Vehicle control method, device, readable storage medium and vehicle

Abstract

The present invention proposes a vehicle control method, device, readable storage medium and vehicle, the vehicle control method comprises: obtaining a first parameter set during vehicle operation, the first parameter set comprising vehicle operation parameters and time series data corresponding to the operation parameters; preprocessing the first parameter set to determine a second parameter set; clustering analysis of the second parameter set to determine the analysis result of the vehicle load condition; and determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition. In the technical solution of the present invention, the data of the vehicle load condition is subdivided by a clustering algorithm to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and then determine the optimal speed torque throttle curve strategy matching the load condition, and realize precise control of vehicle operation according to the speed torque throttle curve strategy, thereby ensuring the stability of vehicle operation and reducing the energy consumption of the vehicle.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a vehicle control method, a vehicle control device, a readable storage medium and a vehicle. Background Art

[0002] The existing classification schemes for vehicle operating conditions do not clearly propose a method for subdividing the operating condition categories based on data, do not consider the possibility of abnormal data in a large amount of data, and cannot achieve precise control of vehicle operation based on the results of the vehicle operating condition segmentation. Summary of the invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art or related art.

[0004] To this end, a first aspect of the present invention is to provide a vehicle control method.

[0005] A second aspect of the present invention is to provide a control device for a vehicle.

[0006] A third aspect of the present invention is to provide a readable storage medium.

[0007] A fourth aspect of the present invention is to provide a vehicle.

[0008] In view of this, according to the first aspect of the present invention, a vehicle control method is proposed, including: obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operating parameters of the vehicle and time series data corresponding to the operating parameters; preprocessing the first parameter set to determine a second parameter set; performing cluster analysis on the second parameter set to determine an analysis result of the vehicle load condition; and determining an engine speed torque throttle curve strategy associated with the vehicle load condition based on the analysis result of the vehicle load condition.

[0009] It should be noted that the executor of the vehicle control method proposed in the present invention may be the control device of the vehicle. In order to more clearly illustrate the vehicle control method proposed in the present invention, the following technical scheme will exemplify the executor of the vehicle control method as the control device of the vehicle.

[0010] In this technical solution, the first parameter set includes the vehicle's operating parameters and time series data corresponding to the operating parameters. The vehicle's operating parameters mainly include parameters such as load, mileage, fuel consumption, speed, etc., and the time series data is used to indicate the time series information when the above parameters are collected.

[0011] Specifically, the control device obtains a first parameter set during the operation of the vehicle, and pre-processes the first parameter set to determine a second parameter set. Specifically, the technical solution proposed in the present invention is used for the analysis of vehicle load conditions, and the first parameter set obtained includes not only the operating parameters of the vehicle load condition, but also other operating parameters, and abnormal data may also exist in the first parameter set. Therefore, after obtaining the first parameter set, the control device needs to pre-process the first parameter set to determine a second parameter set with higher availability and effectiveness to ensure the smooth progress of subsequent steps.

[0012] Furthermore, the control device performs cluster analysis on the second parameter set, that is, subdivides the data of the vehicle load condition, determines the number of interval divisions of the load condition and the segmented intervals of the load condition, and then determines the analysis result of the vehicle load condition.

[0013] Furthermore, the control device determines the speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis result of the determined vehicle load condition. Specifically, according to the analysis result of the vehicle load condition, the number of interval divisions of the vehicle load condition and the segmented intervals of the load condition can be understood, and different load condition intervals correspond to appropriate speed torque throttle curve strategies. Therefore, the control device can determine the speed torque throttle curve strategy that matches the load condition according to the analysis result of the vehicle load condition, thereby realizing precise control of the vehicle operation.

[0014] In the technical solution of the present invention, after obtaining the first parameter set during the operation of the vehicle, the control device pre-processes the first parameter set to determine a second parameter set with higher availability and effectiveness. After determining the second parameter set, the control device performs cluster analysis on the second parameter set to determine the segmented intervals and the number of divided intervals of the vehicle load condition. The control device determines the speed torque throttle curve strategy corresponding to the segmented intervals and the number of divided intervals based on this information. In the technical solution of the present invention, the data of the vehicle load condition is subdivided by a clustering algorithm to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and then the optimal speed torque throttle curve strategy matching the load condition is determined. According to the speed torque throttle curve strategy, precise control of the vehicle operation is achieved, the stability of the vehicle operation is ensured, and the energy consumption of the vehicle operation can be saved.

[0015] In addition, the vehicle control method according to the present invention may also have the following additional technical features:

[0016] In the above technical solution, the step of preprocessing the first parameter set to determine the second parameter set specifically includes: removing abnormal data in the first parameter set, and / or replacing the abnormal data with preset parameters corresponding to the abnormal data to determine the third parameter set; screening out data related to the load condition in the third parameter set to determine the second parameter set.

[0017] In this technical solution, the specific steps of preprocessing the first parameter set are: the control device screens out abnormal data in the first parameter set, and directly removes the abnormal data in the first parameter set that has a low correlation with the vehicle load condition data; for the abnormal data in the first parameter set that has a high correlation with the vehicle load condition data, the abnormal data is replaced with a preset parameter corresponding to the abnormal data, and then a third parameter set without abnormal data is determined. Specifically, the above preset parameters can be calculated based on the non-abnormal data in the first parameter set, or can be set based on the historical parameters of the vehicle operation.

[0018] Furthermore, the control device screens the third parameter set, screens out data related to the load condition in the third parameter set, and determines the second parameter set. Specifically, the data related to the load condition mainly includes load parameters, mileage parameters, fuel consumption parameters, speed parameters, etc.

[0019] In this technical solution, since the first parameter may contain abnormal data and data irrelevant to the load condition, these data will affect the analysis results of the data in the subsequent steps, which is not conducive to cluster analysis and will lead to inaccurate cluster analysis results. Therefore, in the technical solution of the present invention, the control device determines the third parameter set by removing or replacing the abnormal data in the first parameter with preset parameters, and then screens out the parameters related to the load in the third parameter set to determine the second parameter set with higher availability and effectiveness. In this way, the accuracy of the analysis results of the vehicle load condition determined by cluster analysis of the second parameter set is improved.

[0020] In the above technical scheme, the steps of clustering the second parameter set to determine the analysis results of the vehicle load condition specifically include: analyzing the second parameter set according to the first clustering algorithm to determine the first target number of clustering clusters and the first load condition segmentation interval; analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clustering clusters and the second load condition segmentation interval; determining the analysis results of the vehicle load condition based on the first target number of clustering clusters, the first load condition segmentation interval, the second target number of clustering clusters and the second load condition segmentation interval.

[0021] In this technical solution, the first clustering algorithm is a K-means clustering algorithm, and the second clustering algorithm is a Birch clustering algorithm.

[0022] Specifically, the control device performs cluster analysis on the second parameter set, and the steps of determining the analysis result of the vehicle load condition are as follows: the control device performs cluster analysis on the second parameter set according to the first clustering algorithm, and determines the first set of optimal number of cluster clusters and segmentation intervals, that is, the first target number of cluster clusters and the first load condition segmentation interval; the control device performs cluster analysis on the second parameter set according to the second clustering algorithm, and determines the second set of optimal number of cluster clusters and segmentation intervals, that is, the second target number of cluster clusters and the second load condition segmentation interval; the control device performs weighted average calculation on the above two sets of information, and obtains the final vehicle load condition segmentation interval and the number of interval divisions, and then determines the analysis result of the vehicle load condition.

[0023] In this technical solution, in the step of clustering the second parameter set, the control device uses two different clustering algorithms to cluster the second parameter set respectively, and the two clustering results are weighted averaged to obtain the final analysis result of the vehicle load condition. In this way, the accuracy and credibility of the analysis result of the determined vehicle load condition are improved.

[0024] In the above technical scheme, the second parameter set is analyzed according to the first clustering algorithm, and the steps of determining the first target clustering cluster number and the first load condition segmentation interval specifically include: determining the first load condition data according to the second parameter set; compressing the data other than the load data in the first load condition data into a column of data by principal component analysis, and determining the second load condition data based on the column of data and the load data in the first load condition data; performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the first sub-clustering cluster number and the first clustering result; performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the second sub-clustering cluster number and the second clustering result; determining the first target clustering cluster number and the first load condition segmentation interval based on the first sub-clustering cluster number, the first clustering result, the second sub-clustering cluster number and the second clustering result.

[0025] In this technical solution, the first load condition data is used to indicate the data determined by screening and processing the operating parameters of the vehicle in the second parameter set that has been running continuously for more than a certain period of time. It can be understood that the first clustering algorithm (i.e., the K-means clustering algorithm) is more suitable for clustering analysis of small and medium-sized data sets. Therefore, in the technical solution of the present invention, before performing cluster analysis through the first clustering algorithm, the control device needs to screen out data with higher reference value in the second parameter set, that is, reduce the size of the data set to increase the rate of cluster analysis.

[0026] Specifically, the step of the control device analyzing the second parameter set according to the first clustering algorithm is: the control device determines the first load condition data according to the second parameter set, and performs dimensionality reduction and compression processing on the first load condition data to determine the second load condition data with higher simplicity.

