Automobile endurance mileage prediction method and device
By constructing a probability density distribution model and a grey prediction model, and combining them with weighting coefficients, the problem of inaccurate range prediction in existing technologies has been solved, and accurate and reliable range prediction for electric vehicles and hybrid vehicles has been achieved.
Patent Information
- Application Number
- CN202411142186.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-22
AI Technical Summary
In the existing technology, the range prediction methods for electric vehicles and hybrid vehicles fail to accurately consider the vehicle's historical driving conditions, resulting in a large deviation between the estimated results and the actual situation, and failing to provide accurate and reliable remaining range information.
By constructing a probability density distribution model and a grey prediction model for driving mileage per unit energy consumption, and combining weighting coefficients, the prediction results are dynamically adjusted to take into account factors such as driving habits and road conditions, achieving a balance between stability and sensitivity.
It achieves accurate and reliable prediction of remaining driving range, balancing the stability and sensitivity of the prediction, and provides users with accurate remaining driving range information.
Smart Images

Figure CN118991441A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of battery management technology, and more specifically, to a method and device for predicting vehicle range. Background Art
[0002] Currently, in the field of electric vehicles and hybrid vehicles, accurately predicting and displaying the remaining mileage is an important function to improve user experience, plan itineraries, and optimize energy management. In the prior art, the remaining mileage is usually estimated based on the battery state of charge. The SOC value of the battery is measured in real time by a sensor, that is, the percentage of the battery's current remaining power to its total capacity, and a simple calculation is performed in combination with the rated capacity of the battery to obtain an estimate of the remaining mileage. However, in practice, it is found that the complex and changeable driving conditions during actual driving have a significant impact on the vehicle's energy consumption. The existing methods do not take into account the vehicle's historical driving conditions, which leads to a large deviation between the estimated results and the actual situation, and it is impossible to provide users with accurate and reliable remaining mileage information. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide a method and device for predicting the remaining cruising range of an automobile, which can accurately predict the remaining cruising range, take into account the stability and sensitivity of the prediction, and provide users with accurate and reliable remaining cruising range information.
[0004] The first aspect of the present application provides a method for predicting the cruising range of an automobile, comprising:
[0005] Obtain the target vehicle's driving history data and the pre-built grey prediction model;
[0006] Constructing a probability density distribution model of unit energy consumption mileage according to the driving history data;
[0007] Acquire current vehicle data of the target vehicle;
[0008] Perform prediction based on the probability density distribution model and the current vehicle data to obtain a first range prediction value;
[0009] Perform prediction based on the grey prediction model and the current vehicle data to obtain a second range prediction value;
[0010] The final cruising range prediction result is calculated based on the pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value.
[0011] In the above implementation process, the target vehicle's driving history data and the pre-built gray prediction model are first obtained; then the probability density distribution model of the unit energy consumption driving mileage is constructed based on the driving history data; in actual application, the target vehicle's current vehicle data is first obtained; then the first range prediction value is obtained based on the probability density distribution model and the current vehicle data; then the second range prediction value is obtained based on the gray prediction model and the current vehicle data; finally, the final range prediction result is calculated based on the pre-calculated weight coefficient, the first range prediction value, and the second range prediction value. It can be seen that this method can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0012] Furthermore, constructing a probability density distribution model of unit energy consumption mileage according to the driving history data includes:
[0013] Preprocessing the driving history data to obtain preprocessed data;
[0014] Based on the preprocessed data, construct a functional relationship between the driving mileage and the accumulated power consumption;
[0015] Calculate the functional relationship between unit energy consumption and power according to the functional relationship between the driving mileage and the accumulated power consumption;
[0016] Discretely sampling the functional relationship between the unit energy consumption and the power consumption to obtain sampling data;
[0017] Performing data statistics on the sampled data to obtain statistical results;
[0018] A probability density distribution model of unit energy consumption mileage is constructed based on the statistical results.
[0019] Furthermore, the preprocessing of the driving history data to obtain preprocessed data includes:
[0020] Performing verification processing on the driving history data to obtain first processed data;
[0021] Screening the first processed data to obtain screened second processed data;
[0022] performing cleaning processing on the second processed data to obtain cleaned third processed data;
[0023] The third processed data is processed to remove abnormal values and error values to obtain pre-processed data.
