks, jointly ensuring the safety of the system.
[0039] Level 1: Vehicle controller control function.
[0040] Receives operating data from various vehicle sensors, such as accelerator pedal position, engine speed, vehicle speed, etc. Based on the received data, the vehicle controller implements basic control functions, such as calculating required torque, adjusting engine operating status, etc., and sends the processed control instructions to the engine electronic control system to adjust the torque output.
[0041] Level 2: Torque monitoring function.
[0042] Receive the control instructions output by the first layer and the actual torque output data of the engine electronic control system, monitor and compare the actual torque output data, and compare it with the required torque calculated by the first layer. When there is an error between the actual torque output and the required torque, the second layer will trigger the corresponding monitoring mechanism. When the actual torque output exceeds the safety range or there is an abnormality, the second layer will output a torque error signal and may trigger an alarm or protective measures.
[0043] Level 3: System monitoring and troubleshooting.
[0044] The torque error signal output by the second layer and the status information of other related vehicle systems are received, and the torque error signal is further analyzed, and combined with the status information of other vehicle systems, it is determined whether there is a system failure or safety hazard. When a fault is confirmed, the third layer will start the fault handling mechanism. According to the type and severity of the fault, the third layer will output the corresponding fault handling instructions, such as limiting torque output, shutting down the engine, etc., to ensure the safety of the vehicle and personnel.
[0045] The three-layer torque monitoring model realizes comprehensive and real-time monitoring of the torque output of the engine electronic control system through layered monitoring. This model not only improves the stability and reliability of the system, but also reduces the safety hazards caused by abnormal torque. At the same time, this model also provides a useful reference for subsequent safety monitoring strategies.
[0046] In an optional embodiment, the three-layer torque monitoring model can be customized and optimized according to the characteristics of the specific vehicle and engine electronic control system to meet different safety monitoring requirements.
[0047] S102, performing Jacobella test on multiple actual torque errors to determine the test return value.
[0048] Here, the Jarque-Bera Test is a statistical method used to test whether the sample data conforms to the normal distribution. It calculates a statistic based on the skewness and kurtosis of the sample data, and then compares it with the critical value of the statistic to determine whether the sample data deviates from the normal distribution.
[0049] See also Figure 2 , Figure 2 This is a flow chart for determining the test return value provided by the embodiment of the present application. Figure 2 As shown in , the vehicle torque error evaluation method provided by the embodiment of the present application includes:
[0050] S201, determining a plurality of actual torque errors as a plurality of torque error samples;
[0051] First, the collected multiple actual torque errors (i.e., the difference between the actual torque of the vehicle and the expected torque) are sorted into multiple torque error samples. These samples can be divided based on time intervals, mileage or other relevant factors.
[0052] Make sure that each sample contains a sufficient number of data points to allow for reliable statistical analysis.
[0053] The division of samples should be based on reasonable grounds to ensure independence and representativeness among samples.
[0054] S202, using Jacobella test function to test multiple torque error samples to determine the skewness and kurtosis of the multiple torque error samples;
[0055] Specifically, each torque error sample is tested using the Jacobella test function, which calculates the skewness and kurtosis of the sample and assesses the normality of the sample data based on these two statistics.
[0056] Skewness is a statistic that measures the degree of asymmetry in data distribution. Positive skewness means that the data is skewed to the right, and negative skewness means that the data is skewed to the left. Kurtosis is a statistic that measures the sharpness of data distribution. High kurtosis means that the data distribution is sharper than the normal distribution, and low kurtosis means that the data distribution is flatter than the normal distribution.
[0057] The Jacobella test function generates a test statistic based on the calculated values of skewness and kurtosis.
[0058] S203. Determine the test return value of the Jacobella test according to the statistics of skewness and kurtosis.
[0059] Here, the test statistic calculated by the Jacobella test function is compared with the critical value at the preset significance level. According to the comparison result, the test return value h of the Jacobella test is determined, that is, the test return value can determine whether the sample data obeys the normal distribution.
[0060] S103: Determine whether the test return value of the Jacobella test is a preset constant.
[0061] In this step, after the Jacobella test is completed, its return value is first checked to determine whether the statistic of the Jacobella test is less than or equal to the critical value under the preset significance level.
[0062] If the returned value (i.e., statistic) is less than or equal to the critical value, it is generally believed that the sample data may follow a normal distribution; if it is greater than the critical value, the normality hypothesis is rejected. Here, the critical value is the preset constant 0.