[0027] Specifically, the control device performs dimensionality reduction and compression processing on the first load condition data by filtering out data other than the load data in the first load condition data, compressing these data into a column of data through principal component analysis, and then combining this column of data with the load data to determine the second load condition data.

[0028] Furthermore, the control device performs cluster analysis on the first load condition data according to the first clustering algorithm to determine the first clustering result and the optimal number of clustering clusters, that is, the first clustering result and the first sub-clustering cluster number; the control device performs cluster analysis on the second load condition data according to the first clustering algorithm to determine the second clustering result and the optimal number of clustering clusters, that is, the second sub-clustering cluster number and the second clustering result; the control device determines the first target clustering cluster number and the first load condition segmentation interval by combining the first and second clustering results and the optimal number of clustering clusters.

[0029] In this technical solution, in the step of clustering the second parameter set according to the first clustering algorithm, the control device performs clustering analysis on the first load condition data and the second load condition data after dimensionality reduction and compression processing to obtain two groups of analysis results (i.e., clustering is performed twice using the first clustering algorithm), and the first target cluster number and the first load condition segmentation interval are determined by combining the above two groups of analysis results. In this way, the accuracy and credibility of the analysis results of the vehicle load condition determined in the subsequent process are improved.

[0030] In the above technical solution, the step of determining the first load condition data according to the second parameter set specifically includes: extracting multiple motion segment data in the second parameter set, the motion segment data is used to indicate the operating parameters of the vehicle that has been running continuously for more than a first preset time period; processing the multiple motion segment data into multiple single-line data, and determining the first load condition data based on the multiple single-line data.

[0031] In this technical solution, the motion segment data is used to indicate the operating parameters of the vehicle collected when the vehicle continuously travels for more than a first preset time period. The first preset time period is generally 1 hour, which is determined based on the actual situation of the vehicle.

[0032] Specifically, the control device selects multiple motion segment data in the second parameter set whose continuous driving time of the vehicle exceeds the first preset time, processes the multiple motion segment data into multiple single-line data, and determines the first load condition data based on the multiple single-line data. Specifically, the control device combines multiple single-line data into a single-line data by constructing business features and multi-dimensional data binning features.

[0033] In this technical solution, considering that there are operating parameters of vehicles with shorter continuous driving time in the second parameter set, these data are of low reference value and will affect the rate of cluster analysis by the first clustering algorithm. Therefore, in the technical solution of the present invention, before cluster analysis is performed by the first clustering algorithm, the control device needs to screen out multiple motion segment data in the second parameter set whose continuous driving time of vehicles exceeds the first preset time, and process them to determine the first load condition data suitable for cluster analysis. In this way, the robustness of the subsequent steps is guaranteed, thereby improving the accuracy of the analysis results of the vehicle load condition.

[0034] In the above technical scheme, cluster analysis is performed on the first load condition data according to the first clustering algorithm, and the steps of determining the first sub-clustering cluster number and the first clustering result specifically include: setting the initial parameters of the first clustering algorithm, setting the clustering cluster number to the first preset cluster number, and clustering analysis on the first load condition data; increasing the clustering cluster number in sequence until the clustering cluster number is greater than the second preset cluster number; drawing a silhouette coefficient graph with a clustering cluster number equal to the first preset cluster number to the second preset cluster number; determining the clustering cluster number with the largest silhouette coefficient in the silhouette coefficient graph as the first sub-clustering cluster number, and taking the clustering result corresponding to the first sub-clustering cluster number as the first clustering result.

[0035] In this technical solution, the above-mentioned number of clusters is used to indicate a control parameter for adjusting the complexity of the clustering algorithm; the above-mentioned silhouette coefficient is an indicator for evaluating the clustering effect in the clustering algorithm, and the larger the silhouette coefficient, the better the clustering effect.

[0036] Specifically, the control device performs cluster analysis on the first load condition data according to the first clustering algorithm as follows: the control device sets the initial parameters of the first clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the first load condition data into the first clustering algorithm for cluster analysis.

[0037] Further, the control device sequentially adjusts the number of clusters from the first preset number of clusters to a number greater than the second preset number of clusters, and draws a silhouette coefficient diagram where the number of clusters is equal to the first preset number of clusters to the second preset number of clusters. Specifically, the clustering effect of different numbers of clusters can be intuitively determined through the silhouette coefficient diagram, so in the technical solution of the present invention, the control device needs to draw a silhouette coefficient diagram based on the silhouette coefficient where the number of clusters is equal to the first preset number of clusters to the second preset number of clusters.

[0038] Furthermore, the control device compares the silhouette coefficients in the silhouette coefficient graph, determines the cluster number with the largest silhouette coefficient in the graph as the first sub-clustering cluster number, and uses the clustering result corresponding to the first sub-clustering cluster number as the first clustering result.

[0039] In the above technical scheme, cluster analysis is performed on the second load condition data according to the first clustering algorithm, and the steps of determining the second sub-clustering cluster number and the second clustering result specifically include: setting the initial parameters of the first clustering algorithm, setting the clustering cluster number to the first preset cluster number, and clustering analysis is performed on the second load condition data; increasing the clustering cluster number in sequence until the clustering cluster number is greater than the second preset cluster number; determining the error square sum curve based on the error square sum of the clustering cluster number equal to the first preset cluster number to the second preset cluster number; determining the clustering cluster number corresponding to the inflection point in the error square sum curve as the second sub-clustering cluster number, and using the clustering result corresponding to the second sub-clustering cluster number as the second clustering result.

[0040] In this technical solution, the above-mentioned sum of squared errors is an indicator for evaluating the clustering effect in the clustering algorithm. Specifically, as the number of clusters increases, the sample division will be more refined and the sum of squared errors will decrease. When the number of clusters is less than the optimal number of clusters, the increase in the number of clusters will greatly increase the degree of aggregation of each cluster, so the decrease in the sum of squared errors is large. When the number of clusters reaches the optimal value, the return on the degree of aggregation obtained by increasing the number of clusters will decrease rapidly, the decrease in the sum of squared errors will drop sharply, and then the continued increase in the number of clusters will tend to be flat. It can be seen that the optimal number of clusters can be directly determined based on the inflection point in the sum of squared errors curve.

[0041] Specifically, the control device performs cluster analysis on the second load condition data according to the first clustering algorithm as follows: the control device sets the initial parameters of the first clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the second load condition data into the first clustering algorithm for cluster analysis.

[0042] Furthermore, the control device sequentially adjusts the number of clusters from a first preset number of clusters to a number greater than a second preset number of clusters, calculates the sum of square errors when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters, and draws a sum of square errors curve based on the sum of square errors.

[0043] Furthermore, the number of clusters corresponding to the inflection point in the error square sum curve is determined as the second sub-clustering number, and the clustering result corresponding to the second sub-clustering number is used as the second clustering result.

[0044] In the above technical scheme, the second parameter set is analyzed according to the second clustering algorithm, and the steps of determining the second target number of clustering clusters and the second load condition segmentation interval specifically include: compressing the data other than the load data in the second parameter set into a column of data by principal component analysis, and determining the third load condition data based on the column of data and the load data in the second parameter set; performing cluster analysis on the third load condition data according to the second clustering algorithm to determine the second target number of clustering clusters and the second load condition segmentation interval.

[0045] In this technical solution, the step of the control device analyzing the second parameter set according to the second clustering algorithm is: the control device directly performs dimensionality reduction and compression processing on the second parameter set to determine the third load condition data with higher simplicity. It can be understood that the second clustering algorithm (i.e., the Birch clustering algorithm) is more suitable for analyzing large-scale data sets. Therefore, in this technical solution, before performing cluster analysis through the second clustering algorithm, the control device does not need to screen the second parameter set, but can directly perform dimensionality reduction and compression processing on it. In this way, the steps of cluster analysis are simplified, and the comprehensiveness of the data of cluster analysis is improved.

[0046] Specifically, the process of dimensionality reduction and compression processing on the second parameter set is to filter out the data except the load data in the second parameter set, compress these data into a column of data through principal component analysis, and then combine this column of data with the load data to determine the third load condition data.

[0047] Furthermore, the control device performs cluster analysis on the third load condition data according to the second clustering algorithm to determine a second target number of cluster clusters and a second load condition segmentation interval.

[0048] In this technical solution, the control device does not directly perform cluster analysis on the second parameter set, but instead performs cluster analysis on the third load condition data determined based on the dimensionality reduction and compression processing of the second parameter set, thereby reducing the amount of data for cluster analysis and improving the rate of cluster analysis.

[0049] In the above technical scheme, cluster analysis is performed on the third load condition data according to the second clustering algorithm, and the steps of determining the second target cluster number and the second load condition segmentation interval specifically include: setting the initial parameters of the second clustering algorithm, setting the cluster number to the first preset cluster number, and clustering analysis on the third load condition data; increasing the cluster number in sequence until the cluster number is greater than the second preset cluster number; comparing the variance ratio standards of the cluster number equal to the first preset cluster number to the second preset cluster number, and determining the cluster number corresponding to the largest variance ratio standard as the second target cluster number; and determining the second load condition segmentation interval according to the clustering results corresponding to the second target cluster number.

[0050] In this technical solution, the above-mentioned variance ratio standard (Calinski-Harabasz is abbreviated as CHI or CH index) is an indicator for evaluating the clustering effect in the clustering algorithm. Specifically, the larger the variance ratio standard, the better the clustering effect.