[0024] Furthermore, before calculating the final cruising range prediction result according to the pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value, the method further includes:
[0025] Acquire historical prediction data; wherein the historical prediction data includes a final cruising range prediction result predicted at a first moment, a final cruising range prediction result predicted at a second moment, and an actual vehicle mileage from the first moment to the second moment; wherein the second moment is reached after a preset time period from the first moment;
[0026] Calculating a prediction error based on the historical prediction data;
[0027] Calculating an error average according to the prediction error;
[0028] A weight coefficient is calculated according to the error average value.
[0029] Furthermore, the method further comprises:
[0030] Acquire new sample data of the target vehicle's travel;
[0031] Calculating the data weight corresponding to the new driving sample data by using a pre-constructed weight function;
[0032] Calculating new model parameters of the probability density distribution model according to the new driving sample data;
[0033] The probability density distribution model is modified according to the data weight and the new model parameters to obtain a modified target probability density distribution model.
[0034] Furthermore, the parameter correction of the probability density distribution model according to the data weight and the new model parameter to obtain a corrected target probability density distribution model includes:
[0035] Obtaining old model parameters of the probability density distribution model;
[0036] Perform parameter fusion processing on the old model parameters and the new model parameters according to the data weights to obtain target model parameters;
[0037] The model parameters of the probability density distribution model are updated to the target model parameters to obtain a revised target probability density distribution model.
[0038] A second aspect of the present application provides a vehicle cruising range prediction device, the vehicle cruising range prediction device comprising:
[0039] A first acquisition unit is used to acquire the driving history data of the target vehicle and a pre-built grey prediction model;
[0040] A construction unit, used to construct a probability density distribution model of unit energy consumption mileage according to the driving history data;
[0041] A second acquisition unit, used to acquire current vehicle data of the target vehicle;
[0042] A first prediction unit, configured to perform prediction according to the probability density distribution model and the current vehicle data to obtain a first range prediction value;
[0043] A second prediction unit, configured to perform prediction based on the grey prediction model and the current vehicle data to obtain a second range prediction value;
[0044] The first calculation unit is used to calculate a final cruising range prediction result according to a pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value.
[0045] In the above implementation process, the first acquisition unit acquires the target vehicle's driving history data and a pre-constructed gray prediction model; then the construction unit constructs a probability density distribution model of unit energy consumption mileage based on the driving history data; in actual applications, the second acquisition unit first acquires the target vehicle's current vehicle data; then, the first prediction unit predicts based on the probability density distribution model and the current vehicle data to obtain a first range prediction value; then, the second prediction unit predicts based on the gray prediction model and the current vehicle data to obtain a second range prediction value; finally, the first calculation unit calculates the final range prediction result based on the pre-calculated weight coefficient, the first range prediction value, and the second range prediction value. It can be seen that the device can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0046] Furthermore, the construction unit comprises:
[0047] A preprocessing subunit, used for preprocessing the driving history data to obtain preprocessed data;
[0048] A first construction subunit is used to construct a functional relationship between the driving mileage and the accumulated power consumption based on the pre-processed data;
[0049] A calculation subunit, used for calculating the functional relationship between unit energy consumption and power consumption according to the functional relationship between the driving mileage and the accumulated power consumption;
[0050] A sampling subunit, used for discretely sampling the functional relationship between the unit energy consumption and the power consumption to obtain sampling data;
[0051] A statistical subunit, used for performing data statistics on the sampled data to obtain statistical results;
[0052] The second construction subunit is used to construct a probability density distribution model of unit energy consumption mileage based on the statistical results.
[0053] Furthermore, the preprocessing subunit includes:
[0054] A verification module, used for performing verification processing on the driving history data to obtain first processed data;
[0055] A screening module, used for screening the first processed data to obtain screened second processed data;
[0056] a cleaning module, used for cleaning the second processed data to obtain cleaned third processed data;
[0057] The removal module is used to remove abnormal values and error values from the third processed data to obtain pre-processed data.
[0058] Furthermore, the vehicle cruising range prediction device further includes:
[0059] a third acquisition unit, configured to acquire historical prediction data before calculating the final cruising range prediction result according to the pre-calculated weight coefficient, the first cruising range prediction value, and the second cruising range prediction value; wherein the historical prediction data includes the final cruising range prediction result predicted at the first moment, the final cruising range prediction result predicted at the second moment, and the actual vehicle mileage from the first moment to the second moment; wherein the second moment is reached after a preset time period of the first moment;
[0060] A second calculation unit, used for calculating a prediction error based on the historical prediction data;
[0061] The second calculation unit is further used to calculate an error average value according to the prediction error;
[0062] The second calculation unit is further used to calculate a weight coefficient according to the error average value.
