Kinetic energy recovery method and system based on user characteristic differences
By real-time monitoring and analysis of driver characteristics and climatic conditions, dynamically optimized kinetic energy recovery solutions, solving the problems of low kinetic energy recovery efficiency and poor use comfort in the existing technology, and achieving the generation of personalized solutions and improving energy efficiency.
Patent Information
- Application Number
- CN202410526873.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-04-29
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-04-29
AI Technical Summary
The existing kinetic energy recovery methods and systems based on user characteristics differences cannot effectively optimize the kinetic energy recovery plan to adapt to the characteristics and environmental conditions of different drivers, resulting in unsatisfactory kinetic energy recovery efficiency and reduced use comfort.
By obtaining user characteristic parameters and driving climate parameters in real time, comparing standards and actual parameters, analyzing user characteristic differences and the impact of climate differences, dynamically updating and optimizing initial kinetic energy recovery plans, and generating personalized advanced kinetic energy recovery plans to ensure that the plan is always in Optimal state.
It improves energy recovery efficiency, reduces energy consumption and exhaust emissions, extends the service life of the brake system, and improves the performance and reliability of the system.
Smart Images

Figure CN118457248B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of automobile control, and in particular to a kinetic energy recovery method and system based on user characteristic differences. Background Art
[0002] User characteristic differences refer to the characteristic differences between different users or the same user in different situations. For example, in car driving, different drivers may have different driving habits, such as driving speed, frequency of braking, turning habits, etc. These can all be regarded as user characteristics, and the user characteristics of different users are also different in their vehicle operating habits when driving.
[0003] Kinetic energy recovery, also known as energy recovery or regenerative braking, is a technology that converts kinetic energy into reusable energy during the braking process of a vehicle. By converting the vehicle's kinetic energy into storable and reusable electrical energy when the vehicle brakes, the vehicle can improve energy efficiency, reduce energy consumption and reduce environmental pollution when in use. This also makes the kinetic energy recovery function widely used in electric vehicles and hybrid vehicles, and will continue to play an important role in the future development of automotive technology.
[0004] Patent CN202311007554X provides an adaptive braking energy recovery method, system and vehicle, which includes obtaining target vehicle information and driving behavior data within a target time, confirming the start and end positions of a single trip based on the target vehicle information and the corresponding behavior time to obtain an operating scenario feature matrix for the single trip, extracting target driving behavior data within the single trip based on the operating scenario feature matrix for the single trip to obtain a driving habit feature matrix, and optimizing braking energy recovery in stages using a pre-constructed braking energy recovery optimization function based on the operating scenario feature matrix and driving habit feature matrix for the single trip to obtain a target braking energy recovery strategy, and performing energy recovery according to the target braking energy recovery strategy.
[0005] However, when the above method is used, the kinetic energy recovery scheme is not optimized and adjusted according to the differences between the user's driving characteristics and standard driving, resulting in that the kinetic energy recovery efficiency of the existing kinetic energy recovery method and system based on user characteristic differences is not ideal when used, and the kinetic energy recovery scheme cannot match the driver's characteristics, resulting in a significant reduction in the driver's comfort, and the kinetic energy recovery scheme cannot be updated in real time, resulting in the kinetic energy recovery scheme cannot be optimized according to the driver's characteristics and the environmental characteristics during driving, making the applicability of the kinetic energy recovery method and system based on user characteristic differences not ideal when used.
[0006] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention
[0007] In view of the problems in the related art, the present invention proposes a kinetic energy recovery method and system based on user characteristic differences to overcome the above-mentioned technical problems existing in the existing related art.
[0008] To this end, the specific technical solution adopted by the present invention is as follows:
[0009] According to one aspect of the present invention, a kinetic energy recovery method based on user characteristic differences comprises the following steps:
[0010] S1. Obtain vehicle information parameters, and generate an initial kinetic energy recovery plan according to the vehicle information parameters;
[0011] S2, presetting standard user characteristic parameters and standard climate parameters, and obtaining user driving characteristic parameters and driving climate parameters in real time, and generating actual user characteristic parameters and actual driving climate parameters;
[0012] S3, comparing the standard user characteristic parameters with the actual user characteristic parameters to obtain a user characteristic difference value;
[0013] S4. Preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result;
[0014] S5. Verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan;
[0015] S6. Compare the standard climate parameters with the actual driving climate parameters to obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery solution according to the climate difference impact parameters;
[0016] S7. Verify and adjust the optimized advanced kinetic energy recovery plan, obtain a suitable kinetic energy recovery plan, and store and feedback the suitable kinetic energy recovery plan.
[0017] As a preferred solution, standard user characteristic parameters and standard climate parameters are preset, and user driving characteristic parameters and driving climate parameters are acquired in real time, and the generation of actual user characteristic parameters and actual driving climate parameters includes the following steps:
[0018] S21. Obtain the identity information of the driver, and monitor the driving data and climate data of the vehicle in real time;
[0019] S22, cleaning the collected driving data and driving climate data of the vehicle to obtain driving characteristic data and driving climate data;
[0020] S23, presetting a driving feature extraction rule and a climate feature extraction rule, and extracting feature values from the driving feature data and the driving climate data according to the driving feature extraction rule and the climate feature extraction rule;
[0021] S24, verifying the characteristic values extracted from the driving characteristic data and the driving climate data, and constructing actual user characteristic parameters and actual driving climate parameters according to the verified characteristic values of the driving characteristic data and the driving climate data.
[0022] As a preferred solution, a kinetic energy recovery scheme update rule is preset, and the user characteristic difference value is analyzed according to the kinetic energy recovery scheme update rule. The initial kinetic energy recovery scheme is updated according to the analysis result, including the following steps:
[0023] S41, analyzing the user characteristic difference value according to the kinetic energy recovery solution update rule to obtain the time series characteristic difference parameter threshold;
[0024] S42, presetting a timing feature difference parameter optimization threshold, and comparing the timing feature difference parameter optimization threshold with the timing feature difference parameter threshold;
[0025] S43, generating an initial kinetic energy recovery adjustment plan according to a comparison result between the timing characteristic difference parameter optimization threshold and the timing characteristic difference parameter threshold;
[0026] S44. The initial kinetic energy recovery plan is updated according to the initial kinetic energy recovery adjustment plan, and verification feedback is performed on the updated initial kinetic energy recovery plan.
