Parameter calibration method, electronic equipment and storage medium
The parameters that have a great impact on vehicle performance indicators were screened through the experimental design method, and the target optimization algorithm was used for calibration, which solved the problem of low parameter calibration efficiency in the existing technology and achieved efficient parameter calibration.
Patent Information
- Application Number
- CN202510012393.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-03
- Publication Date
- 2025-04-29
AI Technical Summary
In the prior art, vehicle parameter calibration efficiency is low because a large number of candidate parameters need to be manually selected and calibrated, including parameters that have little impact on vehicle performance indicators.
The actual parameters that have a significant impact on the vehicle performance indicators are selected from the multiple candidate parameters to be calibrated by the experimental design method, and these parameters are calibrated using the target optimization algorithm to obtain calibration results that meet the expected value.
The number of parameters that need to be calibrated is reduced, the parameter calibration efficiency is improved, and the vehicle performance indicators reach the expected value.
Smart Images
Figure CN119901500A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of vehicle technology, and in particular to a parameter calibration method, an electronic device and a storage medium. Background Art
[0002] In the related art, vehicle parameters are often selected based on manual experience for calibration. However, the above method may result in spending more time to calibrate some parameters that have little impact on vehicle performance indicators, resulting in low calibration efficiency. Summary of the invention
[0003] The embodiments of the present application provide a parameter calibration method, an electronic device, and a storage medium, which can calibrate only the parameters that have a significant impact on vehicle performance indicators, thereby improving parameter calibration efficiency.
[0004] In a first aspect, an embodiment of the present application provides a parameter calibration method, the method comprising: Acquire multiple candidate parameters to be calibrated and vehicle performance indicators; Determining an actual parameter affecting the vehicle performance indicator from the plurality of candidate parameters; The actual parameters are calibrated to obtain calibration results that meet the expected values of the vehicle performance indicators.
[0005] In an embodiment of the present application, actual parameters that have a significant impact on vehicle performance parameters are selected from multiple candidate parameters to be calibrated, and then the actual parameters are calibrated, thereby reducing the number of parameters that need to be calibrated and improving parameter calibration efficiency.
[0006] Optionally, determining an actual parameter affecting the vehicle performance index from the plurality of candidate parameters includes: The actual parameters affecting the vehicle performance index are determined from the multiple candidate parameters according to the experimental design method.
[0007] In the embodiment of the present application, the actual parameters that have a significant impact on the vehicle performance indicators can be conveniently and accurately selected from multiple candidate parameters through the experimental design method.
[0008] Optionally, determining the actual parameter affecting the vehicle performance index from the plurality of candidate parameters according to an experimental design method includes: Determine the initial value range and initial step size of each candidate parameter; Determining a plurality of candidate values corresponding to the candidate parameters based on the initial value range and the initial step size; Inputting the plurality of candidate values into a precompiled vehicle simulation model in sequence to obtain a maximum value and a minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the influence degree threshold, the corresponding candidate parameter is used as the actual parameter to be calibrated; If the difference between the maximum value and the minimum value is not greater than the influence degree threshold, the corresponding candidate parameter is rejected as the actual parameter to be calibrated.
[0009] In the embodiment of the present application, for each candidate parameter, there is an initial value range and an initial step length, which means that each candidate parameter has multiple candidate values within its corresponding initial value range, then these multiple candidate values can be sequentially input into the precompiled vehicle simulation model, so that the maximum and minimum values of the vehicle performance index can be obtained. If the difference between the maximum value and the minimum value is greater than the impact degree threshold, it indicates that the current candidate parameter value changes, and the impact on the vehicle performance index is large, that is, the current candidate parameter has a significant impact on the vehicle performance index, then the current candidate parameter is used as the actual parameter that needs to be calibrated; on the contrary, if the difference between the maximum value and the minimum value is less than the impact degree threshold, it indicates that the current candidate parameter value changes, and the impact on the vehicle performance index is small, that is, the current candidate parameter does not have a significant impact on the vehicle performance index, then the current candidate parameter is screened out and is not used as the actual parameter that needs to be calibrated.
[0010] Optionally, calibrating the actual parameter to obtain a calibration result that meets the expected value of the vehicle performance indicator includes: Determine the optimal value range and the optimal step length of the actual parameters according to the experimental design method; According to the optimal value range and the optimal step length of the actual parameters, the actual parameters are optimized using a target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index.
[0011] In an embodiment of the present application, before the actual parameters are formally calibrated, the optimal value range and the optimal step size of the actual parameters can be determined through the experimental design method, thereby saving the time for optimizing the actual parameters during the formal calibration process, thereby improving the efficiency of the calibration of the actual parameters.