[0063] Furthermore, the vehicle cruising range prediction device further includes:
[0064] A fourth acquisition unit, used to acquire new driving sample data of the target vehicle;
[0065] A third calculation unit, used for calculating the data weight corresponding to the new driving sample data by using a pre-built weight function;
[0066] The third calculation unit is further used to calculate new model parameters of the probability density distribution model according to the new driving sample data;
[0067] A correction unit is used to correct the parameters of the probability density distribution model according to the data weight and the new model parameters to obtain a corrected target probability density distribution model.
[0068] Furthermore, the correction unit comprises:
[0069] An acquisition subunit, used for acquiring old model parameters of the probability density distribution model;
[0070] A fusion subunit, configured to perform parameter fusion processing on the old model parameters and the new model parameters according to the data weights to obtain target model parameters;
[0071] The updating subunit is used to update the model parameters of the probability density distribution model to the target model parameters to obtain a revised target probability density distribution model.
[0072] A third aspect of the present application provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the vehicle range prediction method described in any one of the first aspects of the present application.
[0073] A fourth aspect of the present application provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the vehicle range prediction method described in any one of the first aspect of the present application is executed. BRIEF DESCRIPTION OF THE DRAWINGS
[0074] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0075] Figure 1 A flow chart of a method for predicting vehicle cruising range provided in an embodiment of the present application;
[0076] Figure 2 A schematic diagram of a flow chart of another method for predicting the cruising range of an automobile provided in an embodiment of the present application;
[0077] Figure 3 An example diagram of a functional relationship between mileage and cumulative power consumption and a functional relationship between unit energy consumption and power provided in an embodiment of the present application;
[0078] Figure 4A schematic diagram of the structure of a vehicle range prediction device provided in an embodiment of the present application;
[0079] Figure 5 A schematic diagram of the structure of another vehicle range prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0080] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application.
[0081] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0082] Example 1
[0083] Please see Figure 1 , Figure 1 A flow chart of a method for predicting the cruising range of a vehicle provided in this embodiment. The method for predicting the cruising range of a vehicle includes:
[0084] S101, obtaining the driving history data of the target vehicle and a pre-built grey prediction model.
[0085] In this embodiment, a large amount of driving history data is collected from the vehicle's onboard computer or related sensors, including the speed, mileage, power consumption, real-time voltage, current, etc. of each trip.
[0086] S102: construct a probability density distribution model of unit energy consumption mileage according to the driving history data.
[0087] In this embodiment, the method extracts the mileage that can be traveled per unit energy consumption (for example, per 1 kwh) based on historical driving data, performs mathematical statistics on the mileage, and constructs a probability density distribution model of the mileage per unit energy consumption.
[0088] In this embodiment, the cruising range is affected by many factors, such as driving habits, road conditions, temperature, etc. It is uncertain in time and space, the degree of influence is uncertain, the main influencing factors are uncertain, and the impact of various factors is also uncertain. Therefore, it is difficult to accurately model and calculate, and it is difficult for general models to cover all changes. However, the above factors will eventually be comprehensively reflected in the driving process. The probability density model constructed by this method based on the probability statistics of mileage per unit energy consumption can well contain the information contained in the above factors.
[0089] S103: Acquire current vehicle data of the target vehicle.
[0090] S104: Predict according to the probability density distribution model and current vehicle data to obtain a first range prediction value.
[0091] S105 . Predict according to the grey prediction model and current vehicle data to obtain a second range prediction value.
[0092] S106, calculating a final cruising range prediction result according to a pre-calculated weight coefficient, the first cruising range prediction value, and the second cruising range prediction value.
[0093] In this embodiment, the method constructs a probability density model of unit energy consumption mileage through historical driving data to predict the remaining mileage when driving habits change significantly; on the other hand, the gray prediction model is used to predict the remaining mileage when driving habits change significantly; the above two are weighted averaged, and the weight coefficient is dynamically adjusted according to the error of the probability density prediction model. Both stability and sensitivity are taken into account.
[0094] In this embodiment, the execution subject of the method may be a computing device such as a computer or a server, and no limitation is made in this embodiment.