[0027] As a preferred solution, analyzing the user feature difference values according to the kinetic energy recovery solution update rule, and obtaining the time series feature difference parameter threshold includes the following steps:
[0028] S411, defining a time series feature collection rule, and performing time domain division on the user feature difference value according to the time series feature collection rule to obtain time domain feature parameters;
[0029] S412, preprocessing the time domain characteristic parameters and performing time series analysis;
[0030] S413, obtaining timing characteristic parameters according to the timing analysis results, and extracting the timing characteristic parameters;
[0031] S414: verify and optimize the extracted time series feature parameters, and output the optimized time series feature values as time series feature difference parameter thresholds.
[0032] As a preferred solution, preprocessing the time domain characteristic parameters and performing time series analysis includes the following steps:
[0033] S4121, performing data cleaning on the time domain characteristic parameters, and performing standardization processing on the cleaned time domain characteristic parameters;
[0034] S4122, performing Fourier transform on the standardized time domain characteristic parameters to obtain the frequency domain of the time domain characteristic parameters, and performing correlation analysis;
[0035] S4123, constructing a time series prediction model according to the correlation analysis results of the time domain characteristic parameters in the frequency domain, and calculating the time series characteristic prediction parameters according to the time series prediction model;
[0036] S4124, analyzing and evaluating the time series feature prediction parameters, adjusting the time series feature prediction parameters according to the evaluation results, and outputting the adjusted time series feature prediction parameters as the time series analysis results.
[0037] As a preferred solution, constructing a time series prediction model according to the correlation analysis results of the time domain characteristic parameters in the frequency domain, and calculating the time series characteristic prediction parameters according to the time series prediction model includes the following steps:
[0038] S41231. Use the ARIMA algorithm to build a time series prediction model based on the correlation analysis results of the time domain characteristic parameters in the frequency domain;
[0039] S41232, using the correlation analysis results of the time domain feature parameters in the frequency domain as input data to train the time series prediction model, and adjusting the hyperparameters of the time series prediction model based on the training results;
[0040] S41233, verifying the adjusted time series prediction model, and calculating time series feature prediction parameters according to the verified time series prediction model;
[0041] S41234. Verify and feedback the time series feature prediction parameters obtained by calculation.
[0042] As a preferred solution, the calculation formula for constructing a time series prediction model using the ARIMA algorithm based on the correlation analysis results of the time domain characteristic parameters in the frequency domain is:
[0043] Y t =a1Y t-1 +a2Y t-2 +…a n Y t-n +v t -v1B t-1 -v2B t-2 -…v n B t-n ;
[0044] Among them, Y t is the target variable value of the correlation analysis result at time t;
[0045] a1 is the autoregressive parameter of the correlation analysis result at time t-1;
[0046] a2 is the autoregressive parameter of the correlation analysis result at time t-2;
[0047] a n is the autoregressive parameter of the correlation analysis result at time tn;
[0048] Y t-1 The value of the target variable at time t-1 as a result of the correlation analysis;
[0049] Y t-2 The value of the target variable at time t-2 as a result of the correlation analysis;
[0050] Y t-n The value of the target variable at time tn as a result of the correlation analysis;
[0051] v t is the error term of the correlation analysis results at time t;
[0052] v1 is the moving average parameter of the correlation analysis result at time t-1;
[0053] v2 is the moving average parameter of the correlation analysis result at time t-2;
[0054] v n is the moving average parameter of the correlation analysis result at time tn;
[0055] B t-1 is the error value of the correlation analysis result at time t-1;
[0056] B t-2 is the error value of the correlation analysis result at time t-2;
[0057] B t-n is the error value of the correlation analysis result at time tn.
[0058] As a preferred solution, comparing the standard climate parameters with the actual driving climate parameters to obtain the climate difference impact parameters, and optimizing the advanced kinetic energy recovery solution according to the climate difference impact parameters includes the following steps:
[0059] S61, removing outliers from actual driving climate parameters and presetting climate difference indicators;
[0060] S62, calculating a percentage change difference value between the standard climate parameter and the actual driving climate parameter;
[0061] S63, comparing the percentage change difference value with the climate difference index, and obtaining the climate difference impact parameter according to the comparison result;
[0062] S64. Preset an optimization adjustment plan, match the optimization adjustment plan with the climate difference impact parameters, and optimize the advanced kinetic energy recovery plan based on the matching results.
[0063] As a preferred solution, comparing the percentage change difference value with the climate difference index, and obtaining the climate difference impact parameter according to the comparison result includes the following steps:
[0064] S631, presetting a climate difference index threshold, and comparing the percentage change difference value with the climate difference index threshold;
[0065] S632, classifying the climate difference impact parameters according to the comparison result between the percentage change difference value and the climate difference index threshold;
[0066] S633. Carry out quantitative analysis based on the classification of climate difference impact parameters, and summarize the climate difference impact parameters based on the quantitative analysis results.
[0067] According to another aspect of the present invention, a kinetic energy recovery system based on user characteristic differences comprises:
[0068] A vehicle information acquisition module, used to acquire vehicle information parameters and generate an initial kinetic energy recovery plan according to the vehicle information parameters;
[0069] An actual parameter acquisition module is used to preset standard user characteristic parameters and standard climate parameters, and acquire user driving characteristic parameters and driving climate parameters in real time to generate actual user characteristic parameters and actual driving climate parameters;
[0070] A difference value acquisition module is used to compare the standard user characteristic parameters with the actual user characteristic parameters to obtain the user characteristic difference value;
[0071] An update judgment module is used to preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result;
[0072] A verification and judgment module, used to verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan;
[0073] A climate comparison module is used to compare standard climate parameters with actual driving climate parameters, obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery solution based on the climate difference impact parameters;
[0074] The scheme feedback module is used to verify and adjust the optimized advanced kinetic energy recovery scheme, obtain the applicable kinetic energy recovery scheme, and store and feedback the applicable kinetic energy recovery scheme.
