Electric drive system evaluation method and device and computer equipment

Abstract

The invention relates to an electric drive system evaluation method and device and computer equipment. The method comprises the following steps: obtaining an evaluation dimension index of a to-be-tested electric drive system; user portraits corresponding to different vehicles are constructed according to the evaluation dimension indexes and the vehicle information of the different vehicles provided with the to-be-tested electric drive systems; adopting a minimum sample evaluation method to extract an evaluation sample set of the to-be-tested electric drive system from the user portraits corresponding to different vehicles; based on the system evaluation model, according to the evaluation sample set, evaluating the to-be-tested electric drive system; wherein the system evaluation model is constructed according to a gradient boosting decision tree model. By adopting the method, the electric driving system can be comprehensively and accurately evaluated.

Description

Technical Field

[0001] The present application relates to the technical field of electric vehicle testing, and in particular to an electric drive system evaluation method, device and computer equipment. Background Art

[0002] With the rapid development of electric vehicle technology, the performance evaluation of electric drive systems has become one of the key factors in promoting the progress of the electric vehicle industry. The performance of electric drive systems involves many aspects, including technical indicators such as electrical performance, power performance, safety performance, reliability, and vehicle matching.

[0003] At present, the existing electric drive system evaluation methods based on bench or vehicle test data use a method that relies on expert experience to select evaluation indicators for grading and scoring. The test population coverage is not comprehensive and the iterative verification cost is high. On the other hand, the electric drive system evaluation methods based on vehicle operation big data use preset formulas or rules to calculate and process data to form evaluation indicators. They are easily interfered by extreme or abnormal values, have large granularity, complex calculation process, and the evaluation results cover a single dimension, making it impossible to comprehensively and accurately evaluate the electric drive system. Summary of the invention

[0004] Based on this, it is necessary to provide an electric drive system evaluation method, device and computer equipment that can comprehensively and accurately evaluate the electric drive system in response to the above technical problems.

[0005] In a first aspect, the present application provides an electric drive system evaluation method, comprising:

[0006] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0007] Constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested;

[0008] Using a minimum sample evaluation method, an evaluation sample set of the electric drive system to be tested is extracted from user portraits corresponding to different vehicles;

[0009] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed according to a gradient boosting decision tree model.

[0010] In one of the embodiments, for each vehicle, the vehicle information of the vehicle includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes basic vehicle information and attribute information of the electric drive system to be tested installed on the vehicle.

[0011] In one embodiment, constructing user profiles corresponding to different vehicles based on the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested includes:

[0012] For each vehicle equipped with the electric drive system to be tested, extracting the index value corresponding to the evaluation dimension index from the vehicle information of the vehicle, and constructing the associated index set of the vehicle based on the evaluation dimension index and the index value corresponding to the evaluation dimension index;

[0013] constructing an information matrix of the vehicle according to evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle;

[0014] Inputting the information matrix corresponding to the vehicle into a text topic classification model to obtain an initial indicator set for marking the driving style label for the associated indicator set; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model;

[0015] A user profile corresponding to the vehicle is constructed based on an initial indicator set corresponding to the vehicle.

[0016] In one embodiment, the information matrix of the vehicle is constructed according to the evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle, including:

[0017] Extracting a first dimension index associated with the driving style and a first index value corresponding to the first dimension index from the associated index set corresponding to the vehicle; wherein the data type of the first index value is a numerical type;

[0018] Extracting a second dimension index associated with the driving style and a second index value corresponding to the second dimension index from the associated index set corresponding to the vehicle; wherein the data type of the second index value is text type;

[0019] An information matrix corresponding to the vehicle is constructed according to the first dimensional index, the first index value, the second dimensional index and the second index value.

[0020] In one embodiment, constructing an information matrix corresponding to the vehicle according to the first dimensional index, the first index value, the second dimensional index and the second index value includes:

[0021] Normalizing the first indicator value, and determining indicator evaluation information corresponding to the first dimensional indicator based on the normalized first indicator value;

[0022] Based on the feature extraction model, determining the indicator feature corresponding to the second dimensional indicator according to the second indicator value;

[0023] An information matrix corresponding to the vehicle is constructed based on the indicator evaluation information corresponding to the first dimensional indicator and the indicator characteristics corresponding to the second dimensional indicator.

[0024] In one embodiment, constructing a user profile corresponding to the vehicle according to an initial indicator set corresponding to the vehicle includes:

[0025] Extracting a third dimension indicator associated with the vehicle model configuration and a third indicator value corresponding to the third dimension indicator from the initial indicator set corresponding to the vehicle;

[0026] Determining a target indicator set for marking the initial indicator set with a vehicle model configuration label according to the third dimension indicator and the third indicator value corresponding to the vehicle;

[0027] A user profile corresponding to the vehicle is constructed based on the target indicator set.

[0028] In one embodiment, a minimum sample evaluation method is used to extract an evaluation sample set of the electric drive system to be tested from user portraits corresponding to different vehicles, including:

[0029] Classify different vehicles according to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles to obtain at least two classification groups;

[0030] Using the minimum sample evaluation method, an intermediate sample set is extracted from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set includes at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators;

[0031] An evaluation sample set is determined from the intermediate sample set according to the information value and the value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set.

[0032] In one embodiment, based on the system evaluation model and according to the evaluation sample set, the electric drive system to be tested is evaluated, including:

[0033] Based on a data-driven objective weighting method, the indicator weight of each evaluation dimension indicator in the evaluation sample set is determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set;

[0034] The evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation result of the electric drive system to be tested.

[0035] In a second aspect, the present application also provides an electric drive system evaluation device, comprising:

[0036] An index acquisition module, used to obtain evaluation dimension indicators of the electric drive system to be tested;

[0037] A portrait construction module, used to construct user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested;

[0038] A sample extraction module, used to extract an evaluation sample set of the electric drive system to be tested from user portraits corresponding to different vehicles using a minimum sample evaluation method;

[0039] A system evaluation module is used to evaluate the electric drive system to be tested based on a system evaluation model and the evaluation sample set; wherein the system evaluation model is constructed based on a gradient boosting decision tree model.

[0040] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0041] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0042] Constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested;

[0043] Using a minimum sample evaluation method, an evaluation sample set of the electric drive system to be tested is extracted from user portraits corresponding to different vehicles;

[0044] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed according to a gradient boosting decision tree model.

[0045] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0046] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0047] Constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested;

[0048] Using a minimum sample evaluation method, an evaluation sample set of the electric drive system to be tested is extracted from user portraits corresponding to different vehicles;

[0049] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed according to a gradient boosting decision tree model.