[0095] In this embodiment, the execution subject of the method may also be a smart device such as a smart phone, a tablet computer, etc., which is not limited in this embodiment.
[0096] It can be seen that the vehicle range prediction method described in this embodiment can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0097] Example 2
[0098] Please see Figure 2 , Figure 2 A flow chart of a method for predicting the cruising range of a vehicle provided in this embodiment. The method for predicting the cruising range of a vehicle includes:
[0099] S201, obtaining the driving history data of the target vehicle and a pre-built grey prediction model.
[0100] In this embodiment, the process of constructing the grey prediction model includes the following steps:
[0101] Acquire the endurance test data of the electric vehicle, wherein the endurance test data includes the corresponding travel mileage of the electric vehicle under an isometric charging state;
[0102] Generate a first vector according to the endurance test data;
[0103] Adding the elements in the first vector in sequence to obtain a plurality of accumulated values, and concatenating the plurality of accumulated values to obtain a second vector;
[0104] sequentially obtaining adjacent average values of elements in the second vector to obtain a plurality of adjacent average values, and concatenating the plurality of adjacent average values to obtain a third vector;
[0105] constructing a differential equation based on the first vector and the third vector;
[0106] Solving the differential equation to obtain a solution formula;
[0107] The solution formula is determined as a grey prediction model.
[0108] S202: Preprocess the driving history data to obtain preprocessed data.
[0109] As an optional implementation, the driving history data is preprocessed to obtain preprocessed data, including:
[0110] Performing verification processing on the driving history data to obtain first processed data;
[0111] Screening the first processed data to obtain screened second processed data;
[0112] Cleaning the second processed data to obtain cleaned third processed data;
[0113] The third processed data is processed to remove abnormal values and error values to obtain preprocessed data.
[0114] S203: constructing a functional relationship between the driving mileage and the accumulated power consumption based on the preprocessed data.
[0115] S204: Calculate the functional relationship between unit energy consumption and power according to the functional relationship between the driving mileage and the accumulated power consumption.
[0116] In this embodiment, for each section of valid driving data in the preprocessed data, a functional relationship between the driving mileage and the accumulated power consumption is constructed, and then the functional relationship between the unit energy consumption and the power consumption is obtained by numerically derivatizing the function.
[0117] Please also read Figure 3 , Figure 3 1 is an example diagram of a functional relationship between mileage and cumulative power consumption and a functional relationship between unit energy consumption and power provided by an embodiment of the present application. Figure 3 As shown, the left coordinate of the above figure is the mileage, and the right side has the numerical derivative of the mileage with respect to the cumulative power consumption.
[0118] S205 , discretely sample the functional relationship between unit energy consumption and power consumption to obtain sampling data.
[0119] S206: Perform data statistics on the sampled data to obtain statistical results.
[0120] S207: construct a probability density distribution model of unit energy consumption mileage based on the statistical results.
[0121] In this embodiment, it is assumed that the unit energy consumption mileage obeys a certain probability distribution (such as normal distribution, log-normal distribution, etc.). The maximum likelihood estimation or other parameter estimation methods can be used to estimate the parameters of the distribution.
[0122] In this embodiment, taking the normal distribution as an example (but not limited to the normal distribution), it is assumed that the unit energy consumption mileage X follows the normal distribution N(u, σ 2 ), then the probability density distribution model is:
[0123]
[0124] Where u is the mean and σ is the standard deviation. u and σ can be estimated based on statistical results and sampled data.
[0125] In this embodiment, through the probability density model based on the unit energy consumption mileage, many uncertain and time-space changing influencing factors such as driving habits, road conditions, temperature, etc. can be comprehensively considered, so as to more accurately predict the remaining mileage when the driving habits have not changed significantly.
[0126] S208. Acquire current vehicle data of the target vehicle.
[0127] S209: Predict according to the probability density distribution model and the current vehicle data to obtain a first range prediction value.
[0128] In this embodiment, it is known that the probability density distribution model of the current unit energy consumption mileage is N(u, σ 2 ), according to the current vehicle data, the current remaining power is E (unit: kWh), and the expected value of the cruising range (i.e., the first cruising range prediction value S1) is predicted by the following formula, and the possible range of the cruising range can be determined by setting a confidence interval (such as a 95% confidence interval).
[0129] S1=u×E.
[0130] S210: Predict according to the grey prediction model and current vehicle data to obtain a second range prediction value.