[0075] The beneficial effects of the present invention are:
[0076] 1. The present invention monitors and analyzes the driving characteristics of the driver in real time, so that the system can generate a personalized kinetic energy recovery plan according to each driver's driving habits, thereby improving the energy recovery efficiency. The system obtains driving climate parameters in real time and optimizes and adjusts them, so that the kinetic energy recovery plan can adapt to various climatic conditions, further improving the energy recovery efficiency. At the same time, the system presets kinetic energy recovery plan update rules and updates the initial kinetic energy recovery plan in real time according to real-time data, ensuring that the kinetic energy recovery plan is always in the optimal state.
[0077] 2. The present invention improves energy recovery efficiency by optimizing the kinetic energy recovery scheme in real time, thereby reducing energy consumption and tail gas emissions, which is beneficial to environmental protection and energy conservation. The updated kinetic energy recovery scheme is verified, judged and adjusted by the system to ensure its effectiveness and reliability, thereby improving the performance of the entire system. In addition, the kinetic energy recovery system can reduce dependence on the traditional braking system during braking, reduce brake wear, and thus extend the service life of the braking system. BRIEF DESCRIPTION OF THE DRAWINGS
[0078] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0079] Figure 1 is a method flow chart of a kinetic energy recovery method based on user characteristic differences according to an embodiment of the present invention;
[0080] Figure 2 4 is a system block diagram of a kinetic energy recovery system based on user characteristic differences according to an embodiment of the present invention.
[0081] In the figure:
[0082] 1. Vehicle information acquisition module; 2. Actual parameter acquisition module; 3. Difference value acquisition module; 4. Update judgment module; 5. Verification judgment module; 6. Climate comparison module; 7. Solution feedback module. DETAILED DESCRIPTION
[0083] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0084] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0085] According to an embodiment of the present invention, a kinetic energy recovery method and system based on user characteristic differences are provided.
[0086] The present invention is further described with reference to the accompanying drawings and specific embodiments. According to one embodiment of the present invention, Figure 1 As shown, the kinetic energy recovery method based on user characteristic differences according to an embodiment of the present invention includes the following steps:
[0087] S1. Obtain vehicle information parameters, and generate an initial kinetic energy recovery plan according to the vehicle information parameters;
[0088] Specifically, basic parameters such as vehicle mass and tire size are obtained from the owner's manual or vehicle technical information. These parameters directly affect the vehicle's kinetic energy. The vehicle's dynamic parameters, such as engine displacement, speed, torque curve, etc., are read through the OBD interface. These parameters determine the vehicle's ability to generate and convert kinetic energy. The vehicle's acceleration performance is measured, and the average acceleration time is obtained through multiple 0-100 acceleration tests. The vehicle's dynamic performance indicators are calculated based on the formula. GPS is used to track and record the vehicle's actual driving data, such as speed, acceleration and deceleration frequency, etc. The owner's daily driving habits and driving environment characteristics are obtained through analysis. Combined with the vehicle parameters and driving habits obtained above, a kinetic energy recovery plan can be preliminarily formulated, a reasonable speed change strategy can be formulated based on engine performance, a suitable deceleration arc can be selected based on vehicle weight and tires, common parking spots can be identified, driving routes can be formulated, smooth and rhythmic driving styles are encouraged, deceleration or acceleration suggestions are made based on the environment, and long-term learning optimization and personalized plans can be formulated.
[0089] S2, presetting standard user characteristic parameters and standard climate parameters, and obtaining user driving characteristic parameters and driving climate parameters in real time, and generating actual user characteristic parameters and actual driving climate parameters;
[0090] Specifically, the preset standard user characteristic parameters and standard climate parameters, and the real-time acquisition of user driving characteristic parameters and driving climate parameters, and the generation of actual user characteristic parameters and actual driving climate parameters include the following steps:
[0091] S21. Obtain the identity information of the driver, and monitor the driving data and climate data of the vehicle in real time;
[0092] Specifically, basic identity information voluntarily provided by the driver, such as name, contact information, etc., and real-time acquisition of vehicle location information require the driver's prior consent. It is also necessary to make it clear that the data is only used to provide services and will not be leaked or used by others. Anonymous data needs to be collected for driving data such as speed, acceleration and deceleration, and personal identity information will not be recorded or tracked. At the same time, all data will be automatically cleared within a certain period of time. Climate data such as temperature and rainfall intensity can be obtained in real time through public channels such as weather forecast interfaces to obtain the approximate climate of the current location without directly connecting to the vehicle system. At the same time, the driver should be clearly informed of the purpose and policy of data collection when using it, and the driver should be allowed to modify or withdraw consent at any time. The data security of all drivers should be strictly protected, and technical means such as encrypted transmission and storage should be adopted to prevent data leakage. Local laws and regulations should be observed, and no actions that may violate privacy should be taken.
[0093] S22, cleaning the collected driving data and driving climate data of the vehicle to obtain driving characteristic data and driving climate data;
[0094] Specifically, check the integrity of the original data, fill in missing values, process outliers, etc., extract key features from the original data, such as speed, acceleration, turning angle, etc. as driving characteristics, temperature, precipitation, etc. as climate characteristics, and then classify and organize the extracted feature data, such as arranging speed data by time series and climate data by geographical location. Considering different vehicle models and driving habits, standardize the feature data, such as converting speed into standard deviation value, perform cluster analysis on the driving feature data, and classify different types of driving modes such as urban driving and highway driving.
[0095] Then, abnormal feature values are filtered out by setting thresholds or using statistical methods, such as speeds that are significantly greater than the speed limit. For a small number of missing values, they are filled with average values or previous and next values. When a large number of missing values appear, the record may need to be discarded. The feature set that is most relevant to driving evaluation and recommendations is selected, such as speed, acceleration, turning, etc., and the processed driving feature data and climate data are output to provide input for subsequent analysis.
[0096] S23, presetting a driving feature extraction rule and a climate feature extraction rule, and extracting feature values from the driving feature data and the driving climate data according to the driving feature extraction rule and the climate feature extraction rule;
[0097] Specifically, dynamic features such as speed and acceleration are extracted, maneuvering features such as turning angles and number of brakes are extracted, features of different road types such as highways and urban roads are extracted, basic meteorological features such as temperature and humidity are extracted, auxiliary meteorological features such as precipitation intensity and wind force are extracted, features under different climatic conditions such as sunny days and rainy days are extracted, various dynamic and maneuvering features defined by the rules are extracted from the cleaned data set, and then various meteorological and conditional features defined by the rules are extracted from the data set, and a driving feature set is output, including various feature values such as speed, acceleration, and turning angle; a climate feature set is output, including various feature values such as temperature and precipitation; random sampling is performed to check whether the feature value extraction complies with the preset rules, and statistical inspection is performed to check whether there are any missing or repeatedly extracted features.