[0012] Optionally, determining the optimal value range of the actual parameter according to an experimental design method includes: Determining an initial value range of the actual parameter; Inputting the upper limit value and the lower limit value of the initial value range into the vehicle simulation model in sequence, and obtaining the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value; If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is greater than the upper limit of the range threshold, the initial value range is narrowed; If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is less than the lower limit of the range threshold, the initial value range is increased until the optimal value range of the actual parameter is determined.
[0013] In the embodiment of the present application, for the actual parameters that have a significant impact on the vehicle performance index, there is a corresponding initial value range. If the upper limit value and the lower limit value of the initial value range are respectively input into the pre-compiled vehicle simulation model, the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value can be obtained. If the difference between the above two vehicle performance indicators is large, it indicates that the current initial value range has a large impact on the vehicle performance index, and the initial value range can be reduced at this time; conversely, if the difference between the above two vehicle performance indicators is small, it indicates that the current initial value range has a small impact on the vehicle performance index, and the initial value range can be increased at this time.
[0014] Optionally, determining the optimal step length of the actual parameter according to an experimental design method includes: Determining an initial step size of the actual parameter; Determining a plurality of candidate values of the actual parameter within the optimal value range of the actual parameter and the initial step size; Inputting the plurality of candidate values into the vehicle simulation model in sequence to obtain the maximum value and the minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the upper limit of the step length threshold, the initial step length is reduced; If the difference between the maximum value and the minimum value is smaller than the lower limit of the step length threshold, the initial step length is increased until the optimal step length of the actual parameter is determined.
[0015] In the embodiment of the present application, for the actual parameters that have a significant impact on the vehicle performance index, there is a corresponding initial step length, so multiple candidate values can be determined based on the initial step length from the optimal value interval of the actual parameters. These multiple candidate values are input into the vehicle simulation model in sequence, and the maximum and minimum values of the vehicle performance index within the optimal value interval can be obtained. If the difference between the maximum and minimum values is large, it indicates that the current initial step length has a greater impact on the vehicle performance index, and the initial step length can be reduced at this time; conversely, if the difference between the maximum and minimum values is small, it indicates that the current initial step length has a smaller impact on the vehicle performance, and the initial step length can be increased at this time until the optimal step length is determined within the optimal value interval.
[0016] Optionally, according to the optimal value range and the optimal step length of the actual parameter, the actual parameter is optimized using the target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance indicator, including: Inputting the actual parameters into a precompiled vehicle simulation model to obtain actual values of the vehicle performance indicators; Constructing an objective function of expected value and actual value of the vehicle performance index; With the purpose of minimizing the objective function, the target optimization algorithm is used to perform optimization according to the optimal value range and the optimal step size of the actual parameter to obtain the calibration result of the actual parameter.
[0017] In an embodiment of the present application, an objective function is constructed by using the expected value and actual value of the vehicle performance index. Then, in the actual parameter calibration process, the objective is to minimize the objective function. An optimization is performed based on the optimal value range and the optimal step size of the actual parameters, thereby accurately determining the calibration result of the actual parameters.
[0018] Optionally, the method further includes: Configuring the calibration results of the actual parameters to the vehicle simulation model; Determining an actual noise from a plurality of candidate noises that can be used to test the robustness of the calibration result; Generating the actual noise, and inputting the actual noise into the vehicle simulation model configured with the calibration result, to obtain a noise environment output value of the vehicle performance index; If the noise environment output value does not meet the robustness requirement, the actual parameter is recalibrated.
[0019] In the embodiment of the present application, after the actual parameter calibration is completed, the calibration result of the actual parameter can be configured into the vehicle simulation model, and at the same time, random noise that can be used to perform robustness testing on the calibration result is input into the vehicle simulation model. If the output value of the vehicle simulation model regarding the vehicle performance index does not meet the robustness requirements, it can be considered that the calibration result of the actual parameter has poor robustness, and the actual parameter needs to be recalibrated to ensure the robustness of the calibration result.
[0020] Optionally, determining actual noise that can be used to test the robustness of the calibration result from a plurality of candidate noises includes: The actual noise that can be used to test the robustness of the calibration result is determined from the plurality of candidate noises based on an experimental design method.
[0021] In the embodiment of the present application, the actual noise that can be used to test the robustness of the calibration result can be conveniently and accurately determined based on the experimental design method, thereby improving the accuracy of the robustness test of the calibration result.