[0131] In this embodiment, the weighted prediction is performed in combination with the grey prediction model, which solves the problem of poor sensitivity when the driving conditions change significantly based only on the probability density model, and takes into account both the stability and sensitivity of the prediction.
[0132] In this embodiment, the current state of charge can be obtained according to the current vehicle data; then the remaining mileage (ie, the second endurance prediction value) can be obtained based on the above solution formula (ie, the gray prediction model) and the current state of charge.
[0133] As an optional implementation, before step S215, the following steps are also included:
[0134] S211. Obtain historical forecast data.
[0135] In this embodiment, the historical prediction data includes the final cruising range prediction result predicted at the first moment, the final cruising range prediction result predicted at the second moment, and the actual vehicle mileage from the first moment to the second moment; wherein the second moment is reached after a preset time period at the first moment.
[0136] S212. Calculate the prediction error based on the historical prediction data.
[0137] In this embodiment, it is assumed that the remaining cruising range predicted by the model at time t1 is S(1), and after driving for a period of time, the cruising range predicted by the model at time t2 is S(2); at the same time, the actual mileage △S during this period is recorded, and a weight coefficient is made according to the error between the predicted mileage and the actual mileage. The smaller the error, the greater the weight, that is, the higher the prediction credibility of the model; at the same time, when the average value of the error is larger in multiple predictions, that is, the model tends to be inapplicable as a whole, the smaller the weight, the smaller the average value of the error, and the greater the weight; the following is a possible weight coefficient design method: First, calculate the error ∈ of each prediction i , the formula is:
[0138] ∈ i =|S i (1) -S i (2) -ΔS i |;
[0139] Here, i represents the i-th prediction.
[0140] S213. Calculate the error average according to the prediction error.
[0141] In this embodiment, the average value of the error is calculated The formula is:
[0142]
[0143] Where n is the total number of predictions.
[0144] S214. Calculate the weight coefficient according to the error average value.
[0145] In this embodiment, the calculation formula of the weight coefficient γ is:
[0146]
[0147] Alternatively, the following calculation formula can be used:
[0148]
[0149] Among them, k, h and m, p are adjustment parameters used to control the influence of single error and average error on weight. By adjusting the value of (k, h) or (m, p), the weight coefficient can reflect the error situation more reasonably according to the characteristics and requirements of actual data.
[0150] In this embodiment, a dynamically adjusted weight coefficient is adopted and determined according to the error of each prediction and the average error of multiple predictions, so that the prediction result is more in line with the actual situation and the credibility and reliability of the prediction are improved.
[0151] S215: Calculate a final range prediction result according to the pre-calculated weight coefficient, the first range prediction value, and the second range prediction value.
[0152] In this embodiment, the first range prediction value predicted by the probability density model method has strong stability and can effectively show the normal driving performance of the driver and his vehicle based on historical data. However, if the driving conditions change significantly, such as a sudden change from the usual urban conditions to long-term high-speed driving conditions, such model-based predictions will lag and lack sensitivity. In order to resolve the contradiction between stability and sensitivity, the model for predicting the second range prediction value is introduced for weighted correction.
[0153] In this embodiment, the final cruising range prediction result = the first cruising range prediction value * weight coefficient + (1-weight coefficient) * the second cruising range prediction value; wherein the weight coefficient is automatically adjusted according to the prediction error.
[0154] In this embodiment, the grey model is selected for prediction, and the prediction results are weighted averaged. The average coefficient adopts a dynamic coefficient, that is, the error of the probability density model prediction is judged by a specific method, and the weight is determined according to the error.
[0155] In this embodiment, it is assumed that the remaining cruising range predicted by the grey model is S2, and the final predicted remaining cruising range is S pre for:
[0156] Spre =γS1+(1-γ)S2
[0157] Among them, γ is the pre-calculated weight coefficient.
[0158] As an optional implementation, after any one of steps S207 to S215, the following steps may be further included to achieve real-time correction of the parameters of the probability density distribution model.
[0159] In this embodiment, many factors that affect the cruising range may change over time, such as seasonal changes in ambient temperature, aging of the vehicle battery, changes in the driver's driving habits, etc., which may cause the original model to become inaccurate, so a model correction mechanism must be added.
[0160] S216: Obtain new driving sample data of the target vehicle.
[0161] S217. Calculate the data weight corresponding to the new driving sample data using a pre-constructed weight function.