[0098] S24, verifying the characteristic values extracted from the driving characteristic data and the driving climate data, and constructing actual user characteristic parameters and actual driving climate parameters according to the verified characteristic values of the driving characteristic data and the driving climate data.
[0099] Specifically, some eigenvalues are randomly selected, and they are manually checked to see if they meet the expected results. Descriptive statistics are performed on the eigenvalues, extreme points and outliers are checked, and it is checked whether there are repeated eigenvalues and whether there are missing eigenvalues. After the inspection is completed, a distribution graph of the eigenvalues is drawn, and then it is checked whether they meet the general rules, and the correlation between the eigenvalues is analyzed, and strongly correlated features are filtered out. According to the verification results, the features that have the greatest impact on the evaluation are selected as input, and then according to the selected driving characteristics, a feature parameter set that describes individual users is constructed. At the same time, according to the selected climate characteristics, a parameter set that describes different climate conditions is constructed, and the verified user feature parameters and driving climate parameters are output to provide a basis for subsequent modeling and application.
[0100] S3, comparing the standard user characteristic parameters with the actual user characteristic parameters to obtain a user characteristic difference value;
[0101] Specifically, based on the universal user feature parameter template obtained from a large number of sample statistics, personalized feature parameters are extracted for a single user based on his or her driving behavior, and then the actual user features and standard features are matched according to the same rules, such as speed corresponding to speed, etc. For each matching feature pair, the difference between the actual value and the standard value is calculated, and then a descriptive statistical analysis is performed on all feature differences to obtain the range of the degree of difference, and abnormal differences that may be caused by extraction or statistical errors are filtered out based on empirical values. Then, based on the degree of influence of the features on the results, important features with the largest differences are screened, and the important feature difference values are output to describe the personality differences between the user and the standard template, and the specific meaning of the difference values is given, such as a higher speed value indicates a more impatient driving mode.
[0102] S4. Preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result;
[0103] Specifically, the preset kinetic energy recovery scheme update rule, analyzing the user characteristic difference value according to the kinetic energy recovery scheme update rule, and updating the initial kinetic energy recovery scheme according to the analysis result includes the following steps:
[0104] S41, analyzing the user characteristic difference value according to the kinetic energy recovery solution update rule to obtain the time series characteristic difference parameter threshold;
[0105] Specifically, the method of analyzing the user characteristic difference value according to the kinetic energy recovery scheme update rule and obtaining the time series characteristic difference parameter threshold comprises the following steps:
[0106] S411, defining a time series feature collection rule, and performing time domain division on the user feature difference value according to the time series feature collection rule to obtain time domain feature parameters;
[0107] Specifically, determine the time series unit, such as hours, days, etc., define the time series feature collection rules, count the feature difference values according to the given time series unit, such as every hour, record the feature difference values under different time series units, such as working hours on weekdays, extract the time series features from the user feature difference values according to the collection rules, count and extract the feature difference values of different time periods according to the rules, encode the time series features, represent different time periods with numbers or some marks, such as 08:00-12:00 on weekdays is represented as 1, statistically describe the time series features, record the statistics of the feature difference values of each time period, such as the mean and standard deviation, construct time domain feature parameters, combine the statistical description parameters of different time periods into a time domain feature parameter set, and output the user time domain feature parameters after time domain division.
[0108] S412, preprocessing the time domain characteristic parameters and performing time series analysis;
[0109] Specifically, the preprocessing of the time domain characteristic parameters and the time series analysis include the following steps:
[0110] S4121, performing data cleaning on the time domain characteristic parameters, and performing standardization processing on the cleaned time domain characteristic parameters;
[0111] Specifically, check whether the time domain characteristic parameter values are complete and whether there are outliers. Then identify and process outliers according to the rules, fill in a small number of missing values with methods such as average values, previous and next values, and arrange the characteristics of different time periods in chronological order. Then standardize the characteristic values of each time period, such as minimum-maximum standardization. You can also use z-score standardization to convert the values into a standard normal distribution. Check whether the standardized time domain characteristic parameter values are reasonable. If there are values that need to be re-standardized, repeat the above standardization process to output the time domain characteristic parameters that have been cleaned and standardized.
[0112] S4122, performing Fourier transform on the standardized time domain characteristic parameters to obtain the frequency domain of the time domain characteristic parameters, and performing correlation analysis;
[0113] Specifically, the time domain feature parameters are converted from the time domain to the frequency domain to obtain the corresponding frequency component representation. The low-frequency components that have a greater impact on the results are screened out according to the energy concentration, and then the amplitude and phase information at each main frequency are extracted as frequency domain features. The features at different frequencies are sorted and summarized into a frequency domain feature parameter set, and the correlation coefficient matrix between the frequency domain features is calculated. The strength of the linear correlation between the features is analyzed, and the strongly correlated features are filtered out according to the threshold to avoid redundant information. The frequency domain feature parameters and correlation analysis results after Fourier transformation are output.
