CN-122009219-A - New energy automobile endurance calibration method, device, equipment, medium and product

CN122009219ACN 122009219 ACN122009219 ACN 122009219ACN-122009219-A

Abstract

The invention discloses a new energy automobile endurance calibration method, device, equipment, medium and product. The method comprises the steps of collecting running data of a vehicle, extracting characteristic parameters representing driving styles of users based on the running data, carrying out iterative cluster analysis, constructing driving behavior images corresponding to different driving styles, configuring corresponding initial cruising calculation weights, calculating error values of actual driving energy consumption and preset standard energy consumption, obtaining an influence coefficient by combining the influence of vehicle speed fluctuation on the energy consumption, carrying out dynamic iterative optimization on the initial cruising calculation weights corresponding to the driving behavior images, setting upper and lower limit boundary constraints, calculating the basic cruising mileage of the vehicle, calculating to obtain real-time dynamic cruising mileage of the vehicle according to the cruising calculation weights corresponding to the driving behavior images and scene correction coefficients corresponding to real-time road conditions, and outputting the real-time dynamic cruising mileage to a vehicle instrument end for display. According to the scheme, the cruising display precision of the new energy automobile can be improved, and cruising anxiety of a user is reduced.

Inventors

YIN PENG
LI SONGSONG
XIANG XIAOLEI
JIANG CONGHUI
GAO JIEPENG
LIU JIAHUI
LIU JIANYU
ZHAO SHUAI

Assignees

一汽解放汽车有限公司

Dates

Publication Date: 20260512
Application Date: 20260407

Claims (10)