[0050] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0051] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0052] Constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested;

[0053] Using a minimum sample evaluation method, an evaluation sample set of the electric drive system to be tested is extracted from user portraits corresponding to different vehicles;

[0054] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed according to a gradient boosting decision tree model.

[0055] The above-mentioned electric drive system evaluation method, device and computer equipment construct user portraits corresponding to different vehicles based on the evaluation dimension indicators of the electric drive system to be tested and the vehicle information of different vehicles equipped with the electric drive system to be tested, and use the minimum sample evaluation method to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles; further, based on the system evaluation model constructed according to the gradient boosting decision tree model, the electric drive system to be tested is evaluated according to the evaluation sample set. The above scheme comprehensively considers the evaluation dimension indicators of the electric drive system to be tested, and based on the dimensional information of different vehicles equipped with the electric drive system to be tested, it ensures the accuracy and comprehensiveness of the user portraits corresponding to different vehicles constructed; at the same time, the minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles, which reduces the amount of data while ensuring the comprehensiveness of the evaluation sample set; further, the system evaluation model is introduced, and based on the evaluation sample set, the comprehensiveness and accuracy of the evaluation results of the electric drive system to be tested are guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1Ais a schematic flow chart of an electric drive system evaluation method in one embodiment;

[0058] Figure 1B A schematic diagram of a static dimension index table and a dynamic dimension index table in one embodiment;

[0059] Figure 2A A schematic diagram of a process for constructing user portraits corresponding to different vehicles in one embodiment;

[0060] Figure 2B is a schematic diagram of a joint dimension index table in one embodiment;

[0061] Figure 3 A schematic diagram of a process of constructing an information matrix of a vehicle in one embodiment;

[0062] Figure 4 A schematic diagram of a process of constructing an information matrix of a vehicle in another embodiment;

[0063] Figure 5 A schematic diagram of a process of constructing a user portrait corresponding to a vehicle in one embodiment;

[0064] Fig. 6A A schematic diagram of a process for extracting an evaluation sample set for an electric drive system to be tested in one embodiment;

[0065] Figure 6B A schematic diagram of classifying vehicles in one embodiment;

[0066] Figure 6C A schematic diagram of an evaluation sample set constructed in one embodiment;

[0067] Figure 7 A schematic diagram of a flow chart of evaluating an electric drive system to be tested in one embodiment;

[0068] Figure 8 is a schematic flow chart of an electric drive system evaluation method in another embodiment;

[0069] Fig. 9 is a structural block diagram of an electric drive system evaluation device in one embodiment;

[0070] FIG. 10 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0072] The electric drive system evaluation method provided in the embodiment of the present application can be applied to an application environment for evaluating the electric drive system. The electric drive system evaluation method provided in the embodiment of the present application is executed by a computer device, which can be a server or a terminal with powerful computing capabilities.

[0073] In one embodiment, Figure 1A As shown, an electric drive system evaluation method is provided, which is described by taking the application of the method to a server as an example, and specifically includes the following steps:

[0074] S101, obtaining evaluation dimension indicators of the electric drive system to be tested.

[0075] The electric drive system to be tested is an electric drive system that needs to be evaluated. The evaluation dimension indicators are indicators of different dimensions used to evaluate the electric drive system to be tested; in the embodiment of the present application, the evaluation dimension indicators include but are not limited to static dimension indicators and dynamic dimension indicators.

[0076] Optionally, there are many methods for obtaining evaluation dimension indicators of the electric drive system to be tested, which are not limited here. One method may be to select indicators that can reflect the performance of the electric drive system as evaluation dimension indicators based on experience or expert scoring methods. In addition, in order to ensure the comprehensiveness and accuracy of the evaluation dimension indicators, another method may be to extract the relevant indicator data, configuration setting data and evaluation feedback data of the associated vehicle and the electric drive system to be tested from the Internet of Vehicles platform of the vehicle equipped with the electric drive system to be tested, and generate evaluation dimension indicators corresponding to each data based on the data; further, the evaluation dimension indicators are divided into static dimension indicators and dynamic dimension indicators according to the data source, data type and update frequency.

[0077] For example, for dynamic dimension indicators, the vehicle cloud-based tracking data stored in the data lake and data warehouse is grouped by vehicle ID and sorted by timestamp to divide the driving segments, and then the data is preprocessed and indicator items are generated. The data preprocessing process includes field screening, outlier cleaning and data integrity verification, specifically filtering out signals such as vehicle power-related and electric drive assembly-related fields from the vehicle background tracking points, and removing abnormal data outside the valid value range of the corresponding signal, and then verifying the data integrity to ensure that the key indicator signals exist and the vehicle data is uploaded completely.

[0078] It should be noted that for vehicle data that have reported electric drive assembly failures, additional processing is required, including additional functional inspections, and supplementary electric drive failure and functional failure indicator labels to form dynamic dimensional indicators together with other indicator items.

[0079] Exemplarily, for static dimension indicators, numerical data or text information introduced by vehicle production / platform / evaluation systems such as sales / vehicle computer / quality assurance are cleaned, processed and formatted, and corresponding indicator values ​​are filled in according to preset rules to form static dimension indicators.

[0080] Optionally, if the static dimension indicators and dynamic dimension indicators in the evaluation dimension indicators are presented in table form, such as Figure 1B The figure shows a schematic diagram of a static dimension index table and a dynamic dimension index table provided in an embodiment of the present application.

[0081] S102, constructing user portraits corresponding to different vehicles based on the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested.

[0082] Among them, for each vehicle, the vehicle information includes vehicle attribute information and vehicle operation information; the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0083] Optionally, a portrait construction model can be constructed based on historical data, and the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive systems to be tested can be input into the constructed portrait construction model, so that the portrait construction model is based on the model parameters, according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive systems to be tested, and outputs user portraits corresponding to different vehicles.

[0084] S103, using a minimum sample evaluation method, extracting an evaluation sample set of the electric drive system to be tested from user portraits corresponding to different vehicles.

[0085] The evaluation sample set is a sample set used to evaluate the electric drive system to be tested.

[0086] Optionally, different vehicles can be classified according to their corresponding user portraits, and the minimum sample estimation method can be used to determine the number of samples required to be drawn in each category. Furthermore, based on the number of samples required to be drawn in each category, a corresponding number of samples can be drawn from each category as an evaluation sample set for the electric drive system to be tested.

[0087] S104, based on the system evaluation model and the evaluation sample set, the electric drive system to be tested is evaluated.