[0051] Specifically, the control device performs cluster analysis on the third load condition data according to the second clustering algorithm as follows: the control device sets the initial parameters of the second clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the third load condition data into the second clustering algorithm for cluster analysis.

[0052] Further, the control device adjusts the number of clusters from the first preset number of clusters to greater than the second preset number of clusters in sequence, and determines the variance ratio standard when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters. Specifically, the larger the variance ratio standard, the tighter the class itself is, and the more dispersed the different classes are, that is, the better the clustering effect is. Therefore, in the technical solution of the present invention, it is necessary to determine the variance ratio standard when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters during the cluster analysis process to evaluate the clustering effect.

[0053] Furthermore, the control device compares the variance ratio standards determined in the above steps, determines the number of clusters corresponding to the largest variance ratio standard as the second target number of clusters, and determines the second load condition segmentation interval according to the clustering result corresponding to the second target number of clusters.

[0054] In the above technical scheme, after determining the speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis results of the vehicle load condition, the vehicle control method also includes: confirming whether the data amount of the fourth parameter set is greater than a preset threshold, the fourth parameter set is used to indicate the newly collected operating parameters of the vehicle; when the data amount of the fourth parameter set is greater than the preset threshold, determining the fifth parameter set according to the fourth parameter set and the first parameter set; performing cluster analysis on the fifth parameter set, and updating or adjusting the speed torque throttle curve strategy according to the cluster analysis results of the fifth parameter set.

[0055] In this technical solution, the fourth parameter set is used to indicate newly collected operating parameters of the vehicle. It can be understood that during the operation of the vehicle, the control device collects the operating parameters of the vehicle in real time.

[0056] Specifically, the control device determines whether the amount of data contained in the fourth parameter set (i.e., the newly collected vehicle operating parameters) is greater than a preset threshold. Specifically, the preset threshold is generally set to 10% of the amount of data in the first parameter set (i.e., the data used for cluster analysis), and is set according to actual conditions.

[0057] Further, if the control device determines that the amount of data included in the fourth parameter set is greater than a preset threshold, the fourth parameter set and the above-mentioned first parameter set are combined together to determine a fifth parameter set.

[0058] Furthermore, the control device performs cluster analysis on the fifth parameter set, and updates or adjusts the speed torque throttle curve strategy according to the cluster analysis result. Specifically, before performing cluster analysis on the fifth parameter set, it also needs to be preprocessed, and the cluster analysis method is the same as the above cluster analysis method.

[0059] In this technical solution, when the amount of data of the newly collected vehicle operating parameters reaches a preset threshold, the control device is also used to merge the newly collected operating parameters with the first parameter set into a data set, and re-cluster the data set, and then re-determine the speed torque throttle curve strategy according to the new cluster analysis results or adjust the previous speed torque throttle curve strategy. In this way, the reliability of the speed torque throttle curve strategy is improved, thereby ensuring the stability of vehicle operation.

[0060] According to a second aspect of the present invention, a vehicle control device is proposed, comprising: an acquisition unit, used to acquire a first parameter set during the operation of the vehicle, the first parameter set including the operating parameters of the vehicle and time series data corresponding to the operating parameters; a processing unit, used to pre-process the first parameter set to determine a second parameter set; the processing unit is also used to perform cluster analysis on the second parameter set to determine an analysis result of the vehicle load condition; the processing unit is also used to determine a speed torque throttle curve strategy of the engine associated with the vehicle load condition based on the analysis result of the vehicle load condition.

[0061] In this technical solution, the first parameter set includes the vehicle's operating parameters and time series data corresponding to the operating parameters. The vehicle's operating parameters mainly include parameters such as load, mileage, fuel consumption, speed, etc., and the time series data is used to indicate the time series information when the above parameters are collected.

[0062] Specifically, the acquisition unit is used to acquire a first parameter set during the operation of the vehicle, and the processing unit is used to pre-process the first parameter set to determine a second parameter set. Specifically, the control device proposed in the present invention is used to analyze the vehicle load condition, and the first parameter set acquired by the acquisition unit includes not only the operating parameters of the vehicle load condition, but also other operating parameters, and abnormal data may also exist in the first parameter set. Therefore, after the first parameter set is acquired by the acquisition unit, it is also necessary to pre-process the first parameter set by the processing unit to determine a second parameter set with higher availability and effectiveness to ensure the smooth progress of subsequent steps.

[0063] Furthermore, the processing unit is also used to perform cluster analysis on the second parameter set, that is, to subdivide the data of the vehicle load condition, determine the number of interval divisions of the load condition and the segmented intervals of the load condition, and then determine the analysis results of the vehicle load condition.

[0064] Furthermore, the processing unit is also used to determine the speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis results of the determined vehicle load condition. Specifically, according to the analysis results of the vehicle load condition, the number of interval divisions of the vehicle load condition and the segmented intervals of the load condition can be understood, and different load condition intervals correspond to appropriate speed torque throttle curve strategies. Therefore, the processing unit can determine the speed torque throttle curve strategy that matches the load condition according to the analysis results of the vehicle load condition, thereby achieving precise control of the vehicle operation.

[0065] In the technical solution of the present invention, after the first parameter set in the vehicle operation process is obtained by the acquisition unit, the first parameter set is preprocessed by the processing unit to determine the second parameter set with higher availability and effectiveness. After determining the second parameter set, the processing unit performs cluster analysis on the second parameter set to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and the processing unit determines the speed torque throttle curve strategy corresponding to the segmented intervals and the number of divided intervals based on this information. In the technical solution of the present invention, the data of the vehicle load condition is subdivided by a clustering algorithm to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and then the optimal speed torque throttle curve strategy matching the load condition is determined, and the precise control of the vehicle operation is achieved according to the speed torque throttle curve strategy, which ensures the stability of the vehicle operation and saves the energy consumption of the vehicle operation.

[0066] According to a third aspect of the present invention, a readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the vehicle control method provided in the first aspect of the present invention are implemented. Therefore, the readable storage medium has all the beneficial effects of the vehicle control method provided in the first aspect of the present invention, which will not be described in detail here.

[0067] According to the fourth aspect of the present invention, a vehicle is proposed, comprising: a control device for the vehicle as proposed in the second aspect of the present invention, and / or a readable storage medium as proposed in the third aspect of the present invention. Therefore, the vehicle has all the beneficial effects of the control device for the vehicle proposed in the second aspect of the present invention and / or the readable storage medium proposed in the third aspect of the present invention, which will not be repeated here.

[0068] Additional aspects and advantages of the present invention will become apparent from the following description or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0070] Figure 1A schematic diagram showing a flow chart of a vehicle control method according to an embodiment of the present invention;

[0071] Figure 2 A second flow chart of a vehicle control method according to an embodiment of the present invention is shown;

[0072] Figure 3 A third flow chart of a vehicle control method according to an embodiment of the present invention is shown;

[0073] Figure 4 A fourth flow chart of a vehicle control method according to an embodiment of the present invention is shown;

[0074] Figure 5 A fifth flow chart of a vehicle control method according to an embodiment of the present invention is shown;

[0075] Figure 6 A sixth flow chart of a vehicle control method according to an embodiment of the present invention is shown;

[0076] Figure 7 FIG7 is a flow chart showing a method for controlling a vehicle according to an embodiment of the present invention;

[0077] Figure 8 FIG8 is a flow chart showing a method for controlling a vehicle according to an embodiment of the present invention;

[0078] Fig. 9 A ninth flowchart of a vehicle control method according to an embodiment of the present invention is shown;

[0079] Fig.10 FIG10 is a flowchart of a vehicle control method according to an embodiment of the present invention;

[0080] Fig.11 A schematic block diagram showing a control device for a vehicle according to an embodiment of the present invention;

[0081] Fig.12 A schematic block diagram of a vehicle according to an embodiment of the present invention is shown;

[0082] Fig.13 A schematic diagram showing the overall flow of a vehicle control method according to an embodiment of the present invention is shown;

[0083] Fig.14 A schematic diagram of a process of performing cluster analysis using a first clustering algorithm according to an embodiment of the present invention is shown;

[0084] Fig.15 A schematic diagram of a process of performing cluster analysis using a second clustering algorithm according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0085] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

[0086] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0087] Combine the following Figures 1 to 15 , the vehicle control method, device, readable storage medium and vehicle provided by the embodiments of the present invention are described in detail through specific embodiments and their application scenarios.

[0088] Embodiment 1:

[0089] Figure 1 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0090] Step S102, obtaining a first parameter set during vehicle operation, the first parameter set including vehicle operation parameters and time series data corresponding to the operation parameters;

[0091] Step S104, preprocessing the first parameter set to determine a second parameter set;

[0092] Step S106, performing cluster analysis on the second parameter set to determine the analysis result of the vehicle load condition;

[0093] Step S108, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0094] It should be noted that the executor of the vehicle control method proposed in this embodiment may be the control device of the vehicle. In order to more clearly illustrate the vehicle control method proposed in this embodiment, the following description will exemplify the executor of the vehicle control method as the control device of the vehicle.

[0095] In this embodiment, the first parameter set includes the vehicle's operating parameters and time series data corresponding to the operating parameters. The vehicle's operating parameters mainly include parameters such as load, mileage, fuel consumption, speed, etc., and the time series data is used to indicate the time series information when the above parameters are collected.