[0022] In a second aspect, an embodiment of the present application provides a parameter calibration device, the device comprising: An acquisition unit, used for acquiring a plurality of candidate parameters to be calibrated and vehicle performance indicators; a determination unit, configured to determine an actual parameter affecting the vehicle performance index from the plurality of candidate parameters; The calibration unit is used to calibrate the actual parameters to obtain calibration results that meet the expected values of the vehicle performance indicators.
[0023] Optionally, the determining unit includes: The actual parameter determination unit is used to determine the actual parameter affecting the vehicle performance index from the multiple candidate parameters according to the experimental design method.
[0024] Optionally, the actual parameter determination unit is specifically used to: Determine the initial value range and initial step size of each candidate parameter; Determining a plurality of candidate values corresponding to the candidate parameters based on the initial value range and the initial step size; Inputting the plurality of candidate values into the vehicle simulation model in sequence to obtain the maximum value and the minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the influence degree threshold, the corresponding candidate parameter is used as the actual parameter to be calibrated; If the difference between the maximum value and the minimum value is not greater than the influence degree threshold, the corresponding candidate parameter is rejected as the actual parameter to be calibrated.
[0025] Optionally, the calibration unit includes: A determination subunit is used to determine the optimal value range and the optimal step length of the actual parameter according to the experimental design method; The calibration subunit is used to optimize the actual parameters according to the optimal value range and the optimal step size of the actual parameters using a target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index.
[0026] Optionally, the determining subunit is specifically used for: Determining an initial value range of the actual parameter; Inputting the upper limit value and the lower limit value of the initial value range into the precompiled vehicle simulation model in sequence, and obtaining the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value; If the performance indicator difference between the vehicle performance indicator corresponding to the upper limit value and the vehicle performance indicator corresponding to the lower limit value is greater than the upper limit of the range threshold, the initial value range is narrowed; if the performance indicator difference between the vehicle performance indicator corresponding to the upper limit value and the vehicle performance indicator corresponding to the lower limit value is less than the lower limit of the range threshold, the initial value range is increased until the optimal value range of the actual parameter is determined.
[0027] Optionally, the determining subunit is specifically used for: Determining an initial step size of the actual parameter; Determining a plurality of candidate values of the actual parameter within the optimal value range of the actual parameter and the initial step size; Inputting the plurality of candidate values into the vehicle simulation model in sequence to obtain a maximum value and a minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the upper limit of the step length threshold, the initial step length is reduced; If the difference between the maximum value and the minimum value is smaller than the lower limit of the step length threshold, the initial step length is increased until the optimal step length of the actual parameter is determined.
[0028] Optionally, the calibration subunit is specifically used for: Inputting the actual parameters into a precompiled vehicle simulation model to obtain actual values of the vehicle performance indicators; Constructing an objective function of expected value and actual value of the vehicle performance index; With the purpose of minimizing the objective function, the target optimization algorithm is used to perform optimization according to the optimal value range and the optimal step size of the actual parameter to obtain the calibration result of the actual parameter.
[0029] Optionally, the device further comprises: A configuration unit, configured to configure the calibration results of the actual parameters to the vehicle simulation model; The determination unit is further used to determine, from a plurality of candidate noises, an actual noise that can be used to test the robustness of the calibration result; A noise simulation unit, configured to generate the actual noise, and input the actual noise into the vehicle simulation model configured with the calibration result, to obtain a noise environment output value of the vehicle performance index; The calibration unit is further used to recalibrate the actual parameter if the noise environment output value does not meet the robustness requirement.
[0030] Optionally, the determining unit is specifically configured to: The actual noise that can be used to test the robustness of the calibration result is determined from the plurality of candidate noises based on an experimental design method.
[0031] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory for storing a computer program and a processor for executing the computer program, wherein when the computer program is executed by the processor, the electronic device is triggered to execute the steps of the method described in any embodiment of the first aspect.
[0032] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the processor executes the steps of the method described in any embodiment of the first aspect.
[0033] It should be understood that the second to fourth aspects of the embodiments of the present application are consistent with the technical solutions of the first aspect of the embodiments of the present application, and the beneficial effects achieved by each aspect and the corresponding feasible implementation methods are similar and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] 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. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0035] Figure 1 A flow chart of a parameter calibration method provided in an embodiment of the present application; Figure 2 A flowchart of a method for selecting actual parameters provided in an embodiment of the present application; Figure 3 A flowchart of another method for selecting actual parameters provided in an embodiment of the present application; Figure 4 A flow chart of a method for calibrating actual parameters provided in an embodiment of the present application; Figure 5 A flowchart of a method for determining an optimal value range of an actual parameter provided in an embodiment of the present application; Figure 6 A flowchart of a method for determining an optimal step length of actual parameters provided in an embodiment of the present application; Figure 7 A schematic diagram of a flow chart of another method for calibrating parameters provided in an embodiment of the present application; Figure 8 A flowchart of a robustness testing method provided in an embodiment of the present application; Fig. 9 A schematic diagram of the structure of a parameter calibration device provided in an embodiment of the present application; Fig.10 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0036] In order to better understand the technical solution of this specification, the embodiments of the present application are described in detail below with reference to the accompanying drawings.