[0063] The Jacobella test is suitable for large sample data. For small sample data, its accuracy may be affected. In addition, before conducting the Jacobella test, the sample data should be preprocessed as necessary, such as removing outliers and smoothing, to improve the accuracy of the test.
[0064] S104: If the test return value of the Jacobella test is a preset constant, multiple evaluation indexes of actual torque errors are calculated, and a performance evaluation result of the vehicle actual torque error is determined according to the evaluation indexes.
[0065] If the statistic of the Jacobella test indicates that the sample data may follow a normal distribution (that is, the statistic is equal to the preset constant), the next step of calculation is performed.
[0066] Calculate the evaluation index of multiple actual torque errors. This can be one or more statistics, such as mean, standard deviation, coefficient of variation, etc., which are used to quantify the dispersion and consistency of the torque errors. Here, the mean is taken as an example.
[0067] Based on the evaluation index, a performance evaluation result of the actual torque error of the vehicle is determined. This may involve comparison with a preset standard or threshold to determine whether the torque error is within an acceptable range.
[0068] S105: If the test return value of the Jacobella test is not a preset constant, directly determine the performance evaluation result of the actual torque error of the vehicle.
[0069] If the Jacobella test statistic indicates that the sample data does not follow a normal distribution (i.e., the statistic is not a preset constant), then the performance evaluation is performed directly.
[0070] In this case, since the data does not conform to the normal distribution, traditional statistical methods based on normal distribution may not be applicable. Nonparametric statistical methods or other evaluation methods suitable for non-normal data can be considered to determine the performance of the vehicle's actual torque error.
[0071] The evaluation results may be more dependent on the specific distribution and characteristics of the data, as well as the actual operating conditions and environment of the vehicle.
[0072] If the sample does not follow a normal distribution: the characteristics of the error data should be analyzed in depth to find out the reasons for the non-normal distribution. Possible reasons include sensor failure, control system instability, external interference, etc. In response to these reasons, corresponding measures can be taken to improve and optimize to improve the performance of the safety monitoring system.
[0073] When conducting the Jacob-Berra test, it is necessary to ensure that the sample data has sufficient quantity and representativeness to avoid bias in the test results due to insufficient or weak sample representation. The significance level of the Jacob-Berra test can be adjusted according to actual conditions. Generally, the lower the significance level, the stricter the requirement for normality of the sample data.
[0074] S106: Output the performance evaluation result and the correction suggestion for reducing the actual torque error of the vehicle.
[0075] The performance evaluation results are determined by:
[0076] If the test return value of the Jacobella test is not a preset constant, it means that the multiple torque error samples do not obey the normal distribution, and the performance evaluation result is determined to be the first evaluation result.
[0077] If it is not a preset constant, the multiple torque error samples do not obey the normal distribution, and the performance evaluation result is directly determined to be the first evaluation result.
[0078] If it is a preset constant, go to the next step.
[0079] Here, the preset constant is 0. If the test return value h=0, the sample obeys the normal distribution and the distribution type needs to be determined. If the test return value h=1, the sample does not obey the normal distribution, the performance of the security monitoring system is poor, and the test is terminated.
[0080] See also Figure 3 , Figure 3 This is a flow chart for determining performance evaluation results provided by an embodiment of the present application. Figure 3 As shown in , the vehicle torque error evaluation method provided by the embodiment of the present application includes:
[0081] S301, if it is detected that the test return value of the Jacobella test is a preset constant, it means that the multiple torque error samples obey the normal distribution, and the average value of the multiple actual torque errors is determined, and the average value is the evaluation index;
[0082] After confirming that multiple torque error samples follow a normal distribution, calculate the average of these samples.
[0083] This average value is used as the evaluation index for subsequent performance evaluation.
[0084] S302, determining whether the average value is equal to a preset constant;
[0085] S303: If the average value is equal to the preset constant, determine that the performance evaluation result is a second evaluation result;
[0086] The average value μ is equal to 0, which is identified as a symmetrical distribution type. This type has a good trade-off between the risk of false triggering of security monitoring and efficient monitoring of security events. The evaluation result of the security monitoring model is excellent. If it is equal to the preset constant, the performance evaluation result is determined to be the second evaluation result. If it is not equal to the preset constant, the next step of judgment is entered.