[0114] S4123, constructing a time series prediction model according to the correlation analysis results of the time domain characteristic parameters in the frequency domain, and calculating the time series characteristic prediction parameters according to the time series prediction model;
[0115] Specifically, the steps of constructing a time series prediction model according to the correlation analysis results of the time domain feature parameters in the frequency domain and calculating the time series feature prediction parameters according to the time series prediction model include the following steps:
[0116] S41231. Use the ARIMA algorithm to build a time series prediction model based on the correlation analysis results of the time domain characteristic parameters in the frequency domain;
[0117] Specifically, the calculation formula for constructing a time series prediction model using the ARIMA algorithm based on the correlation analysis results of the time domain characteristic parameters in the frequency domain is:
[0118] Y t =a1Y t-1 +a2Y t-2 +…a n Y t-n +v t -v1B t-1 -v2B t-2 -…v n B t-n ;
[0119] Among them, Y tis the target variable value of the correlation analysis result at time t;
[0120] a1 is the autoregressive parameter of the correlation analysis result at time t-1;
[0121] a2 is the autoregressive parameter of the correlation analysis result at time t-2;
[0122] a n is the autoregressive parameter of the correlation analysis result at time tn;
[0123] Y t-1 The value of the target variable at time t-1 as a result of the correlation analysis;
[0124] Y t-2 The value of the target variable at time t-2 as a result of the correlation analysis;
[0125] Y t-n The value of the target variable at time tn as a result of the correlation analysis;
[0126] v t is the error term of the correlation analysis results at time t;
[0127] v1 is the moving average parameter of the correlation analysis result at time t-1;
[0128] v2 is the moving average parameter of the correlation analysis result at time t-2;
[0129] v n is the moving average parameter of the correlation analysis result at time tn;
[0130] B t-1 is the error value of the correlation analysis result at time t-1;
[0131] B t-2 is the error value of the correlation analysis result at time t-2;
[0132] B t-n is the error value of the correlation analysis result at time tn.
[0133] S41232, using the correlation analysis results of the time domain feature parameters in the frequency domain as input data to train the time series prediction model, and adjusting the hyperparameters of the time series prediction model based on the training results;
[0134] Specifically, the data is divided into a training data set and a test data set, and a time series prediction model such as RNN, LSTM, etc. is selected. The hyperparameters are initialized and the initial values of the model hyperparameters, such as the number of hidden units, learning rate, etc., are set. The frequency domain features are used as the input model and the time series is used as the output. The time series prediction model is trained on the training data set, and the mapping relationship between the learning features and the time series is learned. The prediction accuracy of the model is evaluated on the test data set, the error index is calculated, and the hyperparameters are adjusted using methods such as random search and grid search. The hyperparameter values are continuously optimized. After each adjustment, the training and evaluation are repeated to select the parameters with the highest accuracy. When the accuracy improvement is not obvious or the maximum number of iterations is reached, the parameter adjustment is stopped, and the optimal hyperparameters are output, and the output is the time series prediction model based on the adjusted parameters.
[0135] S41233, verifying the adjusted time series prediction model, and calculating time series feature prediction parameters according to the verified time series prediction model;
[0136] Specifically, a completely independent validation data set is input into the adjusted time series prediction model, the prediction effect of the model on the new validation data set is observed, and whether the model is overfitting is determined. According to the input time domain characteristics, the trained time series model is used to predict the characteristics of future moments, and the time domain characteristic parameter values at different time points are predicted. The predicted values at each time point are organized into a predicted time domain characteristic parameter set, and the error between the predicted value and the true value is analyzed to determine the prediction effect and the prediction ability of the model. If the prediction effect is not ideal, the possible reasons are analyzed and the model is tuned again, and finally the predicted time domain characteristic parameters of the verified model are output.
[0137] S41234. Verify and feedback the time series feature prediction parameters obtained by calculation.
[0138] Specifically, after the prediction time point, the actual observed time domain feature values are continuously collected, the predicted values are compared with the actual observed values, the absolute error and relative error are calculated, the prediction accuracy of individual features and overall features are analyzed, and the features with large errors are identified. Then, according to the error analysis results, the parameters or structure of the prediction model are adjusted to optimize the prediction effect. The adjusted model is used to re-test the prediction, and it is evaluated whether the prediction accuracy of the adjusted model has been improved. The prediction accuracy evaluation results are then fed back to the user, and suggestions are provided to help the user understand and apply the prediction results. The above process is repeated, and new data is continuously collected to verify the model, and the prediction quality is continuously improved. Data collection and sharing are performed under the premise of ensuring user privacy.
[0139] S4124, analyzing and evaluating the time series feature prediction parameters, adjusting the time series feature prediction parameters according to the evaluation results, and outputting the adjusted time series feature prediction parameters as the time series analysis results.
[0140] Specifically, calculate the error indicators between the prediction parameters and the actual values, such as MAE, MSE, etc., and judge the prediction importance of each feature according to the size of the indicator, then analyze the correlation between the prediction parameters, identify redundant features, filter or downgrade features with low importance or high correlation, retrain the model, optimize the prediction effect, re-predict using the adjusted parameters, compare the prediction accuracy before and after the adjustment, evaluate the effect of the parameter adjustment, output the time series feature prediction parameters after the evaluation and adjustment, and output the prediction parameters as the result of the time series analysis.
[0141] S413, obtaining timing characteristic parameters according to the timing analysis results, and extracting the timing characteristic parameters;
[0142] Specifically, the time series analysis results are obtained, the time series signal is divided into time windows, and the time domain features in each time window are extracted. The extractable features include statistical features such as mean value, standard deviation, maximum value, minimum value, etc. The various time domain features extracted in different time windows are summarized and sorted, and the time domain features with good prediction effect are selected according to the result of feature importance analysis, and the features with poor prediction effect or high correlation are filtered out, and the selected time domain feature parameter set is output, and then the time domain feature parameters are output as the final result of the time series analysis.
[0143] S414: verify and optimize the extracted time series feature parameters, and output the optimized time series feature values as time series feature difference parameter thresholds.
[0144] Specifically, use a new validation data set to predict feature parameters, calculate the error between the predicted value and the actual value, evaluate the quality of feature parameters, adjust the parameters of low-quality features or delete features based on error analysis, retrain the model, optimize feature parameters, calculate the variation range of the optimized feature parameters at different time points, set the variation range as the parameter threshold of the time series feature difference, and then output the calculated time series feature difference parameter threshold, repeat the optimization, and continuously collect new data for verification and optimization to continuously improve the accuracy of feature parameters and thresholds.
[0145] S42, presetting a timing feature difference parameter optimization threshold, and comparing the timing feature difference parameter optimization threshold with the timing feature difference parameter threshold;
[0146] Specifically, preset optimization thresholds, preset a set of timing feature difference parameter optimization thresholds based on empirical values or previous analysis results, obtain the calculated timing feature difference parameter thresholds from the output of the previous step, and then compare the preset optimization thresholds with the original thresholds for corresponding items, analyze the consistency of the two sets of thresholds in numerical value size and change trend, adjust the preset optimization thresholds based on the comparison results to make them closer to the distribution law of the original thresholds, and output the adjusted timing feature difference parameter optimization thresholds.