[0037] It should be clear that the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.
[0038] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit this specification. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.
[0039] In the related art, vehicle parameters are often selected based on manual experience for calibration. However, the above method may result in spending more time to calibrate some parameters that have little impact on vehicle performance indicators, resulting in low calibration efficiency.
[0040] In view of this, an embodiment of the present application provides a parameter calibration method, in which actual parameters that have a significant impact on vehicle performance indicators are screened out from multiple candidate parameters to be calibrated, and then the actual parameters are calibrated, without the need to calibrate other parameters that have no significant impact on the vehicle performance indicators, thereby saving parameter calibration time and improving parameter calibration efficiency.
[0041] The technical solution protected by this application is described in detail below with reference to the accompanying drawings.
[0042] See also Figure 1 , is a flow chart of a parameter calibration method provided in an embodiment of the present application. The method is applied to electronic devices, such as personal computers (PCs), workstations, servers, etc., and the present application does not impose any special restrictions on this. The flow chart of the method is as follows: Step 10: Obtain multiple candidate parameters to be calibrated and vehicle performance indicators.
[0043] In the embodiment of the present application, the multiple candidate parameters to be calibrated can be considered to be selected based on manual experience. The vehicle performance index can be considered to be used to evaluate whether the candidate parameters need to be calibrated, and when it is evaluated to need to be calibrated, to determine the calibration result of the candidate parameters.
[0044] Step 20: Determine the actual parameter that affects the vehicle performance index from multiple candidate parameters.
[0045] In the embodiment of the present application, since multiple candidate parameters are selected based on manual experience, actual parameters that have a significant impact on vehicle performance indicators can be selected from the multiple candidate parameters. The following is a detailed description of how to select the actual parameters from the multiple candidate parameters.
[0046] See also Figure 2 , which is a flow chart of a method for selecting actual parameters provided in an embodiment of the present application. Step 20 can be implemented by executing sub-step 201: Step 201: Determine actual parameters that affect vehicle performance indicators from multiple candidate parameters according to the experimental design method.
[0047] In the embodiment of the present application, the actual parameters that have a significant impact on the vehicle performance indicators can be conveniently and accurately selected from multiple candidate parameters through the experimental design method.
[0048] See also Figure 3 , which is a flowchart of another method for selecting actual parameters provided in an embodiment of the present application. Step 201 can be implemented by executing sub-steps 2011 to 2012: Step 2011: Determine the initial value range and initial step size of each candidate parameter.
[0049] Step 2012: Determine multiple candidate values corresponding to the candidate parameters based on the initial value range and the initial step size.
[0050] Step 2013: Input multiple candidate values into the pre-compiled vehicle simulation model in sequence to obtain the maximum and minimum values corresponding to the vehicle performance indicators.
[0051] Step 2014: If the difference between the maximum value and the minimum value is greater than the influence degree threshold, the corresponding candidate parameter is used as the actual parameter to be calibrated.
[0052] Step 2015: If the difference between the maximum value and the minimum value is not greater than the influence degree threshold, the corresponding candidate parameter is rejected as the actual parameter to be calibrated.
[0053] In the embodiment of the present application, for each candidate parameter, there is an initial value range and an initial step length, and the initial value range and the initial step length can be considered to be set according to artificial experience. This means that there are multiple candidate values for each candidate parameter within its corresponding initial value range, so these multiple candidate values can be sequentially input into the pre-compiled vehicle simulation model, so that multiple values of vehicle performance indicators can be obtained, and the value of a vehicle performance indicator corresponding to one candidate value. From the multiple values of the above-mentioned vehicle performance indicators, the maximum value and the minimum value are selected. If the difference between the maximum value and the minimum value is greater than the impact degree threshold, it indicates that the current candidate parameter value changes, and the impact on the vehicle performance index is large, that is, the current candidate parameter has a significant impact on the vehicle performance index, then the current candidate parameter is used as the actual parameter that needs to be calibrated. If the difference between the maximum value and the minimum value is less than the impact degree threshold, it indicates that the current candidate parameter value changes, and the impact on the vehicle performance index is small, that is, the current candidate parameter does not have a significant impact on the vehicle performance index, then the current candidate parameter is screened out and is not used as the actual parameter that needs to be calibrated.
