Method and System for Vehicle Maintenance Information Recommendation Based on Big Data

Abstract

The present invention relates to the field of vehicle management, and discloses a method and system for recommending vehicle maintenance information based on big data. The method includes: first, collecting vehicle driving data and vehicle owner behavior data in real time through an in-vehicle system; then, obtaining a vehicle maintenance coefficient by processing the vehicle driving data and vehicle owner behavior data; associating and calculating a trend coefficient by combining the difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle with the vehicle maintenance coefficient, comparing the trend coefficient with a preset trend coefficient range, and judging whether to adjust the next maintenance cycle according to the comparison result; if it is judged to extend the next maintenance cycle, obtaining the environmental information of the vehicle and processing it to obtain a vehicle environmental loss coefficient; comparing the vehicle environmental loss coefficient with a vehicle environmental loss threshold to judge whether the vehicle is affected by the environment; if it is judged to be affected by the environment, recalibrating the next maintenance cycle of the vehicle; finally, determining the next maintenance time according to the next maintenance cycle and pushing the maintenance time to the vehicle owner.

Description

Technical Field

[0001] The present invention relates to the field of vehicle management, and particularly to a method and system for recommending vehicle maintenance information based on big data. Background Art

[0002] Vehicle maintenance plays a crucial role in extending the service life of vehicles and ensuring the safe and efficient operation of vehicles. On the one hand, vehicle maintenance involves detecting various components of the vehicle, such as brakes and tires, to ensure vehicle safety. On the other hand, it can replace engine oil, oil filters, and worn components to prevent small problems from becoming big problems and even evolving into safety accidents endangering the driver.

[0003] Currently, the existing vehicle maintenance time generally uses the set driving mileage value or the interval time period as the basis for pushing the next maintenance time. However, due to the different environments of different drivers and vehicles, for example, the driving wear of vehicles in areas with sudden temperature changes, long-term humid areas, or areas with more dust is different. Therefore, there is a large gap between the actual vehicle maintenance cycle and the set maintenance cycle, resulting in a high inefficiency in pushing maintenance information, seriously affecting the actual vehicle driving safety and driver satisfaction. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for recommending vehicle maintenance information based on big data, so as to solve the technical problem that due to the different environments of different drivers and vehicles, for example, the driving wear of vehicles in areas with sudden temperature changes, long-term humid areas, or areas with more dust is different. Therefore, there is a large gap between the actual vehicle maintenance cycle and the set maintenance cycle, resulting in a high inefficiency in pushing maintenance information, seriously affecting the actual vehicle driving safety and driver satisfaction.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for recommending vehicle maintenance information based on big data includes:

[0007] S1: Real-time collect vehicle driving data and vehicle owner behavior data through an in-vehicle system; the vehicle driving data includes driving mileage, driving speed, and braking frequency, and the vehicle owner behavior data includes driving habits, driving routes, and maintenance habits;

[0008] S2: Receive the vehicle driving data and vehicle owner behavior data, and process the vehicle driving data and vehicle owner behavior data to obtain a vehicle maintenance coefficient;

[0009] S3: Associate the difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle with the vehicle maintenance coefficient, calculate to obtain a trend coefficient, compare the trend coefficient with the preset trend coefficient interval, and judge whether to shorten the next maintenance cycle of the vehicle, maintain the vehicle maintenance cycle, or extend the next maintenance cycle of the vehicle according to the comparison result;

[0010] S4: Determine the next maintenance time of the current vehicle according to the next maintenance cycle of the current vehicle, and push the maintenance time to the vehicle owner.

[0011] As a further technical solution, the process of calculating and obtaining the trend coefficient is:

[0012] Through the formula:

[0013]

[0014]

[0015] Calculate to obtain the trend coefficient F i , Q i is the deviation coefficient, D j实 is the actual maintenance mileage value corresponding to the j-th sampling time point, is the average value of the actual maintenance mileage, m is the number of sampling time points, D 实 (t) is the curve of the actual maintenance mileage value changing with time, D 理 (t) is the curve of the theoretical maintenance mileage value changing with time, R i is the vehicle maintenance coefficient, α, β are preset weight factors, t0 is the starting time point, and t1 is the ending time point.