[0147] S43, generating an initial kinetic energy recovery adjustment plan according to a comparison result between the timing characteristic difference parameter optimization threshold and the timing characteristic difference parameter threshold;
[0148] Specifically, analyze the consistency of the thresholds, determine whether the two sets of thresholds are basically consistent in value and change trend, identify the difference points, find out the characteristic points or time periods where there are obvious differences between the two sets of thresholds, analyze the possible reasons for the threshold differences, such as data errors, algorithm problems, etc., set a looser response range for points with large differences, use a more refined response range for consistent points, determine the control parameters, set the upper and lower limits of the control parameters for different characteristic points according to the response strategy, including adjustable parameters such as power generation and frequency, output the preliminary kinetic energy recovery adjustment plan, record the characteristic points and parameters that need to be optimized later, and collect new data to continue to optimize and improve the adjustment plan.
[0149] S44. The initial kinetic energy recovery plan is updated according to the initial kinetic energy recovery adjustment plan, and verification feedback is performed on the updated initial kinetic energy recovery plan.
[0150] Specifically, the initial adjustment plan is implemented in the actual system, operation data is collected, the system response and operation effect during the implementation of the plan are analyzed, and the control parameters with problems or that need to be optimized are found. The parameters are modified for the problem points, such as expanding the response range, adjusting the upper and lower limits, etc. The modified parameters are then updated to the adjustment plan to form an updated plan. The updated plan is used to verify and run in the system, and the operation results before and after the modification are compared to evaluate the update effect. The verification and analysis results and existing problems are then fed back to relevant personnel, and the above process is repeated to continuously collect new data to optimize and improve the plan to ensure efficient operation.
[0151] S5. Verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan;
[0152] Specifically, set evaluation indicators for the scheme's operating effect, such as kinetic energy regulation accuracy, system response time, etc., run the updated initial scheme in the actual system, collect operating data, and calculate the values of various evaluation indicators based on the operating data. Then judge the scheme's operating effect based on the indicator values to see whether it meets the preset standards. If it does not meet the standards, modify the scheme parameters again based on the problem analysis, repeat the verification, and if it meets the standards, confirm that the scheme is an advanced kinetic energy recovery scheme, record the scheme details, output the advanced kinetic energy recovery scheme, and output relevant operating data to support analysis.
[0153] S6. Compare the standard climate parameters with the actual driving climate parameters to obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery solution according to the climate difference impact parameters;
[0154] Specifically, comparing the standard climate parameters with the actual driving climate parameters to obtain the climate difference impact parameters, and optimizing the advanced kinetic energy recovery scheme according to the climate difference impact parameters includes the following steps:
[0155] S61, removing outliers from actual driving climate parameters and presetting climate difference indicators;
[0156] Specifically, driving climate parameter data is collected, outliers are detected using statistical methods, anomaly thresholds are set based on the mean and standard deviation of the parameters, observations that exceed the thresholds are identified as outliers, values identified as outliers are deleted from the original data set, climate difference indicators are preset, the normal range of changes in parameters such as temperature and humidity are statistically analyzed based on historical data, the range is set as the preset value of the climate difference indicator, new data is continuously collected, the anomaly detection method is continuously improved and adjusted, and the range of the preset indicator is adjusted based on new data feedback.
[0157] S62, calculating a percentage change difference value between the standard climate parameter and the actual driving climate parameter;
[0158] Specifically, standard climate parameter and actual driving parameter data are collected, standard parameter and actual parameter values corresponding to the same time point are selected, the numerical difference between the standard parameter and the actual parameter is calculated, the percentage change rate is calculated, and the difference is divided by the standard parameter value to obtain the percentage change rate.
[0159] The formula is: Percentage change rate = (actual parameter value - standard parameter value) / standard parameter value × 100%
[0160] Repeat the calculation for all time points to obtain the percentage change rate value at each time point, and output the percentage change difference sequence between the standard parameters and the actual parameters.
[0161] S63, comparing the percentage change difference value with the climate difference index, and obtaining the climate difference impact parameter according to the comparison result;
[0162] Specifically, comparing the percentage change difference value with the climate difference index and obtaining the climate difference impact parameter according to the comparison result includes the following steps:
[0163] S631, presetting a climate difference index threshold, and comparing the percentage change difference value with the climate difference index threshold;
[0164] Specifically, preset climate difference index thresholds, preset threshold ranges for different climate parameters based on historical data statistics, such as temperature difference ±5%, humidity difference ±10%, etc., obtain percentage change difference values, and compare the percentage change difference sequence between the standard parameters and actual parameters calculated in the previous step. Compare the percentage change value at the same time point with the corresponding threshold to determine whether it exceeds the threshold. If the change value is within the threshold range, it is considered that the difference is within the allowable range. Otherwise, it indicates that the actual climate parameter deviates abnormally from the standard value. Count the number of points exceeding the threshold in the entire time series, and output the abnormal point statistics for subsequent analysis.
[0165] S632, classifying the climate difference impact parameters according to the comparison result between the percentage change difference value and the climate difference index threshold;
[0166] Specifically, the percentage change difference value is compared with the threshold value, and the parameters are divided into three categories according to the comparison results. For example, if all the difference values are within the threshold range, the impact is minimal, if some of the difference values exceed the threshold, the impact is general, and if the difference values mainly exceed the threshold, the impact is greatest. Specific classification standards include the first category abnormal point rate <10%, the second category abnormal point rate 10-30%, and the third category abnormal point rate >30%. The classification results of different parameters are counted separately, such as temperature is class 1, humidity is class 2, wind speed is class 3, etc., and then each parameter is output in correspondence with its degree of influence, such as in a table.
[0167] S633. Carry out quantitative analysis based on the classification of climate difference impact parameters, and summarize the climate difference impact parameters based on the quantitative analysis results.
[0168] Specifically, statistical analysis is performed on parameters of different categories, and the differences in the number of abnormal points and average abnormality rate are compared. The possible causes of the abnormalities are analyzed, such as technical factors such as sensor accuracy. Based on quantitative comparison, the top parameters with the greatest impact are summarized, and the overall impact of these parameters on the system is described. Optimization suggestions are put forward, and improvement plans are provided for these key parameters, such as replacing sensors.