[0096] Specifically, the control device obtains a first parameter set during vehicle operation, and pre-processes the first parameter set to determine a second parameter set. Specifically, this embodiment is used for analyzing vehicle load conditions, and the first parameter set obtained includes not only operating parameters of vehicle load conditions, but also other operating parameters, and abnormal data may also exist in the first parameter set. Therefore, after obtaining the first parameter set, the control device needs to pre-process the first parameter set to determine a second parameter set with higher availability and effectiveness to ensure the smooth progress of subsequent steps.

[0097] Furthermore, the control device performs cluster analysis on the second parameter set, that is, subdivides the data of the vehicle load condition, determines the number of interval divisions of the load condition and the segmented intervals of the load condition, and then determines the analysis result of the vehicle load condition.

[0098] Furthermore, the control device determines the speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis result of the determined vehicle load condition. Specifically, according to the analysis result of the vehicle load condition, the number of interval divisions of the vehicle load condition and the segmented intervals of the load condition can be understood, and different load condition intervals correspond to appropriate speed torque throttle curve strategies. Therefore, the control device can accurately determine the speed torque throttle curve strategy that matches the load condition according to the analysis result of the vehicle load condition, thereby realizing accurate control of the vehicle operation.

[0099] In this embodiment, after obtaining the first parameter set during the operation of the vehicle, the control device preprocesses the first parameter set to determine a second parameter set with higher availability and effectiveness. After determining the second parameter set, the control device performs cluster analysis on the second parameter set to determine the segmented intervals and the number of divided intervals of the vehicle load condition. The control device determines the speed torque throttle curve strategy corresponding to the segmented intervals and the number of divided intervals for the vehicle based on this information. In this embodiment, the data of the vehicle load condition is subdivided by a clustering algorithm to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and then the optimal speed torque throttle curve strategy matching the load condition is determined. According to the speed torque throttle curve strategy, precise control of the vehicle operation is achieved, the stability of the vehicle operation is ensured, and the energy consumption of the vehicle operation can be saved.

[0100] Figure 2 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0101] Step S202, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0102] Step S204, removing abnormal data in the first parameter set, and / or replacing the abnormal data with preset parameters corresponding to the abnormal data, to determine a third parameter set;

[0103] Step S206, filtering out data related to the load condition in the third parameter set to determine the second parameter set;

[0104] Step S208, performing cluster analysis on the second parameter set to determine the analysis result of the vehicle load condition;

[0105] Step S210, determining a speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0106] In this embodiment, the specific steps of preprocessing the first parameter set are: the control device screens out abnormal data in the first parameter set, and directly removes the abnormal data in the first parameter set that has a low correlation with the vehicle load condition data; for the abnormal data in the first parameter set that has a high correlation with the vehicle load condition data, the abnormal data is replaced with a preset parameter corresponding to the abnormal data, and then a third parameter set without abnormal data is determined. Specifically, the above preset parameters can be calculated based on the non-abnormal data in the first parameter set, or can be set based on the historical parameters of the vehicle operation.

[0107] Furthermore, the control device screens the third parameter set, screens out data related to the load condition in the third parameter set, and determines the second parameter set. Specifically, the data related to the load condition mainly includes load parameters, mileage parameters, fuel consumption parameters, speed parameters, etc.

[0108] In this embodiment, since the first parameter may contain abnormal data and data irrelevant to the load condition, these data will affect the analysis results of the data in the subsequent steps, which is not conducive to cluster analysis and will lead to inaccurate cluster analysis results. Therefore, in this embodiment, the control device determines the third parameter set by removing or replacing the abnormal data in the first parameter with preset parameters, and then screens out the parameters related to the load in the third parameter set to determine the second parameter set with higher availability and effectiveness. In this way, the accuracy of the analysis results of the vehicle load condition determined by cluster analysis of the second parameter set is improved.

[0109] Figure 3 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0110] Step S302, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0111] Step S304, preprocessing the first parameter set to determine a second parameter set;

[0112] Step S306, analyzing the second parameter set according to the first clustering algorithm to determine the first target cluster number and the first load condition segmentation interval;

[0113] Step S308, analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clusters and the second load condition segmentation interval;

[0114] Step S310, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number and the second load condition segmentation interval;

[0115] Step S312, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0116] In this embodiment, the first clustering algorithm is a K-means clustering algorithm, and the second clustering algorithm is a Birch clustering algorithm.

[0117] Specifically, the control device performs cluster analysis on the second parameter set, and the steps of determining the analysis result of the vehicle load condition are as follows: the control device performs cluster analysis on the second parameter set according to the first clustering algorithm, and determines the first set of optimal number of cluster clusters and segmentation intervals, that is, the first target number of cluster clusters and the first load condition segmentation interval; the control device performs cluster analysis on the second parameter set according to the second clustering algorithm, and determines the second set of optimal number of cluster clusters and segmentation intervals, that is, the second target number of cluster clusters and the second load condition segmentation interval; the control device performs weighted average calculation on the above two sets of information, and obtains the final vehicle load condition segmentation interval and the number of interval divisions, and then determines the analysis result of the vehicle load condition.

[0118] In this embodiment, in the step of clustering the second parameter set, the control device uses two different clustering algorithms to cluster the second parameter set respectively, and performs weighted average calculation on the two clustering results to obtain the final analysis result of the vehicle load condition. In this way, the accuracy and credibility of the analysis result of the determined vehicle load condition are improved.

[0119] Figure 4 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0120] Step S402, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0121] Step S404, preprocessing the first parameter set to determine a second parameter set;

[0122] Step S406, determining first load condition data according to the second parameter set;

[0123] Step S408, compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data;

[0124] Step S410, performing cluster analysis on the first load condition data according to a first clustering algorithm to determine the number of first sub-clustering clusters and a first clustering result;

[0125] Step S412, performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the number of second sub-clustering clusters and the second clustering result;

[0126] Step S414: determining a first target cluster number and a first load condition segmentation interval according to the first sub-cluster number, the first clustering result, the second sub-cluster number, and the second clustering result.

[0127] Step S416, analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clusters and the second load condition segmentation interval;

[0128] Step S418, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0129] Step S420, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0130] In this embodiment, the first load condition data is used to indicate the data determined by screening and processing the operating parameters of the vehicle in the second parameter set that has been continuously driven for more than a certain period of time. It can be understood that the first clustering algorithm (i.e., the K-means clustering algorithm) is more suitable for clustering analysis of small and medium-sized data sets. Therefore, in this embodiment, before performing cluster analysis through the first clustering algorithm, the control device needs to screen out data with higher reference value in the second parameter set, that is, reduce the size of the data set to increase the rate of cluster analysis.

[0131] Specifically, the step of the control device analyzing the second parameter set according to the first clustering algorithm is: the control device determines the first load condition data according to the second parameter set, and performs dimensionality reduction and compression processing on the first load condition data to determine the second load condition data with higher simplicity.

[0132] Specifically, the control device performs dimensionality reduction and compression processing on the first load condition data by filtering out data other than the load data in the first load condition data, compressing these data into a column of data through principal component analysis, and then combining this column of data with the load data to determine the second load condition data.

[0133] Furthermore, the control device performs cluster analysis on the first load condition data according to the first clustering algorithm to determine the first clustering result and the optimal number of clustering clusters, that is, the first clustering result and the first sub-clustering cluster number; the control device performs cluster analysis on the second load condition data according to the first clustering algorithm to determine the second clustering result and the optimal number of clustering clusters, that is, the second sub-clustering cluster number and the second clustering result; the control device determines the first target clustering cluster number and the first load condition segmentation interval by combining the first and second clustering results and the optimal number of clustering clusters.

[0134] In this embodiment, in the step of clustering the second parameter set according to the first clustering algorithm, the control device performs clustering analysis on the first load condition data and the second load condition data after dimensionality reduction and compression respectively to obtain two groups of analysis results (i.e., clustering is performed twice using the first clustering algorithm), and determines the first target cluster number and the first load condition segmentation interval by combining the above two groups of analysis results. In this way, the accuracy and credibility of the analysis results of the vehicle load condition determined in the subsequent process are improved.

[0135] Figure 5 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0136] Step S502, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0137] Step S504, preprocessing the first parameter set to determine a second parameter set;

[0138] Step S506, extracting a plurality of motion segment data in the second parameter set, where the motion segment data is used to indicate the running parameters of the vehicle when the vehicle continuously travels for more than a first preset time period;

[0139] Step S508, processing the plurality of motion segment data into a plurality of single-line data, and determining first load condition data according to the plurality of single-line data;

[0140] Step S510, compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data;

[0141] Step S512, performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the number of first sub-clustering clusters and the first clustering result;

[0142] Step S514, performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the number of second sub-clustering clusters and the second clustering result;

[0143] Step S516, determining a first target cluster number and a first load condition segmentation interval according to the first sub-cluster number, the first clustering result, the second sub-cluster number and the second clustering result;

[0144] Step S518, analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clusters and the second load condition segmentation interval;

[0145] Step S520, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0146] Step S522, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0147] In this embodiment, the motion segment data is used to indicate the operating parameters of the vehicle collected when the vehicle continuously travels for more than a first preset time period. The first preset time period is generally 1 hour, which is determined based on the actual situation of the vehicle.