[0088] Among them, the system evaluation model is used to evaluate the electric drive system to be tested. In the embodiment of the present application, the system evaluation model is constructed based on the gradient boosting decision tree model.

[0089] Optionally, the evaluation sample set may be input into a system evaluation model so that the system evaluation model processes the evaluation sample set according to model parameters and outputs evaluation results of the electric drive system to be tested, thereby completing the evaluation process of the electric drive system to be tested.

[0090] In the above-mentioned electric drive system evaluation method, user portraits corresponding to different vehicles are constructed based on the evaluation dimension indicators of the electric drive system to be tested and the vehicle information of different vehicles equipped with the electric drive system to be tested, and the minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles; further, based on the system evaluation model constructed according to the gradient boosting decision tree model, the electric drive system to be tested is evaluated according to the evaluation sample set. The above scheme comprehensively considers the evaluation dimension indicators of the electric drive system to be tested, and based on the dimensional information of different vehicles equipped with the electric drive system to be tested, it ensures the accuracy and comprehensiveness of the user portraits corresponding to different vehicles constructed; at the same time, the minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles, which reduces the amount of data while ensuring the comprehensiveness of the evaluation sample set; further, the system evaluation model is introduced, and based on the evaluation sample set, the comprehensiveness and accuracy of the evaluation results of the electric drive system to be tested are guaranteed.

[0091] Optionally, based on the above embodiment, in one embodiment, as Figure 2A As shown, a method for constructing user portraits corresponding to different vehicles is provided, which refines the above S102 and specifically includes the following steps:

[0092] S201, for each vehicle equipped with the electric drive system to be tested, extract the indicator value corresponding to the evaluation dimension indicator from the vehicle information of the vehicle, and construct the vehicle's associated indicator set based on the evaluation dimension indicator and the indicator value corresponding to the evaluation dimension indicator.

[0093] The indicator value corresponding to the evaluation dimension indicator is the value corresponding to the evaluation dimension indicator. The vehicle-related indicator set is a set of evaluation dimension indicators related to the vehicle. In the embodiment of the present application, the related indicator level can be presented in the form of a table.

[0094] Optionally, for each vehicle equipped with the electric drive system to be tested, data corresponding to the evaluation dimension index is extracted from the vehicle information of the vehicle as the index value; further, the evaluation dimension index and the index value are matched to construct an associated index set for the vehicle.

[0095] For example, if the evaluation dimension index includes a dynamic dimension index and a static dimension index, the dynamic dimension index and the static dimension index can be jointly represented, that is, a joint dimension index is constructed, and a joint dimension index table is constructed. The index value corresponding to the evaluation dimension index is extracted from the vehicle information of the vehicle, filled into the joint dimension index table, and the associated indicator set of the vehicle is constructed. Figure 2B As shown, the joint dimensional indicator representation provided in the embodiment of the present application is intended to fill in the corresponding indicator value under the corresponding evaluation dimensional indicator to construct a vehicle-related indicator set.

[0096] S202, constructing a vehicle information matrix according to the evaluation dimension indicators associated with the driving style and the corresponding indicator values ​​in the associated indicator set corresponding to the vehicle.

[0097] The driving styles include but are not limited to calm, normal and aggressive; the vehicle information matrix is ​​a matrix composed of information representing the vehicle driving style.

[0098] Optionally, driving styles can be divided into three categories: calm, normal and aggressive. Evaluation dimension indicators and corresponding indicator values ​​associated with the driving styles are extracted from the associated indicator set corresponding to the vehicle. Based on preset rules, the vehicle information matrix is ​​constructed according to the extracted evaluation dimension indicators and corresponding indicator values.

[0099] S203 inputs the information matrix corresponding to the vehicle into the text topic classification model to obtain an initial indicator set for marking the driving style label on the associated indicator set.

[0100] The text topic classification model is used to classify vehicles; in the embodiment of the present application, the text topic classification model is a Latent Dirichlet Allocation (LDA) model.

[0101] Optionally, the information matrix corresponding to the vehicle is input into a text topic classification model, so that the text topic classification model processes the information matrix according to preset rules, and outputs a driving style label corresponding to the vehicle. Furthermore, the associated indicator set is marked with a driving style label to obtain an initial indicator set.

[0102] S204: construct a user profile corresponding to the vehicle based on the initial indicator set corresponding to the vehicle.

[0103] Optionally, a user profile building tool may be used to build a user profile corresponding to the vehicle based on the driving style label in the initial indicator set corresponding to the vehicle, as well as other evaluation dimension indicators and corresponding indicator values ​​in the initial indicator set.

[0104] In this embodiment, by constructing a vehicle-related indicator set and introducing an information matrix and a text topic classification model, the accuracy of the driving style label of the determined associated indicator set is ensured, and further based on the initial indicator set, the accuracy of the user portrait corresponding to the constructed vehicle is ensured.

[0105] Optionally, in one embodiment, Figure 3 As shown, a method for constructing a vehicle information matrix is ​​provided, which refines the above S202 and specifically includes the following steps:

[0106] S301, extracting a first dimension indicator associated with a driving style and a first indicator value corresponding to the first dimension indicator from a set of associated indicators corresponding to the vehicle.

[0107] Among them, the data type of the first indicator value is a numeric type. For example, the first dimension indicator whose data type is a numeric type includes but is not limited to vehicle speed, acceleration, pedal travel, torque change rate and speed change rate.

[0108] Optionally, a first dimension indicator of a non-numeric data type associated with the driving style and a first indicator value corresponding to the first dimension indicator may be extracted from a set of associated indicators corresponding to the vehicle.

[0109] S302: Extract a second dimension indicator associated with the driving style and a second indicator value corresponding to the second dimension indicator from a set of associated indicators corresponding to the vehicle.

[0110] The data type of the second indicator value is text.

[0111] Optionally, a second dimension indicator of a text-type data type associated with the driving style and a second indicator value corresponding to the second dimension indicator are extracted from a set of associated indicators corresponding to the vehicle.

[0112] S303, constructing an information matrix corresponding to the vehicle according to the first dimension index, the first index value, the second dimension index and the second index value.

[0113] Optionally, based on a preset information matrix generation rule, an information matrix corresponding to the vehicle is constructed according to the first dimension index, the first index value, the second dimension index and the second index value.

[0114] In this embodiment, by introducing the first dimension index and the first index value, as well as the second dimension index and the second index value, the evaluation dimension indexes of all data types are fully considered, and the accuracy of the information matrix corresponding to the constructed vehicle is further guaranteed.