[0016] As a further technical solution, the

[0017] In the formula, X i is the vehicle driving score coefficient, Y i is the vehicle owner behavior score coefficient, γ, ε are preset proportionality coefficients;

[0018] Among them,

[0019] In the formula, x k is the value of the k-th vehicle driving data item, δ k is the preset proportionality coefficient of the k-th vehicle driving data item, n is the number of vehicle driving data items, is the average value of the number of times the vehicle speed change exceeds the preset value v0 within the Δt time period during each historical driving; is the average value of the ratio of the duration when the vehicle speed remains within v i ±5 to the driving duration during each historical driving; Where h is the vehicle maintenance frequency, N is the average number of vehicle parts replaced in each historical maintenance, τ is the dimensionless coefficient, and y1, y2, and y3 are reference coefficients.

[0020] As a further technical solution, the trend coefficient F i is compared with the preset trend coefficient interval [F ilown , F iup in the following process:

[0021] If F i ∈[F iup , +∞], it is determined that the current vehicle loss increases, and the next vehicle maintenance cycle is shortened;

[0022] If F i ∈[F ilown , F iup , it is determined that the current vehicle loss is normal, and the next vehicle maintenance cycle is maintained;

[0023] If F i <F ilown , it is determined that the current vehicle loss decreases, and the next vehicle maintenance cycle is extended.

[0024] As a further technical solution, the following is also included in S3:

[0025] If it is determined to extend the next vehicle maintenance cycle, the environmental information of the vehicle location is obtained, and the vehicle environmental loss coefficient is obtained after processing the environmental information;

[0026] The vehicle environmental loss coefficient is compared with the vehicle environmental loss threshold. If the vehicle environmental loss coefficient exceeds the vehicle environmental loss threshold, it is determined that the vehicle maintenance is affected by the environment, otherwise, it is determined that the vehicle maintenance is not affected by the environment;

[0027] If it is determined that the vehicle maintenance is affected by the environment, the next vehicle maintenance cycle is re-calibrated according to the vehicle environmental loss coefficient and the trend coefficient.

[0028] As a further technical solution, the environmental information includes the number of rainfall times, the number of acid rain times, the duration of the environmental temperature below 0 degrees Celsius, the duration of the environmental temperature above 30 degrees Celsius, the content of particulate matter in the air, and the ratio of the mileage of non-urban roads traveled to the total mileage traveled from the previous maintenance time to the current time. The process of obtaining the vehicle environmental loss coefficient after processing the environmental information is realized based on the trained neural network model.

[0029] As a further technical solution, the process of comparing the vehicle environmental loss coefficient with the vehicle environmental loss threshold is as follows:

[0030] The vehicle environmental loss coefficient P isun is compared with the vehicle environmental loss threshold Pi0 Make a comparison;

[0031] If P isun ≥P i0 , it is determined that vehicle maintenance is affected by the environment;

[0032] If P isun <P i0 , it is determined that vehicle maintenance is not affected by the environment.

[0033] As a further technical solution, the process of recalibrating the next vehicle maintenance cycle according to the vehicle environmental loss coefficient and the trend coefficient is as follows:

[0034] Adjust the extended vehicle maintenance cycle according to

[0035] ;

[0036] In the formula, is the preset standard coefficient, and M 延 is the extended vehicle maintenance cycle.

[0037] A vehicle maintenance information recommendation system based on big data, which is applicable to the above-mentioned method for recommending vehicle maintenance information based on big data.

[0038] Advantages of the present invention:

[0039] (1) First, the present invention collects vehicle driving data and owner behavior data through an in-vehicle system, then obtains a vehicle maintenance coefficient according to the vehicle driving data and owner behavior data, associates the difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle with the vehicle maintenance coefficient, calculates to obtain a trend coefficient, compares the trend coefficient with the preset trend coefficient interval, and determines whether to shorten the next vehicle maintenance cycle, maintain the vehicle maintenance cycle, or extend the next vehicle maintenance cycle according to the comparison result; subsequently, determines the next vehicle maintenance time according to the current vehicle's next maintenance cycle, and pushes the maintenance time to the owner;

[0040] (2) The present invention can linearly improve the efficiency of the maintenance information finally pushed to the owner. Compared with the technology of setting a regular maintenance cycle and giving regular reminders without dynamically adjusting according to the owner behavior data and vehicle driving data, the present invention can predict in real time according to the owner's driving behavior and vehicle driving data, timely predict the loss trend of the vehicle before the next maintenance, and dynamically adjust the vehicle maintenance cycle according to the loss trend, so as to combine the owner's maintenance habits and the actual needs of the vehicle to achieve the purpose of effective push. Description of the Drawings

[0041] The present invention will be further described below with reference to the accompanying drawings.

