A comprehensive optimization method and system for full-link vehicle operation efficiency

Abstract

The present application provides a comprehensive optimization method and system for full-link vehicle operation efficiency. Among them, multiple operation parameters are obtained during the operation of the vehicle; the multiple operation parameters are classified according to the degree of influence on the vehicle operation efficiency to obtain multiple types of operation parameter combinations; the characteristic vectors corresponding to the parameters in the multiple types of operation parameter combinations are extracted; wherein the multiple types of operation parameter combinations are used in order from large to small in terms of influence; based on the characteristic vectors and the state characteristics corresponding to the obtained vehicle operation state, the multiple types of operation parameter combinations are matched with the vehicle operation state to determine the optimal strategy configuration for vehicle operation. The technical solution provided by the present application can effectively improve the accuracy of the comprehensive optimization strategy for vehicle operation efficiency.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of performance gain technology, and in particular, to a comprehensive optimization method and system for full-link vehicle operation performance. Background Art

[0002] With the development of the logistics industry and the sharing economy, vehicles, as important means of transportation, have a direct impact on the cost control and service quality of enterprises. Therefore, how to improve the operational efficiency of vehicles and reduce unnecessary waste of resources through technical means has become an urgent problem to be solved. Especially for enterprises with a large fleet, being able to monitor and optimize the operating status of each vehicle in real time is crucial to improving overall operational efficiency.

[0003] At present, some solutions on the market have attempted to improve vehicle operation efficiency through data analysis, such as collecting data through GPS tracking of vehicle location, fuel consumption monitoring, etc., and using these data for simple statistical analysis to guide scheduling decisions. In addition, some advanced systems will also combine external information such as weather forecasts and traffic conditions to assist in making more reasonable plans.

[0004] However, most of the above methods remain at the basic stage of data collection and simple analysis, lacking effective integration and in-depth mining of complex operating parameters. Especially when faced with massive amounts of data, traditional data analysis methods often cannot accurately identify which operating parameters have a decisive impact on vehicle operating efficiency, resulting in inaccurate optimization strategies, and may even lead to reduced overall benefits due to neglect of certain key factors. Summary of the invention

[0005] The embodiments of the present application provide a comprehensive optimization method and system for full-link vehicle operation efficiency, so as to solve the problem that the optimization strategy in the prior art is not accurate enough.

[0006] In a first aspect, an embodiment of the present application provides a comprehensive optimization method for full-link vehicle operation efficiency, including:

[0007] Obtain multiple operating parameters during vehicle operation;

[0008] Classifying the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency to obtain a plurality of operating parameter combinations;

[0009] Extracting feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations, wherein the plurality of types of operation parameter combinations are arranged in order of influence from greatest to least;

[0010] Based on the feature vector and the acquired state feature corresponding to the vehicle operation state, the plurality of types of operation parameter combinations are matched with the vehicle operation state respectively to determine the optimal strategy configuration for vehicle operation.

[0011] Optionally, the classifying the multiple operating parameters according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations includes:

[0012] identifying a degree of influence of a parameter among the plurality of operating parameters;

[0013] Parameters with the same impact degree are combined to obtain at least one type of operation parameter combination.

[0014] Optionally, identifying the influence degree of a parameter among the plurality of operating parameters comprises:

[0015] Evaluating the influence factors of the parameters among the plurality of operating parameters on the vehicle operating efficiency in a pre-established historical data set;

[0016] The influence degree of the parameter is determined according to the size of the influence factor.

[0017] Optionally, the matching of the plurality of operation parameter combinations with the vehicle operation state respectively based on the feature vector and the acquired state feature corresponding to the vehicle operation state to determine the optimal strategy configuration for vehicle operation includes:

[0018] Based on the feature vector and the state feature corresponding to the acquired vehicle operation state, matching the acquired vehicle operation state with the parameters in the combination of multiple types of operation parameters with the highest degree of influence to obtain a preliminary matching result;

[0019] Starting from the multi-class operating parameter combination of the next impact level of the multi-class operating parameter combination of the highest impact level, determining at least one parameter in the multi-class operating parameter combination of the current impact level that is correlated with the preliminary matching result in the multi-class operating parameter combination of the previous impact level;

[0020] Matching the at least one parameter with the vehicle operation state to obtain a target matching result;

[0021] Using at least one target matching result corresponding to the combination of the multiple types of operating parameters as the optimal strategy configuration for the vehicle operation;

[0022] The matching score corresponding to the target matching result is calculated by the following formula:

[0023] ;

[0024] in, is the eigenvector and state characteristics The matching score between Represents the feature vector and state characteristics The number of elements contained in It is The weight of the parameters, is the eigenvector Middle The value of the element represents the The characteristic value of the operating parameter, It is a state feature Middle The value of an element represents an aspect of the vehicle status. and The eigenvectors are and state characteristics The mean of and The eigenvectors are and state characteristics The standard deviation of is the weight of the bias term, is a predefined constant used to adjust the baseline of the matching score.

