Search

CN-121977861-A - Method for predicting lifetime of tire, vehicle and computer program product

CN121977861ACN 121977861 ACN121977861 ACN 121977861ACN-121977861-A

Abstract

The application provides a service life prediction method of a tire, a vehicle and a computer program product, wherein the method is applied to the technical field of tire detection and comprises the steps of collecting tread images of the tire of the vehicle and collecting running condition data of the vehicle; extracting at least one tread wear characteristic of the tire according to the tire tread image, determining the calculation weight of an initial life prediction model according to the driving condition data and the at least one tread wear characteristic, constructing a life prediction model which is currently used, and inputting the at least one tread wear characteristic into the life prediction model which is currently used so as to output a life prediction result of the vehicle tire. The method solves the problems of high vehicle maintenance cost and long period caused by low detection efficiency and low time consumption of the existing detection method for the tire wear of the vehicle, and can effectively improve the identification accuracy of the tire wear state and the accuracy of life prediction.

Inventors

  • LIAN CHENGLIN

Assignees

  • 长城汽车股份有限公司

Dates

Publication Date
20260505
Application Date
20260112

Claims (10)

  1. 1. A method for predicting the life of a tire, comprising the steps of: Acquiring tread images of vehicle tires and acquiring driving condition data of the vehicle; Extracting at least one tread wear characteristic of the tire according to the tire tread image, determining the calculation weight of an initial life prediction model according to the driving working condition data and the at least one tread wear characteristic, and constructing a currently used life prediction model; The at least one tread wear feature is input to the currently used life prediction model to output a life prediction result of the vehicle tire.
  2. 2. The method of claim 1, wherein determining the computational weight of the initial life prediction model based on the driving condition data and the at least one tread wear feature comprises: Determining a current load, a current braking frequency, and a current wear rate of the vehicle based on the driving condition data and the at least one tread wear feature; And determining the calculation weight of the initial life prediction model according to the current load, the current braking frequency and the current wear rate.
  3. 3. The method of claim 2, wherein the calculated weights of the initial life prediction model include a load weight, a brake frequency weight, and a wear rate weight, the determining the calculated weights of the initial life prediction model based on the current load, the current brake frequency, and the current wear rate comprising: In response to the current load being greater than a preset load value, adjusting the load weight to a first weight value; In response to the current braking frequency being greater than a preset frequency value, adjusting the braking frequency weight to a second weight value; and adjusting the wear rate weight to a third weight value in response to the duration of time that the current wear rate is greater than a preset rate value being greater than a preset duration of time.
  4. 4. The method according to claim 1, further comprising, after outputting the life prediction result of the vehicle tire: and carrying out early warning reminding under the condition that the service life prediction result of the vehicle tire meets the preset early warning condition.
  5. 5. The method of claim 4, wherein the life prediction result includes at least one of a remaining tread thickness, a remaining service life, and a wear prediction rate, and determining whether the vehicle tire meets a preset pre-warning condition based on the life prediction result comprises: and judging that the vehicle tire meets the preset early warning condition under the condition that the thickness of the residual tread is smaller than a first preset value, and/or the residual service life is smaller than a second preset value, and/or the wear prediction rate is larger than a third preset value.
  6. 6. The method as recited in claim 5, further comprising: acquiring the driving mileage, tire state data and tire replacement data of the vehicle; and adjusting the first preset value and/or the second preset value and/or the third preset value according to the driving mileage, the tire state data and the tire replacement data.
  7. 7. The method according to claim 1, characterized by, after outputting the life prediction result of the vehicle tire, comprising: and retraining the currently used life prediction model to predict a life prediction result of the tire by using the retrained life prediction model when the driving mileage of the tire is greater than or equal to a preset mileage and/or the prediction error of the currently used life prediction model is greater than a fourth preset value.
  8. 8. The method of claim 1, comprising, prior to determining the computational weights of the initial life prediction model from the driving condition data: constructing a data set to be trained, wherein the data set to be trained comprises a tire tread abrasion image data set, a corresponding driving working condition data set and a corresponding service life data set; Preprocessing the data set to be trained, and dividing the processed data set to be trained into a training set, a verification set and a test set; Constructing a target neural network, training the target neural network by using the training set, and evaluating the prediction error of the trained target neural network by using the verification set; And after the prediction error of the trained target neural network is smaller than or equal to a preset error value, testing the trained target neural network by using the test set test, and obtaining the initial life prediction model after the test result of the trained target neural network reaches a preset stopping iteration condition.
  9. 9. A vehicle comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the method of predicting the life of a tyre as claimed in any one of claims 1 to 8.
  10. 10. A computer program product, characterized in that it comprises computer program code which, when run on a computer, causes the computer to carry out the method of life prediction of a tyre according to any one of claims 1 to 8.