[0042] Figure 1 This is the flowchart of the method of the present invention. Detailed implementation manners

[0043] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0044] Please refer to Figure 1 As shown, the present invention is a method for recommending vehicle maintenance information based on big data, including:

[0045] S1: Real-time collect vehicle driving data and owner behavior data through an in-vehicle system; the vehicle driving data includes driving mileage, driving speed, and braking frequency, and the owner behavior data includes driving habits, driving routes, and maintenance habits;

[0046] S2: Receive the vehicle driving data and owner behavior data, and process the vehicle driving data and owner behavior data to obtain a vehicle maintenance coefficient;

[0047] S3: Correlate the difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle with the vehicle maintenance coefficient, calculate to obtain a trend coefficient, compare the trend coefficient with a preset trend coefficient interval, and judge whether to shorten the next vehicle maintenance cycle, maintain the vehicle maintenance cycle, or extend the next vehicle maintenance cycle according to the comparison result;

[0048] S4: Determine the next maintenance time of the current vehicle according to the next maintenance cycle of the current vehicle, and push the maintenance time to the vehicle owner.

[0049] The process of calculating the trend coefficient is as follows:

[0050] Through the formula:

[0051]

[0052]

[0053] Calculate to obtain the trend coefficient F i , Q i is the deviation coefficient, D j实 is the actual maintenance mileage value corresponding to the jth sampling time point, is the average value of the actual maintenance mileage, m is the number of sampling time points, D 实 (t) is the curve of the actual maintenance mileage value changing with time, D 理 (t) is the curve of the theoretical maintenance mileage value changing with time, Ri is the vehicle maintenance coefficient, α and β are preset weight factors, t0 is the starting time point, and t1 is the ending time point.

[0054] In this embodiment, first, the vehicle driving data and the owner behavior data are collected through the in-vehicle system, then the vehicle maintenance coefficient is obtained by processing the vehicle driving data and the owner behavior data. The difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle is associated with the vehicle maintenance coefficient to calculate the trend coefficient. The trend coefficient is compared with the preset trend coefficient interval, and according to the comparison result, it is judged whether to shorten the next vehicle maintenance cycle, maintain the vehicle maintenance cycle, or extend the next vehicle maintenance cycle; subsequently, according to the next vehicle maintenance cycle of the current vehicle, the next vehicle maintenance time is determined and pushed to the owner; through the above technical solution, the efficiency of the maintenance information finally pushed to the owner can be directly improved. Compared with the technology of setting a regular maintenance cycle and not dynamically adjusting according to the owner behavior data and vehicle driving data, the present invention can predict in real time according to the driving behavior of the owner and the driving data of the vehicle, timely predict the loss trend of the vehicle before the next maintenance, and dynamically adjust the vehicle maintenance cycle according to the loss trend, so as to integrate the owner's maintenance habits and the actual needs of the vehicle to achieve the purpose of effective push;

[0055] Among them, the method for obtaining the trend coefficient is as follows:

[0056] First, through the formula calculate to obtain Q i , and then substitute it into the following formula calculate to obtain the trend coefficient F i ; Obviously, through the above formula, it can be known that the larger Q i is, the greater the deviation between the actual maintenance mileage value and the preset maintenance mileage value change curve, indicating that the volatility of vehicle loss is greater, and the finally obtained trend coefficient is greater; The larger it is, the greater the dispersion degree of the actual maintenance mileage value of the current vehicle, and the greater the trend coefficient. Through the above technical solution, the loss trend of the current vehicle can be more accurately reflected.

[0057] The

[0058] In the formula, X i is the vehicle driving score coefficient, X i is the owner behavior score coefficient, and γ and ε are preset proportionality coefficients;

[0059] Among them,

[0060] In the formula, x k is the value of the kth vehicle driving data, and δ kδ is the average value of the number of times the vehicle speed change exceeds the preset value v0 within the time period Δt during each historical driving process; is the average value of the ratio of the duration during which the vehicle speed remains at v i ±5 to the driving duration during each historical driving process; In the formula, h is the vehicle maintenance frequency, N is the average value of the number of vehicle parts replaced during each historical maintenance, τ is the dimensionless coefficient, and y1, y2, and y3 are reference coefficients.

[0061] The process of comparing the trend coefficient F i with the preset trend coefficient interval [F ilown , F iup is as follows:

[0062] If F i ∈[F iup , +∞], it is determined that the current vehicle wear increases, and the next vehicle maintenance cycle is shortened;

[0063] If F i ∈[F ilown , F iup , it is determined that the current vehicle wear is normal, and the next vehicle maintenance cycle is maintained;

[0064] If F i <F ilown , it is determined that the current vehicle wear decreases, and the next vehicle maintenance cycle is extended.