[0162] In this embodiment, a pre-constructed weight function w(t), where t represents the timestamp of the data, satisfies the requirement that the weight gradually decreases over time, and specifically, an exponential decay function may be used:
[0163] w(t)=e -λt Where λ is the attenuation coefficient.
[0164] In this embodiment, for each newly collected driving sample data, the data weight w is calculated according to its timestamp. i , specifically, w i =w(t i ).
[0165] S218. Calculate new model parameters of the probability density distribution model according to the new driving sample data.
[0166] In this embodiment, it is assumed that the parameters of the original probability distribution model are the old model parameters N(u, σ 2 Based on the new driving sample data, the parameters are re-estimated using maximum likelihood estimation or other parameter estimation methods to obtain the new model parameters N(u′, σ′ 2 ).
[0167] S219. Modify the parameters of the probability density distribution model according to the data weights and the new model parameters to obtain a modified target probability density distribution model.
[0168] As an optional implementation, the probability density distribution model is modified according to the data weight and the new model parameters to obtain a modified target probability density distribution model, which may also include the following steps:
[0169] Get the old model parameters of the probability density distribution model;
[0170] Perform parameter fusion processing on the old model parameters and the new model parameters according to the data weights to obtain the target model parameters;
[0171] The model parameters of the probability density distribution model are updated to the target model parameters to obtain a revised target probability density distribution model.
[0172] In this embodiment, the old model parameters and the new model parameters can be fused in a weighted average manner to obtain the corrected final parameter N(u final ,σf inal 2 ),in:
[0173] u final =αu′+(1-α)u;
[0174]
[0175] Among them, C is a constant used to adjust the influence of old data. Other parameters are integrated in the same way.
[0176] In this embodiment, a weight function and a model correction mechanism are introduced to adjust the parameters of the probability distribution model according to the passage of time and data changes, adapt to model inaccuracies caused by factors such as ambient temperature, vehicle battery aging, and changes in driving habits, and improve the long-term accuracy and adaptability of the prediction.
[0177] In this embodiment, the method can more accurately and flexibly predict the remaining cruising range, provide the driver with more accurate and practical information, and help to reasonably plan the journey and charging arrangements.
[0178] It can be seen that the vehicle range prediction method described in this embodiment can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0179] Example 3
[0180] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a vehicle range prediction device provided in this embodiment. Figure 4 As shown, the vehicle cruising range prediction device includes:
[0181] A first acquisition unit 310 is used to acquire the driving history data of the target vehicle and a pre-built grey prediction model;
[0182] A construction unit 320, configured to construct a probability density distribution model of unit energy consumption mileage according to the driving history data;
[0183] A second acquisition unit 330 is used to acquire current vehicle data of the target vehicle;
[0184] A first prediction unit 340, configured to perform prediction based on the probability density distribution model and current vehicle data to obtain a first range prediction value;
[0185] A second prediction unit 350, configured to perform prediction based on the grey prediction model and current vehicle data to obtain a second range prediction value;
[0186] The first calculation unit 360 is used to calculate a final cruising range prediction result according to a pre-calculated weight coefficient, a first cruising range prediction value and a second cruising range prediction value.
[0187] In this embodiment, the explanation of the vehicle cruising range prediction device can refer to the description in Example 1 or Example 2, and will not be further elaborated in this embodiment.
[0188] It can be seen that the vehicle range prediction device described in this embodiment can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0189] Example 4
[0190] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of another vehicle cruising range prediction device provided in this embodiment. Figure 5 As shown, the vehicle cruising range prediction device includes:
[0191] A first acquisition unit 310 is used to acquire the driving history data of the target vehicle and a pre-built grey prediction model;
[0192] A construction unit 320, configured to construct a probability density distribution model of unit energy consumption mileage according to the driving history data;
[0193] A second acquisition unit 330 is used to acquire current vehicle data of the target vehicle;
[0194] A first prediction unit 340, configured to perform prediction based on the probability density distribution model and current vehicle data to obtain a first range prediction value;
[0195] A second prediction unit 350, configured to perform prediction based on the grey prediction model and current vehicle data to obtain a second range prediction value;
[0196] The first calculation unit 360 is used to calculate a final cruising range prediction result according to a pre-calculated weight coefficient, a first cruising range prediction value and a second cruising range prediction value.