Description

Method for predicting lifetime of tire, vehicle and computer program product Technical Field The present application relates to the field of tire detection technology, and more particularly, to a method of predicting the life of a tire in the field of tire detection technology, a vehicle, and a computer program product. Background At present, the detection accuracy of the detection mode of tire wear is insufficient, the tire wear state cannot be grasped dynamically, the potential safety hazard of the tire is difficult to check in time, the manual detection mode is time-consuming and low in detection efficiency, and the vehicle maintenance cost is high and the period is long. Disclosure of Invention The application provides a service life prediction method of a tire, a vehicle and a computer program product, which solve the problems of high maintenance cost and long period of the vehicle caused by low detection efficiency and low time consumption in the existing method for detecting the abrasion of the tire of the vehicle, and can effectively improve the identification precision of the abrasion state of the tire and the precision of service life prediction. In a first aspect, a method for predicting the life of a tire is provided, which comprises the steps of collecting a tread image of the tire of a vehicle and collecting driving condition data of the vehicle, extracting at least one tread wear characteristic of the tire according to the tread image of the tire, determining a calculation weight of an initial life prediction model according to the driving condition data and the at least one tread wear characteristic, constructing a life prediction model which is currently used, and inputting the at least one tread wear characteristic into the life prediction model which is currently used so as to output a life prediction result of the tire of the vehicle. With reference to the first aspect, in some possible implementations, determining the computational weight of the initial life prediction model according to the driving condition data and the at least one tread wear feature includes determining a current load, a current braking frequency, and a current wear rate of the vehicle based on the driving condition data and the at least one tread wear feature, and determining the computational weight of the initial life prediction model according to the current load, the current braking frequency, and the current wear rate. According to the technical scheme, the differential quantification of the influence of multiple factors on the abrasion is realized through dynamic weight distribution, so that the abrasion prediction result is more fit with the real-time running state of the vehicle, and the problem of inaccurate prediction of the fixed weight in a heavy load/frequent braking scene is avoided. With reference to the first aspect, in some possible implementations, the calculating weights of the initial life prediction model include a load weight, a brake frequency weight, and a wear rate weight, and the determining the calculating weights of the initial life prediction model according to the current load, the current brake frequency, and the current wear rate includes adjusting the load weight to a first weight value in response to the current load being greater than a preset load value, adjusting the brake frequency weight to a second weight value in response to the current brake frequency being greater than a preset frequency value, and adjusting the wear rate weight to a third weight value in response to a duration of the current wear rate being greater than a preset rate value being greater than a preset duration. Through the technical scheme, a weight adjustment mechanism is set for three factors of load, braking frequency and wear rate, prediction deviation caused by average weight or fixed weight of a traditional model is avoided, so that life prediction precision is improved, misjudgment caused by instantaneous errors of a sensor can be effectively filtered for duration conditions of duration exceeding duration of the wear rate, the weight is adjusted upwards only when a tire enters a stable high-wear stage, frequent early warning interference to a user is avoided, meanwhile, early warning accuracy can be improved, and misinformation (such as misjudgment of high wear under a light load scene) and missing report (such as high wear risk is not identified under a heavy load scene) are reduced. With reference to the first aspect, in some possible implementation manners, after outputting the life prediction result of the vehicle tire, the method further includes performing early warning reminding when the life prediction result of the vehicle tire meets a preset early warning condition. Through the technical scheme, the HUT voice prompt and the instrument indicator lamp at the vehicle end give an alarm, the double sensory warning ensures that a driver can timely sense the tire risk under the complex road condi