[0148] Specifically, the control device selects multiple motion segment data in the second parameter set whose continuous driving time of the vehicle exceeds the first preset time, processes the multiple motion segment data into multiple single-line data, and determines the first load condition data based on the multiple single-line data. Specifically, the control device combines multiple single-line data into a single-line data by constructing business features and multi-dimensional data binning features.

[0149] In this embodiment, considering that there are running parameters of vehicles with shorter continuous driving time in the second parameter set, these data are of low reference value and will affect the rate of cluster analysis by the first clustering algorithm. Therefore, in this embodiment, before cluster analysis is performed by the first clustering algorithm, the control device needs to screen out multiple motion segment data in the second parameter set whose continuous driving time of the vehicle exceeds the first preset time, and process them to determine the first load condition data suitable for cluster analysis. In this way, the robustness of the subsequent steps is guaranteed, thereby improving the accuracy of the analysis results of the vehicle load condition.

[0150] Figure 6 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0151] Step S602, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0152] Step S604, preprocessing the first parameter set to determine a second parameter set;

[0153] Step S606, determining first load condition data according to the second parameter set;

[0154] Step S608, compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data;

[0155] Step S610, setting initial parameters of a first clustering algorithm, setting the number of clusters to a first preset number of clusters, and performing cluster analysis on the first load condition data;

[0156] Step S612, sequentially increasing the number of clusters until the number of clusters is greater than the second preset number of clusters; drawing a silhouette coefficient graph when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters;

[0157] Step S614, determining the cluster number with the largest silhouette coefficient in the silhouette coefficient graph as the first sub-clustering cluster number, and taking the clustering result corresponding to the first sub-clustering cluster number as the first clustering result;

[0158] Step S616, performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the number of second sub-clustering clusters and the second clustering result;

[0159] Step S618, determining a first target cluster number and a first load condition segmentation interval according to the first sub-cluster number, the first clustering result, the second sub-cluster number and the second clustering result;

[0160] Step S620, analyzing the second parameter set according to the second clustering algorithm to determine a second target number of clusters and a second load condition segmentation interval;

[0161] Step S622, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0162] Step S624, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0163] In this embodiment, the above-mentioned number of clusters is used to indicate a control parameter for adjusting the complexity of the clustering algorithm; the above-mentioned silhouette coefficient is an indicator for evaluating the clustering effect in the clustering algorithm, and the larger the silhouette coefficient, the better the clustering effect.

[0164] Specifically, the control device performs cluster analysis on the first load condition data according to the first clustering algorithm as follows: the control device sets the initial parameters of the first clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the first load condition data into the first clustering algorithm for cluster analysis.

[0165] It should be noted that the initial parameters of the first clustering algorithm set by the control device also include a random seed and the number of iterations. These two parameters are set according to actual conditions and are not limited in this embodiment.

[0166] Further, the control device sequentially adjusts the number of clusters from the first preset number of clusters to a number greater than the second preset number of clusters, and draws a silhouette coefficient diagram where the number of clusters is equal to the first preset number of clusters to the second preset number of clusters. Specifically, the clustering effect of different numbers of clusters can be intuitively determined through the silhouette coefficient diagram, so in the technical solution of the present invention, the control device needs to draw a silhouette coefficient diagram based on the silhouette coefficient where the number of clusters is equal to the first preset number of clusters to the second preset number of clusters.

[0167] Furthermore, the control device compares the silhouette coefficients in the silhouette coefficient graph, determines the cluster number with the largest silhouette coefficient in the graph as the first sub-clustering cluster number, and uses the clustering result corresponding to the first sub-clustering cluster number as the first clustering result.

[0168] Figure 7 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0169] Step S702, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0170] Step S704, preprocessing the first parameter set to determine a second parameter set;

[0171] Step S706, determining the first load condition data according to the second parameter set;

[0172] Step S708, compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data;

[0173] Step S710, performing cluster analysis on the first load condition data according to a first clustering algorithm to determine the number of first sub-clustering clusters and a first clustering result;

[0174] Step S712, setting initial parameters of the first clustering algorithm, setting the number of clusters to a first preset number of clusters, and performing cluster analysis on the second load condition data;

[0175] Step S714, sequentially increasing the number of clusters until the number of clusters is greater than the second preset number of clusters, and determining an error square sum curve according to the error square sum when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters;

[0176] Step S716, determining the cluster number corresponding to the inflection point in the error square sum curve as the second sub-clustering cluster number, and taking the clustering result corresponding to the second sub-clustering cluster number as the second clustering result;

[0177] Step S718, determining a first target cluster number and a first load condition segmentation interval according to the first sub-cluster number, the first clustering result, the second sub-cluster number and the second clustering result;

[0178] Step S720, analyzing the second parameter set according to the second clustering algorithm to determine a second target number of clusters and a second load condition segmentation interval;

[0179] Step S722, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0180] Step S724, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0181] In this embodiment, the above-mentioned error square sum is an indicator for evaluating the clustering effect in the clustering algorithm. Specifically, as the number of clusters increases, the sample division will be more refined and the error square sum will decrease. When the number of clusters is less than the optimal number of clusters, the increase in the number of clusters will greatly increase the degree of aggregation of each cluster, so the error square sum will decrease greatly. When the number of clusters reaches the optimal value, the return of the degree of aggregation obtained by increasing the number of clusters will decrease rapidly, the error square sum will decrease sharply, and then the continued increase in the number of clusters will tend to be flat. It can be seen that the optimal number of clusters can be directly determined based on the inflection point in the error square sum curve.

[0182] Specifically, the control device performs cluster analysis on the second load condition data according to the first clustering algorithm as follows: the control device sets the initial parameters of the first clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the second load condition data into the first clustering algorithm for cluster analysis.

[0183] Furthermore, the control device sequentially adjusts the number of clusters from a first preset number of clusters to a number greater than a second preset number of clusters, calculates the sum of square errors when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters, and draws a sum of square errors curve based on the sum of square errors.

[0184] Furthermore, the number of clusters corresponding to the inflection point in the error square sum curve is determined as the second sub-clustering number, and the clustering result corresponding to the second sub-clustering number is used as the second clustering result.

[0185] Figure 8 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0186] Step S802, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0187] Step S804, preprocessing the first parameter set to determine a second parameter set;

[0188] Step S806, analyzing the second parameter set according to the first clustering algorithm to determine the first target cluster number and the first load condition segmentation interval;

[0189] Step S808, compressing the data except the load data in the second parameter set into a column of data by principal component analysis, and determining the third load condition data according to the column of data and the load data in the second parameter set;

[0190] Step S810, performing cluster analysis on the third load condition data according to the second clustering algorithm to determine a second target cluster number and a second load condition segmentation interval;

[0191] Step S812, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0192] Step S814, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0193] In this embodiment, the control device analyzes the second parameter set according to the second clustering algorithm as follows: the control device directly performs dimensionality reduction and compression processing on the second parameter set to determine the third load condition data with higher simplicity. It can be understood that the second clustering algorithm (i.e., the Birch clustering algorithm) is more suitable for analyzing large-scale data sets. Therefore, in this embodiment, before performing clustering analysis through the second clustering algorithm, the control device does not need to screen the second parameter set, but can directly perform dimensionality reduction and compression processing on it. In this way, the steps of clustering analysis are simplified, and the comprehensiveness of the data of clustering analysis is improved.

[0194] Specifically, the process of dimensionality reduction and compression processing on the second parameter set is to filter out the data except the load data in the second parameter set, compress these data into a column of data through principal component analysis, and then combine this column of data with the load data to determine the third load condition data.

[0195] Furthermore, the control device performs cluster analysis on the third load condition data according to the second clustering algorithm to determine a second target number of cluster clusters and a second load condition segmentation interval.

[0196] In this embodiment, the control device does not directly perform cluster analysis on the second parameter set, but performs cluster analysis on the third load condition data determined according to the dimensionality reduction and compression processing of the second parameter set, thereby reducing the amount of data for cluster analysis and improving the rate of cluster analysis.

[0197] Fig. 9 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0198] Step S902, obtaining a first parameter set during the operation of the vehicle, the first parameter set including the operation parameters of the vehicle and time series data corresponding to the operation parameters;

[0199] Step S904, preprocessing the first parameter set to determine a second parameter set;

[0200] Step S906, analyzing the second parameter set according to the first clustering algorithm to determine the first target cluster number and the first load condition segmentation interval;

[0201] Step S908, compressing the data except the load data in the second parameter set into a column of data by principal component analysis, and determining the third load condition data according to the column of data and the load data in the second parameter set;

[0202] Step S910, setting initial parameters of the second clustering algorithm, setting the number of clusters to the first preset number of clusters, and performing cluster analysis on the third load condition data;

[0203] Step S912, sequentially increasing the number of clusters until the number of clusters is greater than the second preset number of clusters;

[0204] Step S914, comparing the variance ratio standards of the clustering number equal to the first preset cluster number to the second preset cluster number, determining the clustering number corresponding to the largest variance ratio standard as the second target clustering number, and determining the second load condition segmentation interval according to the clustering result corresponding to the second target clustering number;

[0205] Step S916, determining the analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval;

[0206] Step S918, determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition.