[0115] Optionally, in one embodiment, Figure 4 As shown, a method for constructing a vehicle information matrix is ​​provided, which refines the above S303 and specifically includes the following steps:

[0116] S401, based on the feature extraction model, determine the indicator feature corresponding to the second dimension indicator according to the first indicator value.

[0117] Among them, the feature extraction model is a model that does not perform feature extraction on the first indicator value.

[0118] Optionally, the first indicator value is input into a feature extraction model so that the feature extraction model extracts keywords from the first indicator value, constructs features, and outputs indicator features corresponding to the second dimension indicator.

[0119] S402, normalize the first indicator value, and determine indicator evaluation information corresponding to the first dimension indicator based on the normalized second indicator value.

[0120] Among them, the indicator evaluation information is information that reflects evaluation words; in the embodiment of the present application, the indicator evaluation information includes but is not limited to high, medium, relatively low, low, etc.

[0121] Optionally, the first indicator value is normalized, and the indicator evaluation information corresponding to the first dimension indicator is determined according to and based on the size of the second indicator value after the normalization.

[0122] For example, if the first dimension indicator is a descriptive / representational indicator, such as vehicle speed, acceleration, pedal travel, torque change rate, and speed change rate, and the first indicator value is , given the limit nominal ,in, is the lower threshold, is the upper threshold, then there is a conversion process:

[0123] (1)

[0124] in, It is the set evaluation category words and indicator evaluation information.

[0125] If the first dimension indicator is an intermediate indicator, such as voltage, current, temperature, etc., the first indicator value is first normalized, and the indicator evaluation information corresponding to the first dimension indicator is determined based on the normalized first indicator value and the limit nominal.

[0126] If the first dimension indicator is an interval indicator, such as speed-torque interval distribution, vehicle speed-mileage interval distribution and electronic and electrical boundary, the first indicator value is also normalized first, and the indicator evaluation information corresponding to the first dimension indicator is determined based on the normalized first indicator value and boundary nominal.

[0127] S403, constructing an information matrix corresponding to the vehicle according to the indicator evaluation information corresponding to the first dimension indicator and the indicator characteristics corresponding to the second dimension indicator.

[0128] Optionally, the indicator evaluation information corresponding to the first dimension indicator and the indicator features corresponding to the second dimension indicator are constructed into an information matrix corresponding to the vehicle. ,have: ,in, is the indicator evaluation information corresponding to the first dimension indicator under the first vehicle, It is the indicator feature corresponding to the second dimension indicator under the first vehicle.

[0129] In this embodiment, the information matrix corresponding to the vehicle is constructed by introducing the index characteristics and the index evaluation information, thereby ensuring the accuracy of the constructed information matrix.

[0130] Optionally, in one embodiment, Figure 5 As shown, a method for constructing a user portrait corresponding to a vehicle is provided, which refines the above S204 and specifically includes the following steps:

[0131] S501, extracting a third dimension indicator associated with the vehicle model configuration and a third indicator value corresponding to the third dimension indicator from an initial indicator set corresponding to the vehicle.

[0132] Among them, vehicle configurations include but are not limited to basic configurations, advanced configurations and high-end configurations.

[0133] Optionally, the vehicle configuration may be divided into basic configuration, advanced configuration and high-level configuration, and the third dimension index associated with the vehicle configuration and the third index value corresponding to the third dimension index may be extracted from the initial index set corresponding to the vehicle.

[0134] S502, determining a target indicator set for marking the initial indicator set with a vehicle model configuration label according to the third dimension indicator and the third indicator value corresponding to the vehicle.

[0135] Optionally, DBSCAN (Density-Based Spatial Clustering of Applications with Noise) can be used to cluster the third dimension index and the third index value corresponding to the vehicle, and output the vehicle configuration label corresponding to the vehicle; further, the initial index set is labeled according to the vehicle configuration label to obtain the target index set.

[0136] S503: Build a user profile corresponding to the vehicle based on the target indicator set.

[0137] In this embodiment, by introducing the third dimension indicator and the third indicator value, as well as the target indicator set, the accuracy of the determined vehicle model configuration label is ensured, and further, the accuracy of the user portrait corresponding to the constructed vehicle is ensured.

[0138] Optionally, in one embodiment, Fig. 6A As shown, a method for extracting an evaluation sample set of an electric drive system to be tested is provided, which refines the above S103 and specifically includes the following steps:

[0139] S601, classify different vehicles according to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles to obtain at least two classification groups.

[0140] Among them, the classification groups are categories obtained according to vehicle configuration labels and driving style labels.

[0141] Optionally, if the vehicle configuration labels are divided into basic configuration, advanced configuration and high-end configuration, and the driving style labels are divided into three categories: calm, ordinary and aggressive, then different vehicles can be divided into basic configuration calm type, basic configuration ordinary type, basic configuration aggressive type, advanced configuration calm type, advanced configuration ordinary type, advanced configuration aggressive type, high-end configuration ordinary type, high-end configuration aggressive type and high-end configuration calm type according to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles.

[0142] For example, Figure 6B As shown, it is a schematic diagram for classifying different vehicles.

[0143] S602, using the minimum sample evaluation method, extracting an intermediate sample set from the user portraits corresponding to each classification group.

[0144] Among them, each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators.

[0145] Optionally, a minimum sample evaluation method is used to determine the minimum number of samples for each classification group. Further, a corresponding number of user portraits of the minimum number of samples are extracted from each classification group as an intermediate sample set.

[0146] For example, if this is the first time to form an evaluation sample set, the number of samples randomly selected from each classification group is the minimum value of the samples in all classification groups. , that is, the total number of sample sets satisfies ,in, and Represents the total number of classification groups and the number of clusters within the classification groups. If this is not the first time to form an evaluation sample set, first take the mean and variance of each classification group, calculate the minimum number of samples required for the evaluation set, take the minimum value of the two calculated sample sizes as the total number of samples, and then extract group samples based on the total number ratio.

[0147] S603, determining an evaluation sample set from the intermediate sample set according to the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set.

[0148] Among them, information value reflects the value contained in the information of the evaluation dimension indicators. The higher the information value, the greater the contribution; the value threshold is the pre-set threshold of information value.

[0149] Optionally, for any evaluation dimension indicator, the delicate value of the evaluation dimension indicator can be calculated by the following formula:

[0150]

[0151] in, For information value, is the proportion of positive corresponding samples in the classification group; is the proportion of non-responding samples in this classification group.

[0152] Optionally, after calculating the information value of each evaluation dimension indicator, the absolute value is taken, and the evaluation dimension indicators whose absolute value of information value is greater than the value threshold and the corresponding indicator values ​​are used as the evaluation sample set.