[0087] S304, if the average value is not equal to the preset constant, determine whether the average value is greater than the preset constant;
[0088] S305: If the average value is greater than a preset constant, determining the performance evaluation result as a third evaluation result;
[0089] The average value μ is greater than 0, which is identified as a left-biased distribution type. This type has a lower risk of false triggering of security monitoring, but a high probability of missing security events. It is necessary to continue to optimize the security monitoring model.
[0090] If it is greater than a preset constant, it may indicate that the torque error is too large, and the performance evaluation result is determined to be the third evaluation result.
[0091] If it is not greater than the preset constant (ie, less than or equal to), proceed to the next step of judgment.
[0092] S306: If the average value is less than a preset constant, determine that the performance evaluation result is a fourth evaluation result.
[0093] The average value μ is less than 0, which is identified as a right-biased distribution type. This type has a high risk of false triggering of safety monitoring and is very likely to cause after-sales complaints. The safety monitoring model needs to be optimized urgently. If it is less than the preset constant, it may indicate that the torque error is too small, and the performance evaluation result is determined to be the fourth evaluation result.
[0094] Specifically, the preset constants (including the constants used to determine the return value of the Jacobella test and the constants used to determine the average value) should be determined according to the actual application background and requirements. These constants may represent performance standards, acceptable ranges or thresholds. Here, the preset constant is 0.
[0095] Different evaluation results correspond to different performance states or problems. For example, the first evaluation result may indicate that the data is abnormal or does not meet the normal distribution assumption; the second evaluation result may indicate that the performance meets the standard; the third evaluation result may indicate that the performance exceeds the standard or the torque error is too large; the fourth evaluation result may indicate that the performance is insufficient or the torque error is too small.
[0096] Although the above process provides a performance evaluation method based on Jacobella test and mean value judgment, it may be necessary to adjust and optimize it according to specific circumstances in actual applications. For example, it is possible to consider introducing other statistics or evaluation indicators to more comprehensively reflect the performance of the actual torque error of the vehicle.
[0097] Before conducting a performance evaluation, you should ensure that the collected torque error data is accurate and complete. The quality of the data directly affects the accuracy and reliability of the evaluation results.
[0098] Specifically, the correction suggestions corresponding to each assessment result are as follows:
[0099] If the performance evaluation result is the first evaluation result, it is determined that the multiple torque error samples do not conform to the normal distribution, and the correction suggestion is to check and process the outliers in the data, transform the data, and improve the skewness and kurtosis of the data;
[0100] Here, the torque error samples do not conform to the normal distribution. The outliers in the data are checked and processed. The distribution of the data is checked by using visualization tools (such as box plots and scatter plots) to identify outliers. For outliers, you can consider deleting, replacing (such as using the mean, median, etc.) or smoothing (such as using the moving average method).
[0101] Try to use mathematical methods such as logarithmic transformation and Box-Cox transformation to improve the skewness and kurtosis of the data. After the transformation, check the normality of the data again to ensure the transformation effect, analyze whether there are systematic deviations or random errors in the data collection process, optimize the data collection method, ensure the accuracy and consistency of data collection, so as to reduce the generation of outliers.
[0102] If the performance evaluation result is the second evaluation result, it is determined that the normal distribution of the multiple torque error samples is a symmetrical distribution type, and no correction is recommended;
[0103] See also Figure 4 , Figure 4 A schematic diagram of the normal distribution of the second evaluation result provided in an embodiment of the present application.
[0104] like Figure 4 As shown in the figure, the actual torque error is the parameter on the y-axis, which represents the difference between the actual torque and the expected torque or the standard torque at a certain time point or a certain frequency. In the figure, the actual torque error is represented by the blue curve, showing its changing trend over time (or frequency).
[0105] Frequency is a parameter on the x-axis, which indicates the number of times a data point appears or the number of time points. In the figure, frequency is used to indicate the distribution of actual torque error at different time points or under different conditions.
[0106] T LO (lower limit of permissible torque) and T UP (Allowable torque upper limit) These two parameters represent the minimum and maximum values of the allowable torque respectively. In the figure, T LO and T UP They are used as two horizontal axes to define the allowable range of actual torque error.
[0107] When the actual torque error is at T LO and T UP If the value is between 0 and 1, it indicates that the torque is within the allowable range; otherwise, it indicates that the torque exceeds the allowable range.