[0197] As an optional implementation, the construction unit 320 includes:
[0198] The preprocessing subunit 321 is used to preprocess the driving history data to obtain preprocessed data;
[0199] A first constructing subunit 322 is used to construct a functional relationship between the driving mileage and the accumulated power consumption based on the pre-processed data;
[0200] The calculation subunit 323 is used to calculate the functional relationship between the unit energy consumption and the power consumption according to the functional relationship between the driving mileage and the accumulated power consumption;
[0201] The sampling subunit 324 is used to perform discrete sampling on the functional relationship between the unit energy consumption and the power consumption to obtain sampling data;
[0202] The statistical subunit 325 is used to perform data statistics on the sampled data to obtain statistical results;
[0203] The second construction subunit 326 is used to construct a probability density distribution model of unit energy consumption mileage based on the statistical results.
[0204] Further, the preprocessing subunit 321 includes:
[0205] A verification module, used for verifying the driving history data to obtain first processed data;
[0206] A screening module, used for screening the first processed data to obtain screened second processed data;
[0207] A cleaning module, used for cleaning the second processed data to obtain cleaned third processed data;
[0208] The removal module is used to remove abnormal values and error values from the third processed data to obtain pre-processed data.
[0209] Furthermore, the vehicle cruising range prediction device also includes:
[0210] The third acquisition unit 370 is used to acquire historical prediction data before calculating the final cruising range prediction result according to the pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value; wherein the historical prediction data includes the final cruising range prediction result predicted at the first moment, the final cruising range prediction result predicted at the second moment, and the actual vehicle mileage from the first moment to the second moment; wherein the second moment is reached after a preset time period of the first moment;
[0211] A second calculation unit 380, configured to calculate a prediction error based on historical prediction data;
[0212] The second calculation unit 380 is further used to calculate an error average value according to the prediction error;
[0213] The second calculation unit 380 is further configured to calculate a weight coefficient according to the error average value.
[0214] Furthermore, the vehicle cruising range prediction device also includes:
[0215] The fourth acquisition unit 390 is used to acquire new sample data of the target vehicle's travel;
[0216] The third calculation unit 400 is used to calculate the data weight corresponding to the new driving sample data by using a pre-built weight function;
[0217] The third calculation unit 400 is further used to calculate new model parameters of the probability density distribution model according to the new driving sample data;
[0218] The correction unit 410 is used to correct the parameters of the probability density distribution model according to the data weight and the new model parameters to obtain a corrected target probability density distribution model.
[0219] Further, the correction unit 410 includes:
[0220] An acquisition subunit 411 is used to acquire old model parameters of the probability density distribution model;
[0221] A fusion subunit 412 is used to perform parameter fusion processing on old model parameters and new model parameters according to data weights to obtain target model parameters;
[0222] The updating subunit 413 is used to update the model parameters of the probability density distribution model to the target model parameters to obtain a modified target probability density distribution model.
[0223] In this embodiment, the explanation of the vehicle cruising range prediction device can refer to the description in Example 1 or Example 2, and will not be further elaborated in this embodiment.
[0224] It can be seen that the vehicle range prediction device described in this embodiment can accurately predict the remaining range, taking into account the stability and sensitivity of the prediction, and providing users with accurate and reliable remaining range information.
[0225] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the vehicle range prediction method in embodiment 1 or embodiment 2 of the present application.
[0226] An embodiment of the present application provides a computer-readable storage medium storing computer program instructions. When the computer program instructions are read and executed by a processor, the vehicle range prediction method in Embodiment 1 or Embodiment 2 of the present application is executed.
[0227] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.
[0228] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
[0229] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0230] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.
[0231] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0232] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
Claims
1. A method for predicting vehicle cruising range, characterized in that: include: Obtain the target vehicle's driving history data and the pre-built grey prediction model; Constructing a probability density distribution model of unit energy consumption mileage according to the driving history data; Acquire current vehicle data of the target vehicle; Perform prediction based on the probability density distribution model and the current vehicle data to obtain a first range prediction value; Perform prediction based on the grey prediction model and the current vehicle data to obtain a second range prediction value; The final cruising range prediction result is calculated based on the pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value.