[0153] For example, Figure 6C The following is a schematic diagram of the constructed evaluation sample set.

[0154] In this embodiment, the minimum sample evaluation method is used to extract the intermediate sample set from each classification group, thereby reducing the amount of data while ensuring the forwardness of the intermediate sample set; further, based on the information value and value threshold of the indicator values ​​corresponding to the evaluation dimension indicators in the intermediate sample set, the evaluation sample set is determined from the intermediate sample set, further reducing the amount of data.

[0155] Optionally, in one embodiment, Figure 7 As shown, a method for evaluating an electric drive system to be tested is provided, which specifically includes the following steps:

[0156] S701, based on the data-driven objective weighting method, determine the indicator weight of each evaluation dimension indicator in the evaluation sample set according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set.

[0157] Optionally, according to the index values ​​corresponding to different vehicles under each evaluation dimension index in the evaluation sample set, a weight calculation matrix is ​​constructed for each evaluation dimension index in the evaluation sample set, and then the weight calculation matrix is ​​applied with a data-driven objective weighting method to calculate the weight factor coefficient value of the index.

[0158] For example, for a matrix consisting of n data samples and m indicator factors ,in , for the evaluation dimension indicators Normalize according to positive indicators, negative indicators, intermediate indicators and interval indicators and then calculate their volatility and conflict The data-driven objective weighting method uses the differences in the indicator values ​​within each evaluation dimension to evaluate and decide how to allocate weights.

[0159] S702: Input the evaluation sample set and the indicator weights of each evaluation dimension indicator into the system evaluation model to obtain the evaluation result of the electric drive system to be tested.

[0160] Optionally, the evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model, so that the system evaluation model analyzes the evaluation sample set and the indicator weights of each evaluation dimension indicator according to the model parameters, and outputs the evaluation results of the electric drive system to be tested.

[0161] In this embodiment, each evaluation dimension indicator in the evaluation sample set is weighted according to a data-driven objective weighting method, thereby highlighting the importance of each evaluation dimension indicator and facilitating subsequent screening and dimensionality reduction of the evaluation dimension indicators.

[0162] Figure 8 FIG. 2 is a flow chart of an electric drive system evaluation method in another embodiment. Based on the above embodiment, this embodiment provides an optional example of an electric drive system evaluation method. Figure 8 The specific implementation process is as follows:

[0163] S801, obtaining evaluation dimension indicators of the electric drive system to be tested.

[0164] S802, for each vehicle equipped with the electric drive system to be tested, extract the indicator value corresponding to the evaluation dimension indicator from the vehicle information of each vehicle, and construct the associated indicator set of each vehicle based on the evaluation dimension indicator and the indicator value corresponding to the evaluation dimension indicator.

[0165] Among them, for each vehicle, the vehicle information of the vehicle includes vehicle attribute information and vehicle operation information; among them, the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0166] S803: extracting a first dimension indicator associated with the driving style and a first indicator value corresponding to the first dimension indicator from the associated indicator set corresponding to each vehicle.

[0167] The data type of the first indicator value is numeric.

[0168] S804: Extract the second dimension indicator associated with the driving style and the second indicator value corresponding to the second dimension indicator from the associated indicator set corresponding to each vehicle.

[0169] Wherein, the data type of the second indicator value is text type;

[0170] S805, constructing an information matrix corresponding to each vehicle according to the first dimension index, the first index value, the second dimension index and the second index value of each vehicle.

[0171] Optionally, based on the feature extraction model, determine the indicator characteristics corresponding to the second dimension indicator according to the first indicator value; normalize the second indicator value, and determine the indicator evaluation information corresponding to the first dimension indicator based on the normalized first indicator value; construct the information matrix corresponding to the vehicle according to the indicator evaluation information corresponding to the first dimension indicator and the indicator characteristics corresponding to the second dimension indicator.

[0172] S806, inputting the information matrix corresponding to each vehicle into the text topic classification model to obtain an initial indicator set for marking the driving style label for the associated indicator set.

[0173] Among them, the text topic classification model is the Latent Dirichlet Allocation (LDA) model.

[0174] S807, extracting the third dimension index associated with the vehicle model configuration and the third index value corresponding to the third dimension index from the initial index set corresponding to each vehicle.

[0175] S808, determining a target indicator set for marking the initial indicator set with a vehicle model configuration label according to the third dimension indicator and the third indicator value corresponding to each vehicle.

[0176] S809: Build a user profile corresponding to each vehicle based on the target indicator set.

[0177] S810, classify different vehicles according to the vehicle configuration labels and driving style labels in the user portraits corresponding to the different vehicles to obtain at least two classification groups.

[0178] S811, using a minimum sample evaluation method to extract an evaluation sample set of the electric drive system to be tested from user portraits corresponding to different vehicles.

[0179] Optionally, the minimum sample evaluation method is used to extract an intermediate sample set from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; and the evaluation sample set is determined from the intermediate sample set based on the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set.

[0180] S812, based on the data-driven objective weighting method, determine the indicator weight of each evaluation dimension indicator in the evaluation sample set according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set.

[0181] S813, input the evaluation sample set and the indicator weight of each evaluation dimension indicator into the system evaluation model to obtain the evaluation result of the electric drive system to be tested.

[0182] The specific process of the above S801-S813 can be found in the description of the above method embodiment. The implementation principle and technical effect are similar and will not be repeated here.

[0183] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0184] Based on the same inventive concept, the embodiment of the present application also provides an electric drive system evaluation device for implementing the electric drive system evaluation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more electric drive system evaluation device embodiments provided below can refer to the limitations of the electric drive system evaluation method above, and will not be repeated here.

[0185] In one embodiment, Fig. 9 As shown, an electric drive system evaluation device 90 is provided, comprising: an indicator acquisition module 10, a portrait construction module 20, a sample extraction module 30 and a system evaluation module 40, wherein:

[0186] The index acquisition module 10 is used to obtain evaluation dimension indicators of the electric drive system to be tested.

[0187] The portrait construction module 20 is used to construct user portraits corresponding to different vehicles based on the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested.

[0188] The sample extraction module 30 is used to extract an evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles by using a minimum sample evaluation method.

[0189] The system evaluation module 40 is used to evaluate the electric drive system to be tested based on the system evaluation model and the evaluation sample set; wherein the system evaluation model is constructed based on the gradient boosting decision tree model.