[0065] In this embodiment, a method for obtaining the vehicle maintenance coefficient is provided. Specifically, the vehicle maintenance coefficient R is calculated through the formula i , where ; in the formula, x k is the value of the k-th vehicle driving data, δ k is the preset proportionality coefficient of the k-th vehicle driving data, n is the number of items of vehicle driving data, is the average value of the number of times the vehicle speed change exceeds the preset value v0 within the time period Δt during each historical driving process; is the average value of the ratio of the duration during which the vehicle speed remains at v i ±5 to the driving duration during each historical driving process; Where h is the maintenance frequency of the vehicle, N is the average number of vehicle parts replaced in each historical maintenance, τ is the dimensionless coefficient, y1, y2, y3 are reference coefficients; where the vehicle-connected driving data includes but is not limited to mileage, driving speed, braking frequency, etc. The above parameters can be obtained through sensors or monitoring equipment and will not be repeated here. Obviously, if the parameter value of each item changes, the final vehicle driving score coefficient X i The higher it is, the greater the vehicle's driving loss is, so the vehicle maintenance factor is greater; similarly, , , i They represent the driving habits of the car owner. The larger the value, the worse the driving habits of the car owner, the greater the damage to the vehicle, and therefore the greater the vehicle maintenance coefficient. By combining the above formulas, the vehicle maintenance coefficient can be accurately calculated and dynamically adjusted to achieve the final trend coefficient F. i The size of the impact.

[0066] The S3 also includes:

[0067] If it is determined that the next maintenance cycle of the vehicle is to be extended, the environmental information of the location of the vehicle is obtained, and the environmental loss coefficient of the vehicle is obtained after the environmental information is processed;

[0068] The vehicle environmental loss coefficient is compared with the vehicle environmental loss threshold, if the vehicle environmental loss coefficient exceeds the vehicle environmental loss threshold, it is determined that the vehicle maintenance is affected by the environment, otherwise, it is determined that the vehicle maintenance is not affected by the environment;

[0069] If it is determined that vehicle maintenance is affected by the environment, the next vehicle maintenance cycle will be recalibrated based on the vehicle environmental loss coefficient and trend coefficient.

[0070] The environmental information includes the number of rainfalls, acid rains, duration of ambient temperature below 0 degrees Celsius, duration of ambient temperature above 30 degrees Celsius, particulate matter content in the air, and the ratio of mileage of non-urban roads to mileage, and the process of obtaining the vehicle environmental loss coefficient after processing the environmental information based on the trained neural network model. It should be noted that the above environmental information can be obtained through the vehicle system or in combination with the local weather. The specific acquisition method is conventional technology and will not be described in detail.

[0071] The process of comparing the vehicle environmental loss coefficient with the vehicle environmental loss threshold is:

[0072] The vehicle environmental loss coefficient P isun and vehicle environmental loss threshold P i0 Make comparisons;

[0073] If P isum≥P i0 , it is determined that vehicle maintenance is affected by the environment;

[0074] If P isum <P i0 , it is determined that vehicle maintenance is not affected by the environment.

[0075] The process of recalibrating the next vehicle maintenance cycle according to the vehicle environmental loss coefficient and the trend coefficient is as follows:

[0076] Adjust the extended vehicle maintenance cycle according to ;

[0077] In the formula, is the preset standard coefficient, and M 延 is the extended vehicle maintenance cycle.

[0078] In this embodiment, in order to avoid extending the maintenance cycle only based on vehicle driving data and owner behavior data, resulting in excessive wear due to environmental influence and inability to ensure safety, the environmental information is used to accurately judge whether the vehicle driving is affected by environmental factors and causes increased loss. The vehicle environmental loss coefficient P isun is compared with the vehicle environmental loss threshold P i0 ; if P isun ≥P i0 , it is determined that vehicle maintenance is affected by the environment; if P isun <P i0 , it is determined that vehicle maintenance is not affected by the environment; once it is determined that there is vehicle environmental loss, it is adjusted after calculation through the formula to ensure the rationality of the next maintenance cycle and improve the efficiency of pushing maintenance information;

[0079] It should be noted that determining the vehicle maintenance cycle and extending or shortening the vehicle maintenance cycle are all determined through comprehensive analysis of historical data and empirical data, which can be directly obtained by the prior art and will not be elaborated here.

[0080] A vehicle maintenance information recommendation system based on big data, which is applicable to the method for recommending vehicle maintenance information based on big data described above.