2. The method for predicting the vehicle cruising range according to claim 1, characterized in that: The method of constructing a probability density distribution model of unit energy consumption mileage according to the driving history data includes: Preprocessing the driving history data to obtain preprocessed data; Based on the preprocessed data, construct a functional relationship between the driving mileage and the accumulated power consumption; Calculate the functional relationship between unit energy consumption and power according to the functional relationship between the driving mileage and the accumulated power consumption; Discretely sampling the functional relationship between the unit energy consumption and the power consumption to obtain sampling data; Performing data statistics on the sampled data to obtain statistical results; A probability density distribution model of unit energy consumption mileage is constructed based on the statistical results.
3. The method for predicting the vehicle's cruising range according to claim 2, characterized in that: The preprocessing of the driving history data to obtain preprocessed data includes: Performing verification processing on the driving history data to obtain first processed data; Screening the first processed data to obtain screened second processed data; performing cleaning processing on the second processed data to obtain cleaned third processed data; The third processed data is processed to remove abnormal values and error values to obtain pre-processed data.
4. The method for predicting the vehicle cruising range according to claim 1, characterized in that: Before calculating the final cruising range prediction result according to the pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value, the method further includes: Acquire historical prediction data; wherein the historical prediction data includes a final cruising range prediction result predicted at a first moment, a final cruising range prediction result predicted at a second moment, and an actual vehicle mileage from the first moment to the second moment; wherein the second moment is reached after a preset time period from the first moment; Calculating a prediction error based on the historical prediction data; Calculating an error average according to the prediction error; A weight coefficient is calculated according to the error average value.
5. The method for predicting the vehicle cruising range according to claim 1, characterized in that: The method further comprises: Acquire new sample data of the target vehicle's travel; Calculating the data weight corresponding to the new driving sample data by using a pre-constructed weight function; Calculating new model parameters of the probability density distribution model according to the new driving sample data; The probability density distribution model is modified according to the data weight and the new model parameters to obtain a modified target probability density distribution model.
6. The method for predicting the vehicle cruising range according to claim 5, characterized in that: The method of performing parameter correction on the probability density distribution model according to the data weight and the new model parameters to obtain a corrected target probability density distribution model includes: Obtaining old model parameters of the probability density distribution model; Perform parameter fusion processing on the old model parameters and the new model parameters according to the data weights to obtain target model parameters; The model parameters of the probability density distribution model are updated to the target model parameters to obtain a revised target probability density distribution model.
7. A vehicle range prediction device, characterized in that: The vehicle cruising range prediction device comprises: A first acquisition unit is used to acquire the driving history data of the target vehicle and a pre-built grey prediction model; A construction unit, used to construct a probability density distribution model of unit energy consumption mileage according to the driving history data; A second acquisition unit, used to acquire current vehicle data of the target vehicle; A first prediction unit, configured to perform prediction according to the probability density distribution model and the current vehicle data to obtain a first range prediction value; A second prediction unit, configured to perform prediction based on the grey prediction model and the current vehicle data to obtain a second range prediction value; The first calculation unit is used to calculate a final cruising range prediction result according to a pre-calculated weight coefficient, the first cruising range prediction value and the second cruising range prediction value.
8. The vehicle cruising range prediction device according to claim 7, characterized in that: The building block comprises: A preprocessing subunit, used for preprocessing the driving history data to obtain preprocessed data; A first construction subunit is used to construct a functional relationship between the driving mileage and the accumulated power consumption based on the pre-processed data; A calculation subunit, used for calculating the functional relationship between unit energy consumption and power consumption according to the functional relationship between the driving mileage and the accumulated power consumption; A sampling subunit, used for discretely sampling the functional relationship between the unit energy consumption and the power consumption to obtain sampling data; A statistical subunit, used for performing data statistics on the sampled data to obtain statistical results; The second construction subunit is used to construct a probability density distribution model of unit energy consumption mileage based on the statistical results.
9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the automobile cruising range prediction method according to any one of claims 1 to 6.
10. A readable storage medium, characterized in that: The readable storage medium stores computer program instructions, and when the computer program instructions are read and executed by a processor, the vehicle range prediction method according to any one of claims 1 to 6 is executed.
Citation Information
Cited By
System and method for predicting endurance mileage of pure electric commercial vehicle and server
CN119849322A
Range prediction system, method and server for pure electric commercial vehicles
CN119849322B
Travel distance evaluation method and device, computer equipment and storage medium
CN120318936A
Electric vehicle endurance measuring and calculating method and system based on intelligent terminal and storage medium
CN121212467A
Method and apparatus for determining an estimated remaining driving range of an electric vehicle
EP4817946A1