[0108] The parameter torque error T refers to a general representation of the actual torque error or a specific torque error value.
[0109] Symmetrical distribution is a term used to describe the type of data distribution. Although the data points or curves of the symmetrical distribution are not directly marked in the figure, it can be inferred whether it is symmetrical based on the shape and distribution of the actual torque error curve. If the actual torque error curve is in the T LO and T UPIf the distribution shows a symmetrical shape (such as a bell curve) and the peak of the curve is on the dotted line of 0, it can be considered to be a symmetrical distribution type.
[0110] Specifically, the normal distribution of torque error samples is a symmetrical distribution type. Maintain the status quo and continue to monitor. Since the data has shown a symmetrical distribution, it shows that the current data collection and processing methods are effective. Continue to maintain the existing data collection and processing process, and conduct performance evaluations regularly to ensure the stability and reliability of the data.
[0111] Although there is no need for immediate correction, you can further analyze the detailed characteristics of the data, explore whether there are other potential optimization points, and consider introducing new features or algorithms to further improve the accuracy and robustness of the model.
[0112] If the performance evaluation result is the third evaluation result, it is determined that the normal distribution of the multiple torque error samples is a left-biased distribution type, and the correction suggestion is to adjust the model threshold, add features, and improve the algorithm to increase the sensitivity to safety events;
[0113] See also Figure 5 , Figure 5 A schematic diagram of the normal distribution of the third evaluation result provided in an embodiment of the present application.
[0114] like Figure 5 As shown in the figure, the actual torque error, frequency, T LO (lower limit of permissible torque) and T UP (upper limit of permissible torque) and Figure 4 Again, I won’t go into details here.
[0115] When the actual torque error is at T LO and T UP When it is between , it means the torque is within the allowable range. Figure 5 In the figure, the dashed line with zero error in the middle is located to the left of the peak of the curve, indicating that the desired torque is greater than the actual torque.
[0116] Here, the normal distribution of the torque error samples is a left-biased distribution. The model threshold is adjusted. According to the distribution characteristics of the left bias, the decision threshold of the model is appropriately adjusted to improve the ability to identify safety events. The optimal threshold setting is determined through methods such as cross-validation.
[0117] Introduce other features related to torque error, such as temperature, humidity, pressure, etc., to improve the expressiveness of the model. Use feature selection methods to screen out features that contribute significantly to model performance. Try to use more advanced machine learning algorithms, such as deep learning, ensemble learning, etc., to improve the accuracy and generalization ability of the model. Tune the algorithm parameters to find the best parameter combination.
[0118] If the performance evaluation result is the fourth evaluation result, it is determined that the normal distribution of multiple torque error samples is a left-biased distribution type, and the correction suggestions are to adjust the model parameters, increase negative sample training, and introduce a classification algorithm to reduce the false trigger rate.
[0119] See also Figure 6 , Figure 6 A schematic diagram of the normal distribution of the fourth evaluation result provided in an embodiment of the present application.
[0120] like Figure 6 As shown in the figure, the actual torque error, frequency, T LO (lower limit of permissible torque) and T UP (upper limit of permissible torque) and Figure 4 Again, I won’t go into details here.
[0121] When the actual torque error is at T LO and T UP When it is between , it means the torque is within the allowable range. Figure 6 In the figure, the dotted line of the middle zero error is located on the right side of the peak of the curve, indicating that the expected torque is smaller than the actual torque, and the actual torque is larger than the expected torque, which is dangerous.
[0122] Here, when the performance evaluation result is the fourth evaluation result, and it is determined that the normal distribution of the plurality of torque error samples is a right-biased distribution type, this means that most of the torque error values in the data are greater than the mean, and the distribution is tilted toward larger values. This right-biased distribution type may indicate that the model is biased or deficient in some aspects, resulting in inaccurate prediction or identification of larger torque errors.
[0123] Review and adjust the model's parameter settings, especially those related to torque error prediction or identification, to optimize the model's performance by adjusting the parameters to better fit the right-biased data distribution.
[0124] Add more negative samples (i.e., samples with smaller torque errors) to the training data set to improve the model's ability to identify smaller torque errors. By increasing the diversity of negative samples, the model can better learn the distribution characteristics in the data and reduce the false trigger rate.