1. The new energy automobile endurance calibration method is characterized by comprising the following steps of: collecting operation data of a vehicle through a vehicle power battery management system BMS and a whole vehicle controller, wherein the operation data comprise user driving behavior data, power battery real-time state data and vehicle running real-time road condition data; based on the operation data, extracting characteristic parameters of the driving style of the user with multi-dimensions, performing iterative cluster analysis on the characteristic parameters through an unsupervised clustering algorithm, performing quantitative classification on the driving behaviors of the user, constructing driving behavior portraits corresponding to different driving styles, and configuring corresponding initial cruising calculation weights for each type of driving behavior portraits; Calculating an error value of actual driving energy consumption and preset standard energy consumption based on user history and real-time driving data, obtaining an influence coefficient of the driving behavior on the cruising deviation by combining the influence quantification of vehicle speed fluctuation on the energy consumption, carrying out dynamic iterative optimization on an initial cruising calculation weight corresponding to the driving behavior image based on the influence coefficient, and setting upper and lower limit boundary constraints for various weights; And calculating the basic endurance mileage of the vehicle, calculating to obtain the real-time dynamic endurance mileage of the vehicle according to the endurance calculation weight corresponding to the driving behavior image of the current user and the scene correction coefficient corresponding to the real-time road condition, and outputting the real-time dynamic endurance mileage to a vehicle instrument end for display.
2. The method of claim 1, wherein after collecting the operation data of the vehicle with the vehicle controller through the vehicle power battery management system BMS, the method further comprises: For the operation data, calculating an arithmetic mean value and a standard deviation of a sample set by taking a continuous preset number of sampling points as a statistical sample set; When the relation between the parameter value of the sampling point and the arithmetic mean value and the standard deviation meets a preset condition, judging that the sampling point is abnormal data, and removing the abnormal data from the sample set; and supplementing the blank value generated after the abnormal data are removed according to the median of the effective data in the statistical sample set.
3. The method of claim 1, wherein constructing driving behavior portraits corresponding to different driving styles comprises: Calculating a sliding average value by taking preset market time as a sliding statistical window from the acceleration average value representing the driving excitation degree, the braking frequency representing the driving smoothness and the energy consumption ratio representing the electric quantity and the mileage conversion efficiency, wherein the energy consumption ratio is the ratio of the electric quantity variation of a battery in the statistical window to the corresponding driving mileage; And carrying out iterative clustering on the three-dimensional feature vectors by adopting a K-means unsupervised clustering algorithm, setting the clustering quantity K=3, and continuously and iteratively updating the clustering center corresponding to the aggressive, smooth and economic driving behavior portraits until the variation of the clustering center is smaller than or equal to the preset value of the preset clustering error threshold value, stopping iteration, and outputting the driving behavior portraits.
4. The method according to claim 3, wherein the dynamically iteratively optimizing the initial endurance calculation weights corresponding to the driving behavior image based on the influence coefficient includes: Calculating the actual energy consumption ratio in the current running process through the running data, and obtaining an energy consumption error corresponding to the current running according to the actual energy consumption ratio and a pre-calibrated reference energy consumption ratio of the vehicle; Dividing the standard deviation of the vehicle speed in the current running process by the arithmetic mean value of the vehicle speed to obtain a vehicle speed fluctuation ratio, and summing the vehicle speed fluctuation ratio with a value 1 to obtain a vehicle speed change influence coefficient; obtaining a driving influence coefficient according to the vehicle speed change influence coefficient and the energy consumption error, wherein the numerical value of the driving influence coefficient has a positive correlation with the influence degree of corresponding driving behavior on the endurance deviation; Taking an arithmetic average value of driving influence coefficients corresponding to 3 last effective driving of the images as an average driving influence coefficient of the images aiming at each type of driving behavior images, multiplying the weight of the current iteration period of the images by the sum of products of 1 and the average driving influence coefficient and a preset iteration coefficient to obtain the optimized weight of the next iteration period of the images; and setting upper and lower limit boundary constraints for the duration calculation weights corresponding to the driving behavior images, and taking the boundary value corresponding to the range as the effective duration calculation weight of the image if the iteratively optimized weight exceeds the constraint range of the corresponding category.
5. The method according to claim 1, wherein the calculating the basic range of the vehicle, according to the range calculation weight corresponding to the driving behavior image to which the current user belongs and the scene correction coefficient corresponding to the real-time road condition, calculates the real-time dynamic range of the vehicle, includes: calculating a basic endurance mileage of the vehicle based on the real-time residual electric quantity of the power battery of the vehicle and a preset reference energy consumption ratio; and calculating real-time dynamic endurance mileage according to the basic endurance mileage, the endurance calculation weight and the scene correction coefficient, wherein the real-time road condition correction coefficient is set in a grading manner according to the real-time road condition acquired by a navigation system of the vehicle.
6. The method according to claim 1, wherein the method further comprises: After the vehicle continuously runs a preset verification driving mileage, acquiring real-time dynamic driving mileage displayed by the instrument ends before and after the vehicle finishes the verification driving mileage, and determining a change difference value; when the cruising deviation rate is larger than or equal to a first preset value and smaller than a second preset value, setting a preset iteration coefficient to be 0.1, and executing 1-time weight iteration optimization operation after each time of the vehicle finishes a preset running process, wherein the first preset value is smaller than the second preset value; And when the endurance deviation rate is greater than or equal to the second preset value, adjusting the preset iteration coefficient to 0.15, executing 1-time weight iteration optimization operation, and after the iteration optimization is completed, recalculating the real-time dynamic endurance mileage of the vehicle and synchronously updating the real-time dynamic endurance mileage to the vehicle instrument end for display.
7. The utility model provides a new energy automobile duration calibrating device which characterized in that includes: The running data acquisition unit is used for acquiring running data of the vehicle through the BMS and the whole vehicle controller, wherein the running data comprise user driving behavior data, real-time state data of the power battery and real-time road condition data of vehicle running; The driving behavior portrayal construction unit is used for extracting characteristic parameters representing the driving style of the user in a multi-dimensional mode based on the operation data, carrying out iterative cluster analysis on the characteristic parameters through an unsupervised clustering algorithm, carrying out quantitative classification on the driving behaviors of the user, constructing driving behavior portrayal corresponding to different driving styles, and configuring corresponding initial cruising calculation weights for each type of driving behavior portrayal; the cruising calculation weight optimizing unit is used for calculating an error value of actual driving energy consumption and preset standard energy consumption based on user history and real-time driving data, obtaining an influence coefficient of the driving behavior on cruising deviation by combining the influence quantification of vehicle speed fluctuation on the energy consumption, carrying out dynamic iterative optimization on an initial cruising calculation weight corresponding to the driving behavior image based on the influence coefficient, and setting upper and lower limit boundary constraint for various weights; The real-time dynamic endurance calculation unit is used for calculating the basic endurance mileage of the vehicle, calculating the real-time dynamic endurance mileage of the vehicle according to the endurance calculation weight corresponding to the driving behavior image of the current user and the scene correction coefficient corresponding to the real-time road condition, and outputting the real-time dynamic endurance mileage to the vehicle instrument end for display.
8. An electronic device, the electronic device comprising: And a memory communicatively coupled to the at least one processor, wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the new energy vehicle range calibration method of any one of claims 1-6.
9. A computer readable storage medium, wherein the computer readable storage medium stores computer instructions for causing a processor to implement the new energy vehicle cruising calibration method according to any one of claims 1-6 when executed.
10. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the new energy vehicle endurance calibration method according to any one of claims 1-6.