[0190] The above-mentioned electric drive system evaluation device constructs user portraits corresponding to different vehicles based on the evaluation dimension indicators of the electric drive system to be tested and the vehicle information of different vehicles equipped with the electric drive system to be tested, and adopts the minimum sample evaluation method to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles; further, based on the system evaluation model constructed according to the gradient boosting decision tree model, the electric drive system to be tested is evaluated according to the evaluation sample set. The above scheme comprehensively considers the evaluation dimension indicators of the electric drive system to be tested, and based on the dimensional information of different vehicles equipped with the electric drive system to be tested, it ensures the accuracy and comprehensiveness of the user portraits corresponding to different vehicles constructed; at the same time, the minimum sample evaluation method is adopted to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles, which reduces the amount of data while ensuring the comprehensiveness of the evaluation sample set; further, the system evaluation model is introduced, and based on the evaluation sample set, the comprehensiveness and accuracy of the evaluation results of the electric drive system to be tested are guaranteed.

[0191] In one embodiment, for each vehicle, the vehicle information includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0192] In one embodiment, the portrait building module 20 includes:

[0193] The indicator construction unit is used to extract the indicator value corresponding to the evaluation dimension indicator from the vehicle information of each vehicle equipped with the electric drive system to be tested, and to construct the vehicle's associated indicator set based on the evaluation dimension indicator and the indicator value corresponding to the evaluation dimension indicator.

[0194] The matrix construction unit is used to construct the vehicle information matrix according to the evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle.

[0195] The label marking unit is used to input the information matrix corresponding to the vehicle into the text topic classification model to obtain an initial indicator set for marking the driving style labels of the associated indicator set; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model.

[0196] The portrait construction unit is used to construct a user portrait corresponding to the vehicle according to an initial indicator set corresponding to the vehicle.

[0197] In one embodiment, the matrix construction unit includes:

[0198] The first extraction subunit is used to extract a first dimension indicator associated with the driving style and a first indicator value corresponding to the first dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the first indicator value is a numerical type.

[0199] The second extraction subunit is used to extract the second dimension indicator associated with the driving style and the second indicator value corresponding to the second dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the second indicator value is text type.

[0200] The matrix construction subunit is used to construct an information matrix corresponding to the vehicle according to the first dimension index, the first index value, the second dimension index and the second index value.

[0201] In one embodiment, the matrix construction subunit is specifically used for:

[0202] Based on the feature extraction model, the indicator feature corresponding to the second dimension indicator is determined according to the first indicator value; the second indicator value is normalized, and based on the normalized first indicator value, the indicator evaluation information corresponding to the first dimension indicator is determined; according to the indicator evaluation information corresponding to the first dimension indicator and the indicator feature corresponding to the second dimension indicator, an information matrix corresponding to the vehicle is constructed.

[0203] In one embodiment, the portrait construction unit is specifically used to:

[0204] From the initial indicator set corresponding to the vehicle, extract the third dimension indicators associated with the vehicle model configuration and the third indicator values ​​corresponding to the third dimension indicators; determine the target indicator set for marking the vehicle model configuration label for the initial indicator set based on the third dimension indicators and the third indicator values ​​corresponding to the vehicle; and construct a user portrait corresponding to the vehicle based on the target indicator set.

[0205] In one embodiment, the sample extraction module 30 is specifically used for:

[0206] According to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles, different vehicles are classified to obtain at least two classification groups; the minimum sample evaluation method is used to extract an intermediate sample set from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; according to the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set, the evaluation sample set is determined from the intermediate sample set.

[0207] In one embodiment, the system evaluation module 40 is specifically used for:

[0208] Based on the data-driven objective weighting method, the indicator weights of each evaluation dimension indicator in the evaluation sample set are determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set; the evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation results of the electric drive system to be tested.

[0209] Each module in the above-mentioned electric drive system evaluation device can be implemented in whole or in part by software, hardware or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0210] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG10. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as evaluation dimension indicators of the electric drive system to be tested. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an electric drive system evaluation method is implemented.

[0211] Those skilled in the art will understand that the structure shown in FIG. 10 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0212] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0213] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0214] According to the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the user profiles corresponding to different vehicles are constructed;

[0215] The minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles;

[0216] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed based on the gradient boosting decision tree model.

[0217] In one embodiment, for each vehicle, the vehicle information includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0218] In one embodiment, when the processor executes the computer program to construct user profiles corresponding to different vehicles based on the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the following steps are also implemented:

[0219] For each vehicle equipped with the electric drive system to be tested, the index value corresponding to the evaluation dimension index is extracted from the vehicle information of the vehicle, and the associated index set of the vehicle is constructed based on the evaluation dimension index and the index value corresponding to the evaluation dimension index; the information matrix of the vehicle is constructed according to the evaluation dimension index associated with the driving style and the corresponding index value in the associated index set corresponding to the vehicle; the information matrix corresponding to the vehicle is input into the text topic classification model to obtain the initial index set for marking the associated index set with the driving style label; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model; according to the initial indicator set corresponding to the vehicle, a user profile corresponding to the vehicle is constructed.

[0220] In one embodiment, when the processor executes the computer program to construct the information matrix of the vehicle according to the evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle, the following steps are also implemented:

[0221] Extract the first dimension indicator associated with the driving style and the first indicator value corresponding to the first dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the first indicator value is a numeric type; extract the second dimension indicator associated with the driving style and the second indicator value corresponding to the second dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the second indicator value is a text type; construct an information matrix corresponding to the vehicle according to the first dimension indicator, the first indicator value, the second dimension indicator and the second indicator value.

[0222] In one embodiment, when the processor executes the computer program to construct the information matrix corresponding to the vehicle according to the first dimension index, the first index value, the second dimension index and the second index value, the following steps are also implemented:

[0223] Based on the feature extraction model, the indicator feature corresponding to the second dimension indicator is determined according to the first indicator value; the second indicator value is normalized, and based on the normalized first indicator value, the indicator evaluation information corresponding to the first dimension indicator is determined; according to the indicator evaluation information corresponding to the first dimension indicator and the indicator feature corresponding to the second dimension indicator, an information matrix corresponding to the vehicle is constructed.

[0224] In one embodiment, when the processor executes the computer program to construct a user profile corresponding to the vehicle based on the initial indicator set corresponding to the vehicle, the processor further implements the following steps:

[0225] From the initial indicator set corresponding to the vehicle, extract the third dimension indicators associated with the vehicle model configuration and the third indicator values ​​corresponding to the third dimension indicators; determine the target indicator set for marking the vehicle model configuration label for the initial indicator set based on the third dimension indicators and the third indicator values ​​corresponding to the vehicle; and construct a user portrait corresponding to the vehicle based on the target indicator set.