[0169] S64. Preset an optimization adjustment plan, match the optimization adjustment plan with the climate difference impact parameters, and optimize the advanced kinetic energy recovery plan based on the matching results.
[0170] Specifically, based on the summary results, an optimization adjustment target is preset for each parameter with the greatest impact. For example, for temperature parameters, the sensor accuracy is improved by 1°C, and for humidity parameters, the number of sensors is increased. These optimization targets are preset as optimization adjustment schemes, and the influencing parameters are matched with the corresponding optimization schemes. For example, the temperature parameters are matched with the scheme that improves the sensor accuracy by 1°C. These suggestions are used to optimize the advanced kinetic energy recovery scheme, such as replacing the temperature sensor model, increasing the number of humidity sensor points, etc. The optimized scheme is re-verified to detect whether the optimization effect meets expectations and whether the parameter impact is further reduced. The optimization scheme is further improved based on the verification results.
[0171] S7. Verify and adjust the optimized advanced kinetic energy recovery plan, obtain a suitable kinetic energy recovery plan, and store and feedback the suitable kinetic energy recovery plan.
[0172] Specifically, record indicators such as kinetic energy recovery efficiency and system operation stability, compare indicators before and after optimization, and verify the optimization effect. If the effect is not ideal, adjust parameters, adjust sensor thresholds, optimize control algorithms, etc. Repeat the steps until you are satisfied. When the indicators meet expectations, obtain applicable kinetic energy recovery plans, store applicable plans in the database, form a callable library, monitor operation data in real time, continuously optimize according to feedback results, improve verification plans, continuously improve plan performance, and update improved plans to the database.
[0173] According to another embodiment of the present invention, Figure 2 As shown, a kinetic energy recovery system based on user characteristic differences is provided, the system comprising:
[0174] The vehicle information acquisition module 1 is used to acquire vehicle information parameters and generate an initial kinetic energy recovery plan according to the vehicle information parameters;
[0175] The actual parameter acquisition module 2 is used to preset standard user characteristic parameters and standard climate parameters, and acquire user driving characteristic parameters and driving climate parameters in real time to generate actual user characteristic parameters and actual driving climate parameters;
[0176] The difference value acquisition module 3 is used to compare the standard user characteristic parameters with the actual user characteristic parameters to obtain the user characteristic difference value;
[0177] An update judgment module 4 is used to preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result;
[0178] Verification and judgment module 5, used to verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan;
[0179] A climate comparison module 6 is used to compare the standard climate parameters with the actual driving climate parameters, obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery scheme according to the climate difference impact parameters;
[0180] The scheme feedback module 7 is used to verify and adjust the optimized advanced kinetic energy recovery scheme, obtain an applicable kinetic energy recovery scheme, and store and feedback the applicable kinetic energy recovery scheme.
[0181] To sum up, with the help of the above-mentioned technical scheme of the present invention, the present invention monitors and analyzes the driving characteristics of the driver in real time, so that the system can generate a personalized kinetic energy recovery plan according to each driver's driving habits, thereby improving the energy recovery efficiency. The system obtains driving climate parameters in real time and optimizes and adjusts them, so that the kinetic energy recovery plan can adapt to various climatic conditions, further improving the energy recovery efficiency. At the same time, the system presets the kinetic energy recovery plan update rules, and updates the initial kinetic energy recovery plan in real time according to real-time data, ensuring that the kinetic energy recovery plan is always in the optimal state.
[0182] In addition, the present invention improves energy recovery efficiency by optimizing the kinetic energy recovery scheme in real time, thereby reducing energy consumption and tail gas emissions, which is beneficial to environmental protection and energy conservation. The updated kinetic energy recovery scheme is verified, judged and adjusted by the system to ensure its effectiveness and reliability, thereby improving the performance of the entire system. In addition, the kinetic energy recovery system can reduce dependence on traditional braking systems during braking and reduce brake wear, thereby extending the service life of the braking system.
[0183] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A kinetic energy recovery method based on user characteristic differences, characterized in that: The following steps are involved: S1. Obtain vehicle information parameters, and generate an initial kinetic energy recovery plan according to the vehicle information parameters; S2, presetting standard user characteristic parameters and standard climate parameters, and obtaining user driving characteristic parameters and driving climate parameters in real time, and generating actual user characteristic parameters and actual driving climate parameters; S3, comparing the standard user characteristic parameters with the actual user characteristic parameters to obtain a user characteristic difference value; S4. Preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result; S5. Verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan; S6. Compare the standard climate parameters with the actual driving climate parameters to obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery solution according to the climate difference impact parameters; S7. Verify and adjust the optimized advanced kinetic energy recovery plan to obtain a suitable kinetic energy recovery plan, and store and feedback the suitable kinetic energy recovery plan; The preset standard user characteristic parameters and standard climate parameters, and the real-time acquisition of user driving characteristic parameters and driving climate parameters, and the generation of actual user characteristic parameters and actual driving climate parameters include the following steps: S21. Obtain the identity information of the driver, and monitor the driving data and climate data of the vehicle in real time; S22, cleaning the collected driving data and driving climate data of the vehicle to obtain driving characteristic data and driving climate data; S23, presetting a driving feature extraction rule and a climate feature extraction rule, and extracting feature values from the driving feature data and the driving climate data according to the driving feature extraction rule and the climate feature extraction rule; S24, verifying the characteristic values extracted from the driving characteristic data and the driving climate data, and constructing actual user characteristic parameters and actual driving climate parameters according to the verified characteristic values of the driving characteristic data and the driving climate data; The preset kinetic energy recovery scheme update rule, analyzing the user characteristic difference value according to the kinetic energy recovery scheme update rule, and updating the initial kinetic energy recovery scheme according to the analysis result include the following steps: S41, analyzing the user characteristic difference value according to the kinetic energy recovery solution update rule to obtain the time series characteristic difference parameter threshold; S42, presetting a timing feature difference parameter optimization threshold, and comparing the timing feature difference parameter optimization threshold with the timing feature difference parameter threshold; S43, generating an initial kinetic energy recovery adjustment plan according to a comparison result between the timing characteristic difference parameter optimization threshold and the timing characteristic difference parameter threshold; S44, updating the initial kinetic energy recovery plan according to the initial kinetic energy recovery adjustment plan, and providing verification feedback on the updated initial kinetic energy recovery plan; Analyzing the user characteristic difference value according to the kinetic energy recovery solution update rule to obtain the time series characteristic difference parameter threshold comprises the following steps: S411, defining a time series feature collection rule, and performing time domain division on the user feature difference value according to the time series feature collection rule to obtain time domain feature parameters; S412, preprocessing the time domain characteristic parameters and performing time series analysis; S413, obtaining timing characteristic parameters according to the timing analysis results, and extracting the timing characteristic parameters; S414: verify and optimize the extracted time series feature parameters, and output the optimized time series feature values as time series feature difference parameter thresholds.