[0125] Consider introducing a classification algorithm into the model to more accurately distinguish torque errors of different sizes. The classification algorithm can help the model better identify the categories of torque errors (such as small errors, medium errors, large errors, etc.) and make more accurate predictions or decisions based on this.
[0126] The raw data is preprocessed, such as standardization and normalization, to eliminate the dimensional differences between different features, perform feature engineering, extract useful features related to torque error, and remove redundant or irrelevant features. Through data preprocessing and feature engineering, the generalization ability and accuracy of the model can be improved.
[0127] Use methods such as cross-validation to evaluate the performance of the model, and tune the model based on the evaluation results. By continuously iterating and optimizing the model, the false trigger rate can be gradually reduced and the accuracy of the model can be improved.
[0128] In summary, for the correction suggestions of different evaluation results, appropriate measures should be taken according to the specific situation to optimize the performance and accuracy of the model. At the same time, it is also necessary to continuously monitor the distribution of data and the performance of the model so as to adjust the strategy and method in time.
[0129] The vehicle torque error evaluation method and vehicle provided in the embodiment of the present application determine multiple actual torque errors of the vehicle within a preset time period, each actual torque error being the difference between the actual torque of the vehicle and the expected torque; perform Jacobella test on multiple actual torque errors to determine the test return value; determine whether the test return value of the Jacobella test is a preset constant; if the test return value of the Jacobella test is a preset constant, calculate the evaluation index of multiple actual torque errors, and determine the performance evaluation result of the vehicle's actual torque error according to the evaluation index; if the test return value of the Jacobella test is not a preset constant, directly determine the performance evaluation result of the vehicle's actual torque error; output the performance evaluation result and the correction suggestion for reducing the vehicle's actual torque error. Through the present application, the accuracy of torque control is improved according to the performance evaluation result, the torque error is effectively reduced, and the safety and performance of the vehicle are improved.
[0130] This application can be applied to all new energy vehicles. The customer service relies on the development experience of personnel and the subjective evaluation method to deal with the defect of inaccurate adjustment of the actual torque error. A suitable evaluation system has been established to objectively quantify the safety monitoring results.
[0131] Based on the same inventive concept, a vehicle torque error evaluation device corresponding to the vehicle torque error evaluation method is also provided in the embodiment of the present application. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the above-mentioned vehicle torque error evaluation method in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0132] See also Figure 7 , Figure 7 This is a schematic diagram of the structure of the vehicle torque error evaluation device provided in the embodiment of the present application. Figure 7 As shown in FIG. , the vehicle torque error evaluation device 700 includes:
[0133] An actual torque error determination module 701 is used to determine a plurality of actual torque errors of the vehicle within a preset time period, each actual torque error being a difference between an actual torque of the vehicle and an expected torque;
[0134] A test return value determination module 702 is used to perform Jacobella test on multiple actual torque errors to determine a test return value;
[0135] A judging module 703 is used to judge whether the test return value of the Jacobella test is a preset constant;
[0136] A first performance evaluation result determination module 704 is configured to calculate a plurality of evaluation indexes of actual torque errors if the test return value of the Jacobella test is the preset constant, and determine a performance evaluation result of the actual torque error of the vehicle according to the evaluation indexes;
[0137] The second performance evaluation result determination module 705 is used to directly determine the performance evaluation result of the actual torque error of the vehicle if the test return value of the Jacobella test is not the preset constant;
[0138] The output module 706 is used to output the performance evaluation result and correction suggestions for reducing the actual torque error of the vehicle.
[0139] See also Figure 8 , Figure 8 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 8 As shown in , the electronic device 800 includes a processor 810 , a memory 820 and a bus 830 .
[0140] The memory 820 stores machine-readable instructions executable by the processor 810. When the electronic device 800 is running, the processor 810 communicates with the memory 820 via the bus 830. When the machine-readable instructions are executed by the processor 810, the above-mentioned Figure 1 The specific implementation of the steps of the vehicle torque error evaluation method in the method embodiment shown can be found in the method embodiment, which will not be repeated here.
[0141] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The specific implementation of the steps of the vehicle torque error evaluation method in the method embodiment shown can be found in the method embodiment, which will not be repeated here.
[0142] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0143] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.