[0054] Step 30: Calibrate the actual parameters to obtain calibration results that meet the expected values of vehicle performance indicators.
[0055] In the embodiment of the present application, after the actual parameters to be calibrated are selected from multiple candidate parameters to be calibrated, the actual parameters can be calibrated for the purpose of achieving the expected values of the vehicle performance indicators, thereby obtaining the calibration results of the actual parameters.
[0056] The calibration process of the actual parameters is described in detail below.
[0057] See also Figure 4 , which is a flow chart of a method for calibrating actual parameters provided in an embodiment of the present application. Step 30 can be specifically performed by executing sub-steps 301 to 302: Step 301: Determine the optimal value range and optimal step length of the actual parameters according to the experimental design method.
[0058] See also Figure 5, is a flowchart of a method for determining an optimal value range of an actual parameter provided in an embodiment of the present application. Step 301 can be implemented by executing sub-steps 3011 to 3014: Step 3011: Determine the initial value range of the actual parameter.
[0059] Step 3012: input the upper limit value and the lower limit value of the initial value range into the precompiled vehicle simulation model in turn to obtain the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value.
[0060] Step 3013: If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is greater than the upper limit of the range threshold, the initial value range is narrowed.
[0061] Step 3014: If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is less than the lower limit of the range threshold, the initial value range is increased until the optimal value range of the actual parameter is determined.
[0062] In the embodiment of the present application, for the actual parameter that has a significant impact on the vehicle performance index, there is a corresponding initial value range. If the upper limit value and the lower limit value of the initial value range are respectively input into the pre-compiled vehicle simulation model, the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value can be obtained. If the difference between the above two vehicle performance indicators is large, it indicates that the current initial value range has a large impact on the vehicle performance index, and the initial value range can be reduced at this time; conversely, if the difference between the above two vehicle performance indicators is small, it indicates that the current initial value range has a small impact on the vehicle performance index, and the initial value range can be increased at this time. If the difference between the above two vehicle performance indicators is between the lower limit of the range threshold and the upper limit of the range threshold, it can be considered that the value range is more appropriate and no adjustment is required. Based on the above value range adjustment rules, the optimal value range of the actual parameter can be determined.
[0063] See also Figure 6 , is a flowchart of a method for determining an optimal step length of actual parameters provided in an embodiment of the present application. Step 301 can be implemented by executing sub-steps 3015 to 3014: Step 3015: Determine the initial step size of the actual parameters.
[0064] Step 3016: Determine multiple candidate values of the actual parameter within the optimal value range of the actual parameter and the initial step size.
[0065] Step 3017: Input multiple candidate values into the vehicle simulation model in sequence to obtain the maximum and minimum values corresponding to the vehicle performance indicators.
[0066] Step 3018: If the difference between the maximum value and the minimum value is greater than the upper limit of the step length threshold, the initial step length is reduced.
[0067] Step 3019: If the difference between the maximum value and the minimum value is less than the lower limit of the step size threshold, the initial step size is increased until the optimal step size of the actual parameter is determined.
[0068] In the embodiment of the present application, for the actual parameter that has a significant impact on the vehicle performance index, there is a corresponding initial step length, so multiple candidate values can be determined based on the initial step length from the optimal value interval of the actual parameter. These multiple candidate values are input into the vehicle simulation model in sequence, and the maximum and minimum values of the vehicle performance index within the optimal value interval can be obtained. If the difference between the maximum value and the minimum value is large, it indicates that the current initial step length has a greater impact on the vehicle performance index, and the initial step length can be reduced at this time; conversely, if the difference between the maximum value and the minimum value is small, it indicates that the current initial step length has a smaller impact on the vehicle performance, and the initial step length can be increased at this time. If the difference between the maximum value and the minimum value is between the lower limit of the step length threshold and the upper limit of the step length threshold, the initial step length is not adjusted. Based on the above step length adjustment rule, the optimal step length can be determined within the optimal value interval.
[0069] Step 302: According to the optimal value range and optimal step length of the actual parameters, the actual parameters are optimized using a target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index.
[0070] In the embodiment of the present application, the target optimization algorithm can select a corresponding optimization algorithm according to the number of actual parameters to be optimized, and the present application does not impose any special restrictions on this.
[0071] See also Figure 7 , which is a flow chart of another method for calibrating parameters provided in an embodiment of the present application. Step 302 can be specifically performed by executing sub-steps Step 3021: Input the actual parameters into the precompiled vehicle simulation model to obtain the actual values of the vehicle performance indicators.
[0072] Step 3022: Construct an objective function of the expected value and actual value of the vehicle performance index.