[0207] In this embodiment, the variance ratio standard is an indicator for evaluating the clustering effect in the clustering algorithm. Specifically, the larger the variance ratio standard, the better the clustering effect.

[0208] Specifically, the control device performs cluster analysis on the third load condition data according to the second clustering algorithm as follows: the control device sets the initial parameters of the second clustering algorithm, sets the number of clusters to a first preset number of clusters, and then substitutes the third load condition data into the second clustering algorithm for cluster analysis.

[0209] It should be noted that the initial parameters of the second clustering algorithm set by the control device also include a threshold and a calculation label. These two parameters are set according to actual conditions, and the present invention does not limit them here.

[0210] Further, the control device sequentially adjusts the number of clusters from the first preset number of clusters to greater than the second preset number of clusters, and respectively determines the variance ratio standard when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters. Specifically, the larger the variance ratio standard, the tighter the class itself is, and the more dispersed the different classes are, that is, the better the clustering effect is. Therefore, in this embodiment, it is necessary to respectively determine the variance ratio standard when the number of clusters is equal to the first preset number of clusters to the second preset number of clusters during the cluster analysis process to evaluate the clustering effect.

[0211] Furthermore, the control device compares the variance ratio standards determined in the above steps, determines the number of clusters corresponding to the largest variance ratio standard as the second target number of clusters, and determines the second load condition segmentation interval according to the clustering result corresponding to the second target number of clusters.

[0212] Fig.10 A schematic flow chart of a vehicle control method according to an embodiment of the present invention is shown, wherein the control method comprises:

[0213] Step S1002, obtaining a first parameter set during vehicle operation, the first parameter set including vehicle operation parameters and time series data corresponding to the operation parameters;

[0214] Step S1004, preprocessing the first parameter set to determine a second parameter set;

[0215] Step S1006, performing cluster analysis on the second parameter set to determine the analysis result of the vehicle load condition;

[0216] Step S1008, determining a speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis result of the vehicle load condition;

[0217] Step S1010, confirming whether the data volume of the fourth parameter set is greater than a preset threshold, the fourth parameter set being used to indicate the newly collected operating parameters of the vehicle;

[0218] Step S1012, when the data amount of the fourth parameter set is greater than a preset threshold, determining a fifth parameter set according to the fourth parameter set and the first parameter set;

[0219] Step S1014: performing cluster analysis on the fifth parameter set, and updating or adjusting the speed torque throttle curve strategy according to the cluster analysis result of the fifth parameter set.

[0220] In this embodiment, the fourth parameter set is used to indicate newly collected operating parameters of the vehicle. It can be understood that during the operation of the vehicle, the control device collects the operating parameters of the vehicle in real time.

[0221] Specifically, the control device determines whether the amount of data contained in the fourth parameter set (i.e., the newly collected vehicle operating parameters) is greater than a preset threshold. Specifically, the preset threshold is generally set to 10% of the amount of data in the first parameter set (i.e., the data used for cluster analysis), and is set according to actual conditions.

[0222] Further, if the control device determines that the amount of data included in the fourth parameter set is greater than a preset threshold, the fourth parameter set and the above-mentioned first parameter set are combined together to determine a fifth parameter set.

[0223] Furthermore, the control device performs cluster analysis on the fifth parameter set, and updates or adjusts the speed torque throttle curve strategy according to the cluster analysis result. Specifically, before performing cluster analysis on the fifth parameter set, it also needs to be preprocessed, and the cluster analysis method is the same as the above cluster analysis method.

[0224] In this embodiment, when the amount of data of the newly collected vehicle operating parameters reaches a preset threshold, the control device is further used to merge the newly collected operating parameters with the first parameter set into a data set, and re-cluster the data set, and then re-determine the speed torque throttle curve strategy according to the new cluster analysis result or adjust the previous speed torque throttle curve strategy. In this way, the reliability of the speed torque throttle curve strategy is improved, thereby ensuring the stability of vehicle operation.

[0225] Embodiment 2:

[0226] Fig.11 A schematic block diagram of a control device of a vehicle according to an embodiment of the present invention is shown, wherein the control device 1100 of the vehicle includes: an acquisition unit 1102, used to acquire a first parameter set during vehicle operation, the first parameter set including vehicle operating parameters and time series data corresponding to the operating parameters; a processing unit 1104, used to pre-process the first parameter set to determine a second parameter set; the processing unit 1104 is also used to perform cluster analysis on the second parameter set to determine an analysis result of the vehicle load condition; the processing unit 1104 is also used to determine a speed torque throttle curve strategy of the engine associated with the vehicle load condition based on the analysis result of the vehicle load condition.

[0227] In this embodiment, the first parameter set includes the vehicle's operating parameters and time series data corresponding to the operating parameters. The vehicle's operating parameters mainly include parameters such as load, mileage, fuel consumption, speed, etc., and the time series data is used to indicate the time series information when the above parameters are collected.

[0228] Specifically, the acquisition unit 1102 is used to acquire a first parameter set during the operation of the vehicle, and the processing unit 1104 is used to pre-process the first parameter set to determine a second parameter set. Specifically, the control device proposed in the present invention is used to analyze the vehicle load condition, and the first parameter set acquired by the acquisition unit 1102 includes not only the operating parameters of the vehicle load condition, but also other operating parameters, and abnormal data may also exist in the first parameter set. Therefore, after the first parameter set is acquired by the acquisition unit 1102, the first parameter set needs to be pre-processed by the processing unit 1104 to determine a second parameter set with higher availability and effectiveness to ensure the smooth progress of subsequent steps.

[0229] Furthermore, the processing unit 1104 is also used to perform cluster analysis on the second parameter set, that is, to subdivide the data of the vehicle load condition, determine the number of interval divisions of the load condition and the segmented intervals of the load condition, and then determine the analysis results of the vehicle load condition.

[0230] Furthermore, the processing unit 1104 is also used to determine the speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis results of the determined vehicle load condition. Specifically, according to the analysis results of the vehicle load condition, the number of interval divisions of the vehicle load condition and the segmented intervals of the load condition can be understood, and different load condition intervals correspond to appropriate speed torque throttle curve strategies. Therefore, the processing unit 1104 can determine the speed torque throttle curve strategy that matches the load condition according to the analysis results of the vehicle load condition, thereby achieving precise control of the vehicle operation.

[0231] In this embodiment, after the first parameter set during the operation of the vehicle is obtained by the acquisition unit 1102, the first parameter set is preprocessed by the processing unit 1104 to determine a second parameter set with higher availability and effectiveness. After determining the second parameter set, the processing unit 1104 performs cluster analysis on the second parameter set to determine the segmented intervals and the number of divided intervals of the vehicle load condition. The processing unit 1104 determines the speed torque throttle curve strategy corresponding to the segmented intervals and the number of divided intervals based on this information. In this embodiment, the data of the vehicle load condition is subdivided by a clustering algorithm to determine the segmented intervals and the number of divided intervals of the vehicle load condition, and then the optimal speed torque throttle curve strategy matching the load condition is determined. According to the speed torque throttle curve strategy, precise control of the vehicle operation is achieved, the stability of the vehicle operation is ensured, and the energy consumption of the vehicle operation can be saved.

[0232] Furthermore, in this embodiment, in the step of preprocessing the first parameter set to determine the second parameter set, the processing unit 1104 is also used to remove abnormal data in the first parameter set, and / or replace the abnormal data with preset parameters corresponding to the abnormal data to determine the third parameter set; and filter out the data related to the load condition in the third parameter set to determine the second parameter set.

[0233] Furthermore, in this embodiment, in the step of performing cluster analysis on the second parameter set to determine the analysis results of the vehicle load condition, the processing unit 1104 is also used to analyze the second parameter set according to the first clustering algorithm to determine the first target number of clustering clusters and the first load condition segmentation interval; analyze the second parameter set according to the second clustering algorithm to determine the second target number of clustering clusters and the second load condition segmentation interval; determine the analysis results of the vehicle load condition based on the first target number of clustering clusters, the first load condition segmentation interval, the second target number of clustering clusters and the second load condition segmentation interval.

[0234] Further, in this embodiment, in the step of analyzing the second parameter set according to the first clustering algorithm to determine the first target clustering cluster number and the first load condition segmentation interval, the processing unit 1104 is also used to determine the first load condition data according to the second parameter set; compress the data other than the load data in the first load condition data into a column of data through principal component analysis, and determine the second load condition data based on the column of data and the load data in the first load condition data; perform cluster analysis on the first load condition data according to the first clustering algorithm to determine the first sub-clustering cluster number and the first clustering result; perform cluster analysis on the second load condition data according to the first clustering algorithm to determine the second sub-clustering cluster number and the second clustering result; determine the first target clustering cluster number and the first load condition segmentation interval based on the first sub-clustering cluster number, the first clustering result, the second sub-clustering cluster number and the second clustering result.

[0235] Furthermore, in this embodiment, in the step of determining the first load condition data according to the second parameter set, the processing unit 1104 is also used to extract multiple motion segment data in the second parameter set, the motion segment data being used to indicate operating parameters of the vehicle that has been running continuously for more than a first preset time period; the multiple motion segment data are processed into multiple single-line data, and the first load condition data is determined based on the multiple single-line data.