Description

New energy automobile endurance calibration method, device, equipment, medium and product Technical Field The invention relates to the technical field of battery management, in particular to a new energy automobile endurance calibration method, device, equipment, medium and product. Background Currently, a main flow technical scheme of estimating the endurance mileage of a new energy automobile in industry mostly adopts a fixed parameter estimation model based on a battery state of charge (SOC), the scheme takes the reference energy consumption ratio of the residual battery power and calibration before the delivery of the automobile as a core calculation basis, the model architecture is simple, the engineering landing is easy, and the scheme is a basic scheme adopted by the current multiple production automobile types. In order to solve the problem of estimation deviation of a fixed parameter model, the prior art also carries out multidirectional optimization exploration, wherein part of schemes introduce parameters such as ambient temperature, battery health Status (SOH) and the like to correct a basic endurance value, part of researches attempt to extract single characteristic parameters of driving behaviors to adjust a battery discharge strategy, and part of schemes combine navigation road condition information to carry out scene adaptation on endurance estimation. However, the prior art still has technical defects which are difficult to overcome, and the main stream fixed parameter model does not consider the personalized difference of the driving behaviors of users, so that the deviation between the apparent endurance and the actual driving mileage of different users under different working conditions is obvious. Aiming at the problem of large deviation between the display and the actual endurance in the prior art, a new energy automobile endurance calibration method is needed to improve the endurance display precision and reduce the endurance anxiety of users. Disclosure of Invention The invention provides a new energy automobile endurance calibration method, device, equipment, medium and product, which are used for improving the endurance display precision of the new energy automobile and reducing the endurance anxiety of a user. According to one aspect of the invention, a new energy automobile endurance calibration method is provided, which comprises the following steps: collecting operation data of a vehicle through a vehicle power battery management system BMS and a whole vehicle controller, wherein the operation data comprise user driving behavior data, power battery real-time state data and vehicle running real-time road condition data; based on the operation data, extracting characteristic parameters of the driving style of the user with multi-dimensions, performing iterative cluster analysis on the characteristic parameters through an unsupervised clustering algorithm, performing quantitative classification on the driving behaviors of the user, constructing driving behavior portraits corresponding to different driving styles, and configuring corresponding initial cruising calculation weights for each type of driving behavior portraits; Calculating an error value of actual driving energy consumption and preset standard energy consumption based on user history and real-time driving data, obtaining an influence coefficient of the driving behavior on the cruising deviation by combining the influence quantification of vehicle speed fluctuation on the energy consumption, carrying out dynamic iterative optimization on an initial cruising calculation weight corresponding to the driving behavior image based on the influence coefficient, and setting upper and lower limit boundary constraints for various weights; And calculating the basic endurance mileage of the vehicle, calculating to obtain the real-time dynamic endurance mileage of the vehicle according to the endurance calculation weight corresponding to the driving behavior image of the current user and the scene correction coefficient corresponding to the real-time road condition, and outputting the real-time dynamic endurance mileage to a vehicle instrument end for display. Optionally, after collecting the operation data of the vehicle by the vehicle power battery management system BMS and the whole vehicle controller, the method further comprises: For the operation data, calculating an arithmetic mean value and a standard deviation of a sample set by taking a continuous preset number of sampling points as a statistical sample set; When the relation between the parameter value of the sampling point and the arithmetic mean value and the standard deviation meets a preset condition, judging that the sampling point is abnormal data, and removing the abnormal data from the sample set; and supplementing the blank value generated after the abnormal data are removed according to the median of the effective data in the statistical sample set. Optionally,