[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0145] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0146] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, or in other words, the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0147] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
Claims
1. A vehicle torque error evaluation method, characterized in that: include: Determining a plurality of actual torque errors of the vehicle within a preset time period, each actual torque error being a difference between an actual torque of the vehicle and an expected torque; Perform Jacobella test on multiple actual torque errors to determine the test return value; Determine whether the return value of the Jacobella test is a preset constant; If the test return value of the Jacobella test is the preset constant, calculating a plurality of evaluation indexes of the actual torque errors, and determining a performance evaluation result of the vehicle actual torque error according to the evaluation indexes; If the test return value of the Jacobella test is not the preset constant, directly determining the performance evaluation result of the actual torque error of the vehicle; The performance evaluation result and correction suggestions for reducing the actual torque error of the vehicle are output.
2. The method according to claim 1, characterized in that The actual torque error is determined by: Obtain vehicle operation data within a preset time period; The vehicle operation data is input into a torque monitoring model to obtain a plurality of actual torque errors, wherein the torque monitoring model is used to monitor the torque output of the engine electronic control system through a hierarchical monitoring method.
3. The method according to claim 2, characterized in that The Jacobella test is performed on a plurality of actual torque errors to determine the test return value, including: determining a plurality of actual torque errors as a plurality of torque error samples; Jacobella test function is used to test multiple torque error samples to determine the skewness and kurtosis of multiple torque error samples; A test return value of the Jacobella test is determined based on the statistics of the skewness and the kurtosis.
4. The method according to claim 3, characterized in that If the test return value of the Jacobella test is not the preset constant, directly determining the performance evaluation result of the actual torque error of the vehicle includes: If the test return value of the Jacobella test is not a preset constant, it means that the multiple torque error samples do not obey the normal distribution, and the performance evaluation result is determined to be the first evaluation result.
5. The method according to claim 4, characterized in that The performance evaluation results are determined by: If it is detected that the test return value of the Jacobella test is a preset constant, it means that the multiple torque error samples obey the normal distribution, and the average value of the multiple actual torque errors is determined, and the average value is the evaluation index; Determining whether the average value is equal to a preset constant; If the average value is equal to the preset constant, determining the performance evaluation result as the second evaluation result; If the average value is not equal to the preset constant, determining whether the average value is greater than the preset constant; If the average value is greater than the preset constant, determining the performance evaluation result to be a third evaluation result; If the average value is less than the preset constant, the performance evaluation result is determined to be a fourth evaluation result.
6. The method according to claim 5, characterized in that The outputting of the performance evaluation result and the correction suggestion for reducing the actual torque error of the vehicle includes: If the performance evaluation result is the first evaluation result, it is determined that the multiple torque error samples do not conform to the normal distribution, and the correction suggestion is to check and process the outliers in the data, transform the data, and improve the skewness and kurtosis of the data; If the performance evaluation result is the second evaluation result, it is determined that the normal distribution of the plurality of torque error samples is a symmetrical distribution type, and no correction is recommended; If the performance evaluation result is the third evaluation result, it is determined that the normal distribution of the plurality of torque error samples is a left-biased distribution type, and the correction suggestion is to adjust the model threshold, add features, and improve the algorithm; If the performance evaluation result is the fourth evaluation result, it is determined that the normal distribution of the multiple torque error samples is a right-biased distribution type, and the correction suggestions are to adjust the model parameters, increase negative sample training, and introduce a classification algorithm.
7. A vehicle torque error evaluation device, characterized in that: include: an actual torque error determination module, used to determine a plurality of actual torque errors of the vehicle within a preset time period, each actual torque error being a difference between an actual torque of the vehicle and an expected torque; A test return value determination module is used to perform Jacobella tests on multiple actual torque errors to determine a test return value; A judgment module, used for judging whether the test return value of Jacobella test is a preset constant; a first performance evaluation result determination module, configured to calculate a plurality of evaluation indexes of actual torque errors if the test return value of the Jacobella test is the preset constant, and determine a performance evaluation result of the actual torque error of the vehicle according to the evaluation indexes; a second performance evaluation result determination module, configured to directly determine a performance evaluation result of an actual torque error of the vehicle if the test return value of the Jacobella test is not the preset constant; The output module is used to output the performance evaluation result and correction suggestions for reducing the actual torque error of the vehicle.
8. A vehicle, characterized in that: The vehicle comprises a vehicle torque error evaluation method as described in any one of claims 1-6.
9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of any method as claimed in claims 1 to 6.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are executed.