[0226] In one embodiment, when the processor executes the computer program to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles using the minimum sample evaluation method, the processor also implements the following steps:

[0227] According to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles, different vehicles are classified to obtain at least two classification groups; the minimum sample evaluation method is used to extract an intermediate sample set from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; according to the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set, the evaluation sample set is determined from the intermediate sample set.

[0228] In one embodiment, the processor executes the computer program to evaluate the electric drive system to be tested based on the system evaluation model and the evaluation sample set, and further implements the following steps:

[0229] Based on the data-driven objective weighting method, the indicator weights of each evaluation dimension indicator in the evaluation sample set are determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set; the evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation results of the electric drive system to be tested.

[0230] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0231] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0232] According to the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the user profiles corresponding to different vehicles are constructed;

[0233] The minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles;

[0234] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed based on the gradient boosting decision tree model.

[0235] In one embodiment, for each vehicle, the vehicle information includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0236] In one embodiment, when the processor executes the computer program to construct user profiles corresponding to different vehicles based on the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the following steps are also implemented:

[0237] For each vehicle equipped with the electric drive system to be tested, the index value corresponding to the evaluation dimension index is extracted from the vehicle information of the vehicle, and the associated index set of the vehicle is constructed based on the evaluation dimension index and the index value corresponding to the evaluation dimension index; the information matrix of the vehicle is constructed according to the evaluation dimension index associated with the driving style and the corresponding index value in the associated index set corresponding to the vehicle; the information matrix corresponding to the vehicle is input into the text topic classification model to obtain the initial index set for marking the associated index set with the driving style label; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model; according to the initial indicator set corresponding to the vehicle, a user profile corresponding to the vehicle is constructed.

[0238] In one embodiment, when the processor executes the computer program to construct the information matrix of the vehicle according to the evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle, the following steps are also implemented:

[0239] Extract the first dimension indicator associated with the driving style and the first indicator value corresponding to the first dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the first indicator value is a numeric type; extract the second dimension indicator associated with the driving style and the second indicator value corresponding to the second dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the second indicator value is a text type; construct an information matrix corresponding to the vehicle according to the first dimension indicator, the first indicator value, the second dimension indicator and the second indicator value.

[0240] In one embodiment, when the processor executes the computer program to construct the information matrix corresponding to the vehicle according to the first dimension index, the first index value, the second dimension index and the second index value, the following steps are also implemented:

[0241] Based on the feature extraction model, the indicator feature corresponding to the second dimension indicator is determined according to the first indicator value; the second indicator value is normalized, and based on the normalized first indicator value, the indicator evaluation information corresponding to the first dimension indicator is determined; according to the indicator evaluation information corresponding to the first dimension indicator and the indicator feature corresponding to the second dimension indicator, an information matrix corresponding to the vehicle is constructed.

[0242] In one embodiment, when the processor executes the computer program to construct a user profile corresponding to the vehicle based on the initial indicator set corresponding to the vehicle, the processor further implements the following steps:

[0243] From the initial indicator set corresponding to the vehicle, extract the third dimension indicators associated with the vehicle model configuration and the third indicator values ​​corresponding to the third dimension indicators; determine the target indicator set for marking the vehicle model configuration label for the initial indicator set based on the third dimension indicators and the third indicator values ​​corresponding to the vehicle; and construct a user portrait corresponding to the vehicle based on the target indicator set.

[0244] In one embodiment, when the processor executes the computer program to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles using the minimum sample evaluation method, the processor also implements the following steps:

[0245] According to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles, different vehicles are classified to obtain at least two classification groups; the minimum sample evaluation method is used to extract an intermediate sample set from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; according to the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set, the evaluation sample set is determined from the intermediate sample set.

[0246] In one embodiment, the processor executes the computer program to evaluate the electric drive system to be tested based on the system evaluation model and the evaluation sample set, and further implements the following steps:

[0247] Based on the data-driven objective weighting method, the indicator weights of each evaluation dimension indicator in the evaluation sample set are determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set; the evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation results of the electric drive system to be tested.

[0248] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0249] Obtain evaluation dimension indicators of the electric drive system to be tested;

[0250] According to the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the user profiles corresponding to different vehicles are constructed;

[0251] The minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles;

[0252] Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed based on the gradient boosting decision tree model.

[0253] In one embodiment, for each vehicle, the vehicle information includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes the operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes the basic vehicle information and the attribute information of the electric drive system to be tested installed on the vehicle.

[0254] In one embodiment, when the processor executes the computer program to construct user profiles corresponding to different vehicles based on the evaluation dimension indicators and the vehicle information of different vehicles equipped with the electric drive system to be tested, the following steps are also implemented:

[0255] For each vehicle equipped with the electric drive system to be tested, the index value corresponding to the evaluation dimension index is extracted from the vehicle information of the vehicle, and the associated index set of the vehicle is constructed based on the evaluation dimension index and the index value corresponding to the evaluation dimension index; the information matrix of the vehicle is constructed according to the evaluation dimension index associated with the driving style and the corresponding index value in the associated index set corresponding to the vehicle; the information matrix corresponding to the vehicle is input into the text topic classification model to obtain the initial index set for marking the associated index set with the driving style label; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model; according to the initial indicator set corresponding to the vehicle, a user profile corresponding to the vehicle is constructed.

[0256] In one embodiment, when the processor executes the computer program to construct the information matrix of the vehicle according to the evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle, the following steps are also implemented:

[0257] Extract the first dimension indicator associated with the driving style and the first indicator value corresponding to the first dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the first indicator value is a numeric type; extract the second dimension indicator associated with the driving style and the second indicator value corresponding to the second dimension indicator from the associated indicator set corresponding to the vehicle; wherein the data type of the second indicator value is a text type; construct an information matrix corresponding to the vehicle according to the first dimension indicator, the first indicator value, the second dimension indicator and the second indicator value.

[0258] In one embodiment, when the processor executes the computer program to construct the information matrix corresponding to the vehicle according to the first dimension index, the first index value, the second dimension index and the second index value, the following steps are also implemented:

[0259] Based on the feature extraction model, the indicator feature corresponding to the second dimension indicator is determined according to the first indicator value; the second indicator value is normalized, and based on the normalized first indicator value, the indicator evaluation information corresponding to the first dimension indicator is determined; according to the indicator evaluation information corresponding to the first dimension indicator and the indicator feature corresponding to the second dimension indicator, an information matrix corresponding to the vehicle is constructed.