2. The kinetic energy recovery method based on user characteristic differences according to claim 1, characterized in that: The preprocessing of the time domain characteristic parameters and the time series analysis comprises the following steps: S4121, performing data cleaning on the time domain characteristic parameters, and performing standardization processing on the cleaned time domain characteristic parameters; S4122, performing Fourier transform on the standardized time domain characteristic parameters to obtain the frequency domain of the time domain characteristic parameters, and performing correlation analysis; S4123, constructing a time series prediction model according to the correlation analysis results of the time domain characteristic parameters in the frequency domain, and calculating the time series characteristic prediction parameters according to the time series prediction model; S4124, analyzing and evaluating the time series feature prediction parameters, adjusting the time series feature prediction parameters according to the evaluation results, and outputting the adjusted time series feature prediction parameters as the time series analysis results.
3. The kinetic energy recovery method based on user characteristic differences according to claim 2, characterized in that: The method of constructing a time series prediction model according to the correlation analysis results of the time domain characteristic parameters in the frequency domain and calculating the time series characteristic prediction parameters according to the time series prediction model includes the following steps: S41231. Use the ARIMA algorithm to build a time series prediction model based on the correlation analysis results of the time domain characteristic parameters in the frequency domain; S41232, using the correlation analysis results of the time domain feature parameters in the frequency domain as input data to train the time series prediction model, and adjusting the hyperparameters of the time series prediction model based on the training results; S41233, verifying the adjusted time series prediction model, and calculating time series feature prediction parameters according to the verified time series prediction model; S41234. Verify and feedback the time series feature prediction parameters obtained by calculation.
4. The kinetic energy recovery method based on user characteristic differences according to claim 3, characterized in that: The calculation formula for constructing a time series prediction model using the ARIMA algorithm based on the correlation analysis results of the time domain characteristic parameters in the frequency domain is: Y t =a1Y t-1 +a2Y t-2 +…a n Y t-n +v t -v1B t-1 -v2B t-2 -…v n B t-n Among them, Y t is the target variable value of the correlation analysis result at time t; a1 is the autoregressive parameter of the correlation analysis result at time t-1; a2 is the autoregressive parameter of the correlation analysis result at time t-2; a n is the autoregressive parameter of the correlation analysis result at time tn; Y t-1 The value of the target variable at time t-1 as a result of the correlation analysis; Y t-2 The value of the target variable at time t-2 as a result of the correlation analysis; Y t-n The value of the target variable at time tn as a result of the correlation analysis; v t is the error term of the correlation analysis results at time t; v1 is the moving average parameter of the correlation analysis result at time t-1; v2 is the moving average parameter of the correlation analysis result at time t-2; v n is the moving average parameter of the correlation analysis result at time tn; B t-1 is the error value of the correlation analysis result at time t-1; B t-2 is the error value of the correlation analysis result at time t-2; B t-n is the error value of the correlation analysis result at time tn.
5. The kinetic energy recovery method based on user characteristic differences according to claim 1, characterized in that: The step of comparing the standard climate parameters with the actual driving climate parameters, obtaining the climate difference impact parameters, and optimizing the advanced kinetic energy recovery scheme according to the climate difference impact parameters includes the following steps: S61, removing outliers from actual driving climate parameters and presetting climate difference indicators; S62, calculating a percentage change difference value between the standard climate parameter and the actual driving climate parameter; S63, comparing the percentage change difference value with the climate difference index, and obtaining the climate difference impact parameter according to the comparison result; S64. Preset an optimization adjustment plan, match the optimization adjustment plan with the climate difference impact parameters, and optimize the advanced kinetic energy recovery plan based on the matching results.
6. The kinetic energy recovery method based on user characteristic differences according to claim 5, characterized in that: The step of comparing the percentage change difference value with the climate difference index and obtaining the climate difference impact parameter according to the comparison result comprises the following steps: S631, presetting a climate difference index threshold, and comparing the percentage change difference value with the climate difference index threshold; S632, classifying the climate difference impact parameters according to the comparison result between the percentage change difference value and the climate difference index threshold; S633. Carry out quantitative analysis based on the classification of climate difference impact parameters, and summarize the climate difference impact parameters based on the quantitative analysis results.
7. A kinetic energy recovery system based on user characteristic differences, used to implement the kinetic energy recovery method based on user characteristic differences according to any one of claims 1 to 6, characterized in that: The system includes: A vehicle information acquisition module, used to acquire vehicle information parameters and generate an initial kinetic energy recovery plan according to the vehicle information parameters; An actual parameter acquisition module is used to preset standard user characteristic parameters and standard climate parameters, and acquire user driving characteristic parameters and driving climate parameters in real time to generate actual user characteristic parameters and actual driving climate parameters; A difference value acquisition module is used to compare the standard user characteristic parameters with the actual user characteristic parameters to obtain the user characteristic difference value; An update judgment module is used to preset a kinetic energy recovery plan update rule, analyze the user characteristic difference value according to the kinetic energy recovery plan update rule, and update the initial kinetic energy recovery plan according to the analysis result; A verification and judgment module, used to verify and judge the updated initial kinetic energy recovery plan to obtain an advanced kinetic energy recovery plan; A climate comparison module is used to compare standard climate parameters with actual driving climate parameters, obtain climate difference impact parameters, and optimize the advanced kinetic energy recovery solution based on the climate difference impact parameters; The scheme feedback module is used to verify and adjust the optimized advanced kinetic energy recovery scheme, obtain the applicable kinetic energy recovery scheme, and store and feedback the applicable kinetic energy recovery scheme.