[0073] Step 3023: With the purpose of minimizing the objective function, the target optimization algorithm is used to optimize the optimal value range and the optimal step size of the actual parameters to obtain the calibration results of the actual parameters.
[0074] In an embodiment of the present application, an objective function is constructed based on the expected value and actual value of the vehicle performance index. Then, in the actual parameter calibration process, with the purpose of minimizing the objective function, an optimization is performed based on the optimal value range and optimal step size of the actual parameters, thereby accurately determining the calibration result of the actual parameters.
[0075] In some embodiments, after obtaining the calibration results of the actual parameters, the calibration results may be subjected to a robustness test.
[0076] See also Figure 8 , which is a flowchart of a robustness testing method provided in an embodiment of the present application. After executing step 30, you can continue to execute step Step 40: Configure the calibration results of the actual parameters to the vehicle simulation model.
[0077] Step 50: Determine actual noise from a plurality of candidate noises that can be used to test the robustness of the calibration result.
[0078] In an embodiment of the present application, multiple candidate noises can be considered as artificially selected noises. On this basis, actual noises that can be used to test the robustness of the calibration results can be selected from multiple candidate operations based on the experimental design method, thereby improving the accuracy of the robustness test of the calibration results.
[0079] Step 60: Generate actual noise, and input the actual noise into a vehicle simulation model configured with a calibration result to obtain a noise environment output value of a vehicle performance index.
[0080] Step 70: If the output value in the noise environment does not meet the robustness requirement, the actual parameters are recalibrated.
[0081] In the embodiment of the present application, after the actual parameter calibration is completed, the calibration result of the actual parameter can be configured into the vehicle simulation model, and at the same time, random noise that can be used to perform robustness testing on the calibration result is input into the vehicle simulation model. If the output value of the vehicle simulation model regarding the vehicle performance index does not meet the robustness requirements, it can be considered that the calibration result of the actual parameter has poor robustness, and the actual parameter needs to be recalibrated to ensure the robustness of the calibration result.
[0082] The parameter calibration process is explained in detail below through an example.
[0083] First, based on manual experience, multiple candidate parameters to be calibrated are determined as stator temperature, power distribution coefficient, and compensation charging power. Vehicle performance indicators are: the ratio of the pedal change rate to the simulated vehicle speed change rate, which can be used as a reference for drivability evaluation; the mean square error between the target vehicle speed and the actual vehicle speed, which can be used as a reference for dynamic evaluation; and the loss power (engine, motor, battery), which can be used as an economic evaluation. The above-mentioned stator temperature, power distribution coefficient, and compensation charging power all have corresponding initial value ranges and initial step sizes. The power distribution coefficient can be selected from the above-mentioned multiple candidate parameters as the actual parameter through the experimental design method.
[0084] Then, the optimal range and step size of the actual parameters are determined according to the experimental design method, and the calibration results that meet the expected values of the vehicle performance indicators are determined in combination with the target optimization algorithm. Among them, if the actual parameters that need to be calibrated are less than or equal to 3, the target optimization algorithm can be NSGA-II, and if the number of actual parameters that need to be calibrated is greater than 3, the target optimization algorithm can be NSGA-III. The expected value of the vehicle performance index refers to the ratio of the pedal change rate to the simulated vehicle speed change rate, the mean square error between the target vehicle speed and the actual vehicle speed, and the loss power (engine, motor, battery) are minimized as a whole, or each variable is minimized. It is worth noting that if the expected value of the vehicle performance index refers to the ratio of the pedal change rate to the simulated vehicle speed change rate, the mean square error between the target vehicle speed and the actual vehicle speed, and the loss power (engine, motor, battery) as a whole being minimized, then the ratio of the pedal change rate to the simulated vehicle speed change rate can be assigned a first weight, the mean square error between the target vehicle speed and the actual vehicle speed can be assigned a second weight, and the loss power (engine, motor, battery) can be assigned a third weight, the second weight is greater than the first weight, and the first weight is greater than the third weight.
[0085] After determining the calibration result of the actual parameter, the calibration result can be configured into the vehicle simulation model, and then the actual noise that can be used to test the robustness of the calibration result can be determined from a variety of known noises according to the experimental design method. For example, the actual noise includes: transmission system noise (simulating the dispersion of the entire transmission system due to differences in parts or assembly), road noise (simulating the random distribution of resistance caused by different road adhesion coefficients), power system noise (simulating the dispersion of the entire power system due to differences in parts or assembly), and of course other types of noise can also be included, and this application does not impose any special restrictions on this. On this basis, the actual noise is generated, and the actual noise is input into the vehicle simulation model, and then the output value of the vehicle performance index in the noise environment is observed. For example, the robustness of the calibration result can be evaluated by observing the mean square error of the target speed and the simulated speed, as well as the power loss (engine, motor, battery). If the output value of the vehicle performance index in the noise environment does not meet the robustness requirements after evaluation, the calibration results of the above-mentioned actual parameters are recalibrated until they meet the robustness requirements.