[0236] Furthermore, in this embodiment, in the step of performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the first sub-clustering cluster number and the first clustering result, the processing unit 1104 is also used to set the initial parameters of the first clustering algorithm, set the clustering cluster number to the first preset cluster number, and perform cluster analysis on the first load condition data; increase the clustering cluster number in sequence until the clustering cluster number is greater than the second preset cluster number; draw a silhouette coefficient graph with a clustering cluster number equal to the first preset cluster number to the second preset cluster number; determine the clustering cluster number with the largest silhouette coefficient in the silhouette coefficient graph as the first sub-clustering cluster number, and use the clustering result corresponding to the first sub-clustering cluster number as the first clustering result.

[0237] Furthermore, in this embodiment, in the step of performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the second sub-clustering cluster number and the second clustering result, the processing unit 1104 is also used to set the initial parameters of the first clustering algorithm, set the clustering cluster number to the first preset cluster number, and perform cluster analysis on the second load condition data; increase the clustering cluster number in sequence until the clustering cluster number is greater than the second preset cluster number; determine the error square sum curve based on the error square sum when the clustering cluster number is equal to the first preset cluster number to the second preset cluster number; determine the clustering cluster number corresponding to the inflection point in the error square sum curve as the second sub-clustering cluster number, and use the clustering result corresponding to the second sub-clustering cluster number as the second clustering result.

[0238] Furthermore, in this embodiment, in the step of analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clustering clusters and the second load condition segmentation interval, the processing unit 1104 is also used to compress the data other than the load data in the second parameter set into a column of data through principal component analysis, and determine the third load condition data based on the column of data and the load data in the second parameter set; perform cluster analysis on the third load condition data according to the second clustering algorithm to determine the second target number of clustering clusters and the second load condition segmentation interval.

[0239] Furthermore, in this embodiment, in the step of performing cluster analysis on the third load condition data according to the second clustering algorithm to determine the second target number of cluster clusters and the second load condition segmentation interval, the processing unit 1104 is also used to set the initial parameters of the second clustering algorithm, set the number of cluster clusters to the first preset number of clusters, and perform cluster analysis on the third load condition data; increase the number of cluster clusters in sequence until the number of cluster clusters is greater than the second preset number of clusters; compare the variance ratio standards of the number of cluster clusters equal to the first preset number of clusters to the second preset number of clusters, and determine the number of cluster clusters corresponding to the largest variance ratio standard as the second target number of cluster clusters; and determine the second load condition segmentation interval based on the clustering results corresponding to the second target number of cluster clusters.

[0240] Furthermore, in this embodiment, after determining the speed torque throttle curve strategy of the engine associated with the vehicle load condition based on the analysis results of the vehicle load condition, the processing unit 1104 is also used to confirm whether the data amount of the fourth parameter set is greater than a preset threshold, and the fourth parameter set is used to indicate the newly collected operating parameters of the vehicle; when the data amount of the fourth parameter set is greater than the preset threshold, the fifth parameter set is determined based on the fourth parameter set and the first parameter set; the fifth parameter set is clustered and analyzed, and the speed torque throttle curve strategy is updated or adjusted based on the cluster analysis results of the fifth parameter set.

[0241] Embodiment three:

[0242] According to a third embodiment of the present invention, a readable storage medium is provided, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the vehicle control method provided in the above embodiment are implemented. Therefore, the readable storage medium has all the beneficial effects of the vehicle control method provided in the above embodiment, which will not be described in detail here.

[0243] Embodiment 4:

[0244] Fig.12A schematic block diagram of a vehicle according to an embodiment of the present invention is shown. The vehicle 1200 includes the vehicle control device 1100 as proposed in the above embodiment, and / or the readable storage medium 1202 as proposed in the above embodiment. Therefore, the vehicle 1200 has all the beneficial effects of the vehicle control device 1100 proposed in the above embodiment or the readable storage medium 1202 proposed in the above embodiment, which will not be repeated here.

[0245] Embodiment five:

[0246] This embodiment combines Figures 13 to 15 The vehicle control method proposed in the present invention is exemplarily described.

[0247] like Fig.13 As shown, the overall process of the vehicle control method proposed in this embodiment is:

[0248] Step S1302, data collection;

[0249] Step S1304, outlier data processing;

[0250] Step S1306, filtering the time series data column;

[0251] Step S1308, extracting motion segments;

[0252] Step S1310, feature compression and dimensionality reduction;

[0253] Step S1312, Kmeans cluster analysis of load category quantity;

[0254] Step S1314, Birch cluster analysis of load category quantity;

[0255] Step S1316, clustering effect evaluation;

[0256] Step S1318, determining the optimal load cluster number and cluster segmentation points;

[0257] Step S1320, load condition map strategy matching;

[0258] Step S1322, load strategy application;

[0259] Step S1324, newly added operation data is saved;

[0260] Step S1326, whether the accumulated data exceeds 10% of the analyzed data; if yes, return to step S1306, if not, go to step S1322.

[0261] In this embodiment, the control device performs data collection (ie, obtains the first parameter set). Specifically, the collected data includes operating parameters of the vehicle and time series data corresponding to the operating parameters.

[0262] Furthermore, the control device performs outlier processing on the collected data, and filters out data related to the load condition based on the time series data column to determine the second parameter set.

[0263] Furthermore, the control device extracts motion segments and performs feature compression and dimensionality reduction on the second parameter set to obtain first load condition data, and performs cluster analysis on the first load condition data using a Kmeans clustering algorithm to determine a first cluster analysis result.

[0264] Further, the control device performs cluster analysis on the second parameter set by using a Birch clustering algorithm to determine a second cluster analysis result.

[0265] Furthermore, the clustering effect of the two clustering analysis results obtained by the first clustering algorithm (Kmeans clustering algorithm) and the second clustering algorithm (Birch clustering algorithm) is evaluated to determine the optimal load condition clustering data (i.e., load condition interval score) and clustering segmentation points (i.e., load condition segmentation interval), and then determine the analysis results of the load condition.

[0266] Furthermore, the control device matches the analysis result of the load condition with a map (engine speed torque throttle curve of the vehicle) to determine a speed torque throttle curve strategy of the engine associated with the load condition.

[0267] Furthermore, when it is determined that the amount of data collected about the operating parameters of the newly added vehicle exceeds 10% of the data used for cluster analysis, cluster analysis is performed again, and based on the new cluster analysis results, the engine speed torque throttle curve strategy associated with the load condition is updated and adjusted.

[0268] like Fig.14 As shown, in this embodiment, the specific steps of using the Kmeans clustering algorithm to perform cluster analysis on the first load condition data are:

[0269] Step S1402, obtaining load condition data;

[0270] Step S1404, all data columns except the load are reduced to one column by PCA;

[0271] Step S1406, combining the PCA dimension reduction data with the load column data;

[0272] Step S1408, k-means++ selects the mean vector of the initial clustering;

[0273] Step S1410, the number of clusters K is determined (initial K=2);

[0274] Step S1412, Random_sate=0, max_iter=300;

[0275] Step S1414, the K-means cluster center point is saved and each sample label is saved and output;

[0276] Step S1416, clustering effect evaluation: error sum of squares SSE calculation;

[0277] Step S1418, the clustering result center point and each sample label are saved and output;

[0278] Step S1420, clustering effect evaluation: drawing a silhouette coefficient graph;

[0279] Step S1422, K=K+1;

[0280] Step S1424, determine whether K is greater than 12; if yes, execute steps S1426 and S1428; if no, return to execute step S1410;

[0281] Step S1426, plotting the square sum of error SSE curve to determine the K value of the inflection point;

[0282] Step S1428, comparing the contour curves of all clusters to determine the optimal K value;

[0283] Step S1430, comprehensively determine the optimal number of clusters and load segmentation intervals.

[0284] In this embodiment, Fig.14 The medium load condition data corresponds to the first load condition data mentioned above. The control device compresses all data columns except the load data in the first load condition data into one column through PCA (Principal Component Analysis), and then combines this column of data with the load data to determine the second load condition data.

[0285] Furthermore, k-means++ selects the mean vector of the initial clustering, that is, the first load condition data is directly clustered by the Kmeans clustering algorithm. Specifically, the number of initial clusters is set to 2, random_satate (random seed) is set to 0, and max_iter (number of iterations) is set to 300. The center point and label results of each sample in the clustering process are saved, and the silhouette coefficient is used to evaluate the clustering effect.

[0286] Furthermore, the number of clusters is increased from K=2 to 12 in the manner of K=K+1, a silhouette coefficient graph of the number of clusters K=2 to K=12 is drawn, and the number of clusters with the largest silhouette coefficient in the silhouette coefficient graph is determined as the first optimal number of clusters.

[0287] Furthermore, the Kmeans clustering algorithm is used to perform cluster analysis on the data of the second load condition, and the number of initial clusters is set to 2, random_satate is set to 0, and max_iter is set to 300. The center point and the label result of each sample in the clustering process are saved, and the SSE (Sum of Squared Error) is used to evaluate the clustering effect.

[0288] Furthermore, the number of clusters is increased from K=2 to 12 in the manner of K=K+1, the sum of squared errors of the number of clusters K=2 to K=12 is determined, and the sum of squared error curve of the number of clusters K=2 to K=12 is plotted, and the number of clusters corresponding to the inflection point in the sum of squared error curve is determined as the second optimal number of clusters.