[0260] In one embodiment, when the processor executes the computer program to construct a user profile corresponding to the vehicle based on the initial indicator set corresponding to the vehicle, the processor further implements the following steps:

[0261] From the initial indicator set corresponding to the vehicle, extract the third dimension indicators associated with the vehicle model configuration and the third indicator values ​​corresponding to the third dimension indicators; determine the target indicator set for marking the vehicle model configuration label for the initial indicator set based on the third dimension indicators and the third indicator values ​​corresponding to the vehicle; and construct a user portrait corresponding to the vehicle based on the target indicator set.

[0262] In one embodiment, when the processor executes the computer program to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles using the minimum sample evaluation method, the processor also implements the following steps:

[0263] According to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles, different vehicles are classified to obtain at least two classification groups; the minimum sample evaluation method is used to extract an intermediate sample set from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set contains at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; according to the information value and value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set, the evaluation sample set is determined from the intermediate sample set.

[0264] In one embodiment, the processor executes the computer program to evaluate the electric drive system to be tested based on the system evaluation model and the evaluation sample set, and further implements the following steps:

[0265] Based on the data-driven objective weighting method, the indicator weights of each evaluation dimension indicator in the evaluation sample set are determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set; the evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation results of the electric drive system to be tested.

[0266] It should be noted that the data involved in this application (including but not limited to data used for analysis, data stored, data displayed, etc.) are all information and data fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0267] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0268] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0269] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An electric drive system evaluation method, characterized in that: The method comprises: Obtain evaluation dimension indicators of the electric drive system to be tested; Constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested; Using a minimum sample evaluation method, an evaluation sample set of the electric drive system to be tested is extracted from user portraits corresponding to different vehicles; Based on the system evaluation model, the electric drive system to be tested is evaluated according to the evaluation sample set; wherein the system evaluation model is constructed according to a gradient boosting decision tree model.

2. The method according to claim 1, characterized in that For each vehicle, the vehicle information of the vehicle includes vehicle attribute information and vehicle operation information; wherein the vehicle operation information includes operation information of the electric drive system to be tested installed on the vehicle, and the vehicle attribute information includes basic vehicle information and attribute information of the electric drive system to be tested installed on the vehicle.

3. The method according to claim 1, characterized in that The step of constructing user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested includes: For each vehicle equipped with the electric drive system to be tested, extracting the index value corresponding to the evaluation dimension index from the vehicle information of the vehicle, and constructing the associated index set of the vehicle based on the evaluation dimension index and the index value corresponding to the evaluation dimension index; constructing an information matrix of the vehicle according to evaluation dimension indicators and corresponding indicator values ​​associated with the driving style in the associated indicator set corresponding to the vehicle; Inputting the information matrix corresponding to the vehicle into a text topic classification model to obtain an initial indicator set for marking the driving style label for the associated indicator set; wherein the text topic classification model is a latent Dirichlet allocation (LDA) model; A user profile corresponding to the vehicle is constructed based on an initial indicator set corresponding to the vehicle.

4. The method according to claim 3, characterized in that The step of constructing the information matrix of the vehicle according to the evaluation dimension indexes and corresponding index values ​​associated with the driving style in the associated index set corresponding to the vehicle includes: Extracting a first dimension index associated with the driving style and a first index value corresponding to the first dimension index from the associated index set corresponding to the vehicle; wherein the data type of the first index value is a numerical type; Extracting a second dimension index associated with the driving style and a second index value corresponding to the second dimension index from the associated index set corresponding to the vehicle; wherein the data type of the second index value is text type; An information matrix corresponding to the vehicle is constructed according to the first dimensional index, the first index value, the second dimensional index and the second index value.

5. The method according to claim 4, characterized in that The constructing the information matrix corresponding to the vehicle according to the first dimensional index, the first index value, the second dimensional index and the second index value includes: Based on the feature extraction model, determining the indicator feature corresponding to the second dimensional indicator according to the first indicator value; Normalizing the second indicator value, and determining indicator evaluation information corresponding to the first dimensional indicator based on the normalized first indicator value; An information matrix corresponding to the vehicle is constructed based on the indicator evaluation information corresponding to the first dimensional indicator and the indicator characteristics corresponding to the second dimensional indicator.

6. The method according to claim 3, characterized in that The step of constructing a user profile corresponding to the vehicle according to the initial indicator set corresponding to the vehicle includes: Extracting a third dimension indicator associated with the vehicle model configuration and a third indicator value corresponding to the third dimension indicator from the initial indicator set corresponding to the vehicle; Determining a target indicator set for marking the initial indicator set with a vehicle model configuration label according to the third dimension indicator and the third indicator value corresponding to the vehicle; A user profile corresponding to the vehicle is constructed based on the target indicator set.

7. The method according to claim 6, characterized in that The minimum sample evaluation method is used to extract the evaluation sample set of the electric drive system to be tested from the user portraits corresponding to different vehicles, including: Classify different vehicles according to the vehicle configuration labels and driving style labels in the user portraits corresponding to different vehicles to obtain at least two classification groups; Using the minimum sample evaluation method, an intermediate sample set is extracted from the user portraits corresponding to each classification group; wherein each sample in the intermediate sample set includes at least two evaluation dimension indicators and indicator values ​​corresponding to the evaluation dimension indicators; An evaluation sample set is determined from the intermediate sample set according to the information value and the value threshold of the indicator value corresponding to each evaluation dimension indicator in the intermediate sample set.

8. The method according to claim 1, characterized in that: The step of evaluating the electric drive system to be tested based on the system evaluation model and the evaluation sample set includes: Based on a data-driven objective weighting method, the indicator weight of each evaluation dimension indicator in the evaluation sample set is determined according to the indicator values ​​corresponding to different vehicles under each evaluation dimension indicator in the evaluation sample set; The evaluation sample set and the indicator weights of each evaluation dimension indicator are input into the system evaluation model to obtain the evaluation result of the electric drive system to be tested.

9. An electric drive system evaluation device, characterized in that: The device comprises: An index acquisition module, used to obtain evaluation dimension indicators of the electric drive system to be tested; A portrait construction module, used to construct user portraits corresponding to different vehicles according to the evaluation dimension indicators and vehicle information of different vehicles equipped with the electric drive system to be tested; A sample extraction module, used to extract an evaluation sample set of the electric drive system to be tested from user portraits corresponding to different vehicles using a minimum sample evaluation method; A system evaluation module is used to evaluate the electric drive system to be tested based on a system evaluation model and the evaluation sample set; wherein the system evaluation model is constructed based on a gradient boosting decision tree model.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images