[0086] See also Fig. 9 , an embodiment of the present application provides a parameter calibration device, the device comprising: An acquisition unit 801 is used to acquire a plurality of candidate parameters to be calibrated and vehicle performance indicators; The determination unit 802 is used to determine the actual parameter that affects the vehicle performance index from multiple candidate parameters.
[0087] The calibration unit 803 is used to calibrate the actual parameters to obtain calibration results that meet the expected values of vehicle performance indicators.
[0088] Optionally, the determining unit 802 includes: The actual parameter determination unit is used to determine the actual parameter affecting the vehicle performance index from a plurality of candidate parameters according to the experimental design method.
[0089] Optionally, the actual parameter determination unit is specifically used for: Determine the initial value range and initial step size of each candidate parameter; Determine multiple candidate values of corresponding candidate parameters based on the initial value range and the initial step size; Inputting multiple candidate values into the vehicle simulation model in sequence to obtain the maximum and minimum values corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the influence degree threshold, the corresponding candidate parameter is used as the actual parameter to be calibrated; If the difference between the maximum value and the minimum value is not greater than the influence degree threshold, the corresponding candidate parameter is rejected as the actual parameter to be calibrated.
[0090] Optionally, the calibration unit 803 includes: A determination subunit is used to determine the optimal value range and the optimal step length of the actual parameter according to the experimental design method; The calibration subunit is used to optimize the actual parameters according to the optimal value range and optimal step size of the actual parameters using the target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index.
[0091] Optionally, determine that the subunit is specifically used for: Determine the initial value range of the actual parameters; Inputting the upper limit value and the lower limit value of the initial value range into the precompiled vehicle simulation model in turn, obtaining the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value; If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is greater than the upper limit of the range threshold, the initial value range is narrowed; if the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is less than the lower limit of the range threshold, the initial value range is increased until the optimal value range of the actual parameter is determined.
[0092] Optionally, determine that the subunit is specifically used for: Determine the initial step size of the actual parameters; Determining multiple candidate values of the actual parameter within an optimal value range and an initial step size of the actual parameter; Inputting multiple candidate values into the vehicle simulation model in sequence to obtain the maximum and minimum values corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the upper limit of the step size threshold, the initial step size is reduced; If the difference between the maximum value and the minimum value is less than the lower limit of the step size threshold, the initial step size is increased until the optimal step size of the actual parameter is determined.
[0093] Optionally, the calibration subunit is specifically used for: Input actual parameters into the precompiled vehicle simulation model to obtain actual values of vehicle performance indicators; Construct an objective function of the expected and actual values of vehicle performance indicators; With the purpose of minimizing the objective function, the target optimization algorithm is used to optimize the optimal value range and the optimal step size of the actual parameters to obtain the calibration results of the actual parameters.
[0094] Optionally, the device further comprises: A configuration unit 804, configured to configure the calibration results of the actual parameters to the vehicle simulation model; The determination unit 802 is further configured to determine, from a plurality of candidate noises, an actual noise that can be used to test the robustness of the calibration result; The noise simulation unit 805 is used to generate actual noise and input the actual noise into the vehicle simulation model configured with the calibration result to obtain the noise environment output value of the vehicle performance index; The calibration unit 803 is further configured to recalibrate the actual parameter if the noise environment output value does not meet the robustness requirement.
[0095] Optionally, the determining unit 802 is specifically configured to: The actual noise that can be used to test the robustness of the calibration results is determined from multiple candidate noises based on the experimental design method.
[0096] See also Fig.10 , is an electronic device provided in an embodiment of the present application, the electronic device includes at least one processor 901, the processor 901 is used to execute a computer program stored in a memory, and implement the embodiment of the present application as provided in the following Figure 1-8 Schematic diagram of the process flow of the parameter calibration method shown.
[0097] Optionally, the processor 901 may specifically be a central processing unit, a specific ASIC, or one or more integrated circuits for controlling program execution.
[0098] Optionally, the electronic device may further include a memory 902 connected to the at least one processor 901, and the memory 902 may include a ROM, a RAM, and a disk memory. The memory 902 is used to store data required by the processor 401 when it is running, that is, it stores instructions that can be executed by the at least one processor 901. The at least one processor 901 executes the instructions stored in the memory 902 to execute the following. Figure 1-8 The method shown. Wherein, the number of memories 902 is one or more. Wherein, the number of memories 902 is one or more.