[0289] Furthermore, the control device comprehensively determines the first optimal number of clusters and load segmentation intervals based on the above-mentioned first optimal number of clusters and the second optimal number of clusters and their corresponding clustering results.

[0290] like Fig.15 As shown, in this embodiment, the specific steps of using the Birch clustering algorithm to perform cluster analysis on the second parameter set are:

[0291] Step S1502, obtaining load condition data;

[0292] Step S1504, all data columns except the load are reduced to one column by PCA;

[0293] Step S1506, combining the PCA dimension reduction data with the load column data;

[0294] Step S1508, the number of clusters n_clusters is determined (initial n_clusters=2);

[0295] Step S1510, threshold=0.5, compute_labels=True;

[0296] Step S1512, the clustering result center point and each sample label are saved and output;

[0297] Step S1514, clustering effect evaluation: variance ratio standard CHI calculation;

[0298] Step S1516, n_clusters=n_clusters+1;

[0299] Step S1518, determine whether n_clusters is greater than 12; if yes, execute step S1520; if no, return to execute step S1508;

[0300] Step S1520, n_clusters with the highest CHI score are determined;

[0301] Step S1522, output the optimal number of clusters and load segmentation intervals.

[0302] In this implementation, Fig.15 The medium load operating condition data corresponds to the above-mentioned second parameter set. The control device filters out the data except the load data in the second parameter set, compresses these data into a column of data through PCA, and then combines this column of data with the load data to determine the third load operating condition data.

[0303] Furthermore, the Birch clustering algorithm is used to perform cluster analysis on the data of the third load condition, and the initial number of clusters n_clusters is set to 2, the threshold is set to 0.5, and compute_labels is set to True. The center point of the clustering process and the label result of each sample are saved, and the clustering effect is evaluated using CHI (variance ratio standard).

[0304] Furthermore, the number of clusters is increased from K=2 to 12 in the manner of K=K+1, the variance ratio standards of the number of clusters K=2 to K=12 are compared, and the number of clusters corresponding to the variance ratio standard with the highest score is determined as the optimal number of clusters.

[0305] Furthermore, the control device determines a second optimal number of clusters and load segment intervals based on the determined optimal number of clusters.

[0306] In the description of this specification, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance unless otherwise clearly specified and limited; the terms "connection", "installation", "fixation" and the like should be understood in a broad sense, for example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0307] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "specific embodiments", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0308] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0309] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for controlling a vehicle, It is characterized in that include: Acquire a first parameter set during vehicle operation, the first parameter set including vehicle operation parameters and time series data corresponding to the operation parameters; Preprocessing the first parameter set to determine a second parameter set; Performing cluster analysis on the second parameter set to determine an analysis result of the vehicle load condition; Determining a speed torque throttle curve strategy of an engine associated with the vehicle load condition according to the analysis result of the vehicle load condition; The cluster analysis of the second parameter set to determine the analysis result of the vehicle load condition specifically includes: Analyzing the second parameter set according to a first clustering algorithm to determine a first target cluster number and a first load condition segmentation interval; Analyzing the second parameter set according to a second clustering algorithm to determine a second target number of clusters and a second load condition segmentation interval; Determining an analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval; The analyzing the second parameter set according to the first clustering algorithm to determine the first target cluster number and the first load condition segmentation interval specifically includes: Determining first load condition data according to the second parameter set; Compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data; Performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the number of first sub-clustering clusters and a first clustering result; Performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the number of second sub-clustering clusters and a second clustering result; The first target cluster number and the first load condition segmentation interval are determined according to the first sub-clustering cluster number, the first clustering result, the second sub-clustering cluster number and the second clustering result.

2. The vehicle control method according to claim 1, It is characterized in that The preprocessing of the first parameter set to determine the second parameter set specifically includes: removing abnormal data in the first parameter set, and / or replacing the abnormal data with preset parameters corresponding to the abnormal data, to determine a third parameter set; The data related to the load condition in the third parameter set is filtered out to determine the second parameter set.

3. The vehicle control method according to claim 1, It is characterized in that The determining the first load condition data according to the second parameter set specifically includes: Extracting a plurality of motion segment data from the second parameter set, the motion segment data being used to indicate operating parameters of a vehicle that has been continuously driven for more than a first preset time period; The plurality of motion segment data are processed into a plurality of single-line data, and the first load condition data are determined according to the plurality of single-line data.

4. The vehicle control method according to claim 1, It is characterized in that The performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the number of first sub-clustering clusters and a first clustering result specifically includes: Setting initial parameters of the first clustering algorithm, setting the number of clusters to a first preset number of clusters, and performing cluster analysis on the first load condition data; Sequentially increasing the number of clusters until the number of clusters is greater than a second preset number of clusters; Draw a silhouette coefficient graph where the number of clusters is equal to the first preset number of clusters to the second preset number of clusters; The cluster number with the largest silhouette coefficient in the silhouette coefficient graph is determined as the first sub-cluster number, and the clustering result corresponding to the first sub-cluster number is used as the first clustering result.

5. The vehicle control method according to claim 1, It is characterized in that The performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the second sub-clustering cluster number and the second clustering result specifically includes: Setting initial parameters of the first clustering algorithm, setting the number of clusters to a first preset number of clusters, and performing cluster analysis on the second load condition data; Sequentially increasing the number of clusters until the number of clusters is greater than a second preset number of clusters; Determine a square error curve according to the square error of the cluster number being equal to the first preset number of clusters to the second preset number of clusters; The cluster number corresponding to the inflection point in the error square sum curve is determined as the second sub-cluster number, and the clustering result corresponding to the second sub-cluster number is used as the second clustering result.

6. The vehicle control method according to claim 1, It is characterized in that The analyzing the second parameter set according to the second clustering algorithm to determine the second target number of clusters and the second load condition segmentation interval specifically includes: Compressing the data other than the load data in the second parameter set into a column of data by using a principal component analysis method, and determining the third load condition data according to the column of data and the load data in the second parameter set; The third load condition data is clustered and analyzed according to the second clustering algorithm to determine a second target number of clusters and a second load condition segmentation interval.

7. The vehicle control method according to claim 6, It is characterized in that The performing cluster analysis on the third load condition data according to the second clustering algorithm to determine the second target cluster number and the second load condition segmentation interval specifically includes: Setting initial parameters of the second clustering algorithm, setting the number of clusters to a first preset number of clusters, and performing cluster analysis on the third load condition data; Sequentially increasing the number of clusters until the number of clusters is greater than a second preset number of clusters; Comparing the variance ratio standards of the clustering number being equal to the first preset cluster number to the second preset cluster number, and determining the clustering number corresponding to the largest variance ratio standard as the second target clustering number; The second load condition segmentation interval is determined according to the clustering result corresponding to the second target clustering number.

8. The vehicle control method according to any one of claims 1 to 7, It is characterized in that After determining the engine speed torque throttle curve strategy associated with the vehicle load condition according to the analysis result of the vehicle load condition, the control method further includes: confirming whether the data volume of the fourth parameter set is greater than a preset threshold, wherein the fourth parameter set is used to indicate the newly collected operating parameters of the vehicle; When the amount of data in the fourth parameter set is greater than a preset threshold, determining a fifth parameter set according to the fourth parameter set and the first parameter set; Perform a cluster analysis on the fifth parameter set, and update or adjust the speed torque throttle curve strategy according to the cluster analysis result of the fifth parameter set.

9. A vehicle control device, It is characterized in that include: An acquisition unit, configured to acquire a first parameter set during vehicle operation, wherein the first parameter set includes operating parameters of the vehicle and time series data corresponding to the operating parameters; a processing unit, configured to preprocess the first parameter set to determine a second parameter set; The processing unit is further used to perform cluster analysis on the second parameter set to determine an analysis result of the vehicle load condition; The processing unit is further used to determine a speed torque throttle curve strategy of the engine associated with the vehicle load condition according to the analysis result of the vehicle load condition; The processing unit is further used for: Analyzing the second parameter set according to a first clustering algorithm to determine a first target cluster number and a first load condition segmentation interval; Analyzing the second parameter set according to a second clustering algorithm to determine a second target number of clusters and a second load condition segmentation interval; Determining an analysis result of the vehicle load condition according to the first target cluster number, the first load condition segmentation interval, the second target cluster number, and the second load condition segmentation interval; The processing unit is further used for: Determining first load condition data according to the second parameter set; Compressing the data other than the load data in the first load condition data into a column of data by using a principal component analysis method, and determining the second load condition data according to the column of data and the load data in the first load condition data; Performing cluster analysis on the first load condition data according to the first clustering algorithm to determine the number of first sub-clustering clusters and a first clustering result; Performing cluster analysis on the second load condition data according to the first clustering algorithm to determine the number of second sub-clustering clusters and a second clustering result; The first target cluster number and the first load condition segmentation interval are determined according to the first sub-clustering cluster number, the first clustering result, the second sub-clustering cluster number and the second clustering result.

10. A readable storage medium, It is characterized in that The readable storage medium stores a program or an instruction, and when the program or the instruction is executed by the processor, the steps of the vehicle control method according to any one of claims 1 to 8 are implemented.

11. A vehicle, It is characterized in that include: The vehicle control device as claimed in claim 9; and / or The readable storage medium as claimed in claim 10.

Images