[0081] The above has described a specific embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for recommending vehicle maintenance information based on big data, characterized in that: include: S1: Real-time collection of vehicle driving data and owner behavior data through the vehicle system; The vehicle driving data includes driving mileage, driving speed, and braking frequency, and the owner behavior data includes driving habits, driving routes, and maintenance habits; S2: receiving vehicle driving data and vehicle owner behavior data, and obtaining a vehicle maintenance coefficient according to the vehicle driving data and vehicle owner behavior data; S3: Associating the difference between the preset maintenance mileage value and the actual maintenance mileage value of the current vehicle with the vehicle maintenance coefficient, calculating a trend coefficient, comparing the trend coefficient with a preset trend coefficient range, and judging whether to shorten the vehicle's next maintenance cycle, maintain the vehicle's maintenance cycle, or extend the vehicle's next maintenance cycle based on the comparison result; S4: Determine the next maintenance time of the current vehicle according to the next maintenance cycle of the current vehicle, and push the maintenance time to the owner; The process of calculating the trend coefficient is: By formula: Calculate the trend coefficient , is the coefficient of variation, is the actual maintenance mileage value corresponding to the jth sampling time point, is the actual maintenance mileage mean, m is the number of sampling time points, is the curve of actual maintenance mileage changing with time, is the curve of theoretical maintenance mileage changing with time, is the vehicle maintenance factor, , is the preset weight factor, is the starting time point, is the end time point; Said ; In the formula, is the vehicle driving rating coefficient, is the owner behavior scoring coefficient, , is the preset proportional coefficient; in, ; In the formula is the value of the kth vehicle driving data, is the preset proportional coefficient of the kth vehicle driving data, n is the number of vehicle driving data items, The speed of the vehicle during each driving process in history The vehicle speed change during the time period exceeds the preset value The average of the number of times; The speed of the vehicle is maintained at The average value of the ratio of the duration of driving to the driving duration; , where h is the maintenance frequency of the vehicle, N is the average number of vehicle parts replaced in each maintenance in history, is the dimensionless coefficient, , , is the reference coefficient.

2. The method for recommending vehicle maintenance information based on big data according to claim 1, characterized in that: The trend coefficient With the preset trend coefficient interval The comparison process is: like ∈ , it is judged that the current vehicle loss is increasing, and the next vehicle maintenance cycle is shortened; like ∈ , then the current vehicle loss is judged to be normal and the vehicle is maintained at the next maintenance cycle; like < , it is judged that the current vehicle loss is reduced and the next vehicle maintenance cycle is extended.

3. The method for recommending vehicle maintenance information based on big data according to claim 1, characterized in that: The S3 also includes: If it is determined that the next maintenance cycle of the vehicle is to be extended, the environmental information of the location of the vehicle is obtained, and the environmental loss coefficient of the vehicle is obtained after the environmental information is processed; The vehicle environmental loss coefficient is compared with the vehicle environmental loss threshold, if the vehicle environmental loss coefficient exceeds the vehicle environmental loss threshold, it is determined that the vehicle maintenance is affected by the environment, otherwise, it is determined that the vehicle maintenance is not affected by the environment; If it is determined that vehicle maintenance is affected by the environment, the next vehicle maintenance cycle will be recalibrated based on the vehicle environmental loss coefficient and trend coefficient.

4. The method for recommending vehicle maintenance information based on big data according to claim 3, characterized in that: The environmental information includes the number of rainfalls and acid rains from the last maintenance time to the current time, the length of time the ambient temperature is below 0 degrees Celsius, the length of time the ambient temperature is above 30 degrees Celsius, the content of particulate matter in the air, and the ratio of the mileage of non-urban roads to the mileage. The process of obtaining the vehicle environmental loss coefficient after processing the environmental information is realized based on the trained neural network model.

5. The method for recommending vehicle maintenance information based on big data according to claim 3, characterized in that: The process of comparing the vehicle environmental loss coefficient with the vehicle environmental loss threshold is: Vehicle environmental loss factor Vehicle environmental loss threshold Make comparisons; like ≥ , then it is judged that vehicle maintenance is affected by the environment; like < , it is judged that vehicle maintenance is affected by the environment.

6. The method for recommending vehicle maintenance information based on big data according to claim 5, characterized in that: The process of recalibrating the vehicle's next maintenance cycle based on the vehicle's environmental loss coefficient and trend coefficient is as follows: Make adjustments; In the formula, is the preset standard coefficient, For extended vehicle maintenance cycles.

7. The vehicle maintenance information recommendation system based on big data is characterized by: The system is applicable to the method for recommending vehicle maintenance information based on big data as described in any one of claims 1 to 6.

Images