[0099] The present application also provides a computer storage medium, wherein the computer storage medium stores a computer program, and when the computer program is run on a processor, the processor executes Figure 1-8 method.
[0100] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.
Claims
1. A parameter calibration method, characterized in that: The method comprises: Acquire multiple candidate parameters to be calibrated and vehicle performance indicators; Determining an actual parameter affecting the vehicle performance indicator from the plurality of candidate parameters; The actual parameters are calibrated to obtain calibration results that meet the expected values of the vehicle performance indicators.
2. The method according to claim 1, characterized in that Determining an actual parameter affecting the vehicle performance index from the plurality of candidate parameters includes: The actual parameters affecting the vehicle performance index are determined from the multiple candidate parameters according to the experimental design method.
3. The method according to claim 2, characterized in that Determining actual parameters affecting the vehicle performance index from the plurality of candidate parameters according to an experimental design method includes: Determine the initial value range and initial step size of each candidate parameter; Determining a plurality of candidate values corresponding to the candidate parameters based on the initial value range and the initial step size; Inputting the plurality of candidate values into the vehicle simulation model in sequence to obtain the maximum value and the minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the influence degree threshold, the corresponding candidate parameter is used as the actual parameter to be calibrated; If the difference between the maximum value and the minimum value is not greater than the influence degree threshold, the corresponding candidate parameter is rejected as the actual parameter to be calibrated.
4. The method according to claim 1, characterized in that: Calibrate the actual parameters to obtain calibration results that meet the expected values of the vehicle performance indicators, including: Determine the optimal value range and the optimal step length of the actual parameters according to the experimental design method; According to the optimal value range and the optimal step length of the actual parameters, the actual parameters are optimized using a target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index.
5. The method according to claim 4, characterized in that The optimal value range of the actual parameters is determined according to the experimental design method, including: Determining an initial value range of the actual parameter; Inputting the upper limit value and the lower limit value of the initial value range into the precompiled vehicle simulation model in sequence, and obtaining the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value; If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is greater than the upper limit of the range threshold, the initial value range is narrowed; If the performance index difference between the vehicle performance index corresponding to the upper limit value and the vehicle performance index corresponding to the lower limit value is less than the lower limit of the range threshold, the initial value range is increased until the optimal value range of the actual parameter is determined.
6. The method according to claim 5, characterized in that The optimal step length of the actual parameters is determined according to the experimental design method, including: Determining an initial step size of the actual parameter; Determining a plurality of candidate values of the actual parameter within the optimal value range of the actual parameter and the initial step size; Inputting the plurality of candidate values into the vehicle simulation model in sequence to obtain a maximum value and a minimum value corresponding to the vehicle performance index; If the difference between the maximum value and the minimum value is greater than the upper limit of the step length threshold, the initial step length is reduced; If the difference between the maximum value and the minimum value is smaller than the lower limit of the step length threshold, the initial step length is increased until the optimal step length of the actual parameter is determined.
7. The method according to claim 4, characterized in that According to the optimal value range and the optimal step length of the actual parameter, the actual parameter is optimized by using the target optimization algorithm to obtain a calibration result that meets the expected value of the vehicle performance index, including: Inputting the actual parameters into a precompiled vehicle simulation model to obtain actual values of the vehicle performance indicators; Constructing an objective function of expected value and actual value of the vehicle performance index; With the purpose of minimizing the objective function, the target optimization algorithm is used to perform optimization according to the optimal value range and the optimal step size of the actual parameter to obtain the calibration result of the actual parameter.
8. The method according to any one of claims 1 to 7, characterized in that: The method further comprises: Configuring the calibration results of the actual parameters to the vehicle simulation model; Determining an actual noise from a plurality of candidate noises that can be used to test the robustness of the calibration result; Generating the actual noise, and inputting the actual noise into the vehicle simulation model configured with the calibration result, to obtain a noise environment output value of the vehicle performance index; If the noise environment output value does not meet the robustness requirement, the actual parameter is recalibrated.
9. The method according to claim 8, characterized in that Determining actual noise from a plurality of candidate noises that can be used to test the robustness of the calibration result includes: The actual noise that can be used to test the robustness of the calibration result is determined from the plurality of candidate noises based on an experimental design method.
10. An electronic device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to implement the parameter calibration method according to any one of claims 1 to 9.
11. A computer-readable storage medium, characterized in that: Used to store a computer program, wherein when the computer program is executed by a processor, the parameter calibration method according to any one of claims 1 to 9 is implemented.