[0025] Optionally, the matching of the plurality of operation parameter combinations with the vehicle operation state respectively based on the feature vector and the acquired state feature corresponding to the vehicle operation state to determine the optimal strategy configuration for vehicle operation includes:

[0026] Calculating feature correlations between feature vectors corresponding to parameters in the plurality of types of operation parameter combinations and the acquired state features of vehicle operation;

[0027] determining at least one optimal strategy configuration in which the feature correlation satisfies a correlation condition;

[0028] The correlation score of the feature correlation is calculated by the following formula:

[0029] ;

[0030] in, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in is the eigenvector Middle The value of the element, It is a state feature Middle The value of the element, and Represents the eigenvectors and state characteristics The average value of is the weight of the adjustment term, is a predefined adjustment factor used to fine-tune the relevance score of feature correlation.

[0031] Optionally, the acquiring of a plurality of operating parameters during vehicle operation includes:

[0032] Obtain a parameter list including multiple operation parameters provided by a real-time acquisition system;

[0033] Acquire multiple operating parameters during vehicle operation from the parameter list;

[0034] The step of extracting the feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations comprises:

[0035] A machine learning model is used to extract feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations, wherein the machine learning model is a pre-trained model, and during the training process, the machine learning model uses a preset loss function to maximize the correlation between features while minimizing redundancy.

[0036] Wherein, the preset loss function is defined as:

[0037] ;

[0038] in, is the preset loss function, are model parameters, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in represents the number of eigenvectors, and is the weight of the regularization term, is the redundancy penalty term, is the eigenvector and the eigenvector The correlation score between is the eigenvector and the eigenvector The correlation between the features;

[0039] Wherein, the feature correlation is calculated in the manner of Pearson correlation coefficient, ,in, and The eigenvectors are and No. The value of the element, and The eigenvectors are and The average value of .

[0040] Optionally, the vehicle operation status is vehicle position information that changes in real time;

[0041] The method further comprises:

[0042] Based on the optimal strategy configuration, adjusting the location scheduling of the vehicle; or, based on the optimal strategy configuration, predicting the energy consumption of the vehicle; or, based on the optimal strategy configuration, optimizing the maintenance plan of the vehicle, and determining the optimal maintenance strategy by calculating the ratio of maintenance cost to expected benefit;

[0043] The energy consumption is calculated by the following formula:

[0044] ;

[0045] in, is the predicted time point Energy consumption, It is a recurrent neural network (RNN) that is particularly suitable for processing sequence data and can capture long-term dependencies in time series. It's time point The characteristic vector of It's time point Historical energy consumption data, is the weight of the drift term, is the drift coefficient used to capture long-term trends;

[0046] Among them, the optimal maintenance strategy is calculated by the following formula:

[0047] ;

[0048] in, is the optimal maintenance strategy. is the maintenance cost, Given a maintenance strategy The expected return under is the weight of the penalty term, is a penalty factor used to prevent premature or late maintenance.

[0049] In a second aspect, the embodiment of the present application provides a comprehensive optimization system for full-link vehicle operation efficiency, including:

[0050] An acquisition module, used to acquire multiple operating parameters during the operation of the vehicle;

[0051] A classification module, used to classify the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency, so as to obtain a plurality of types of operating parameter combinations;

[0052] An extraction module, used to extract the feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations; wherein the plurality of types of operation parameter combinations are used in the order of influence from large to small;

[0053] A matching module is used to match the multiple types of operating parameter combinations with the vehicle operating state in order of influence from large to small, based on the feature vector and the state characteristics corresponding to the acquired vehicle operating state, to determine the optimal strategy configuration for vehicle operation.

[0054] In the third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a comprehensive optimization method for full-link vehicle operation efficiency as described in any one of the first aspects.

[0055] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a comprehensive optimization method for full-link vehicle operation efficiency as described in any one of the first aspects.

[0056] In the embodiment of the present application, a plurality of operating parameters are obtained during the operation of the vehicle; the plurality of operating parameters are classified according to the degree of influence on the vehicle operation efficiency to obtain a plurality of operating parameter combinations; the characteristic vectors corresponding to the parameters in the plurality of operating parameter combinations are extracted; wherein the plurality of operating parameter combinations are used to sort the parameters in the plurality of operating parameter combinations in descending order of influence; based on the characteristic vector and the state characteristics corresponding to the obtained vehicle operation state, the plurality of operating parameter combinations are matched with the vehicle operation state to determine the optimal strategy configuration for vehicle operation. The technical solution provided by the present application can effectively improve the accuracy of the comprehensive optimization strategy for vehicle operation efficiency.

[0057] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 A flowchart of a comprehensive optimization method for full-link vehicle operation efficiency provided in an embodiment of the present application;

[0060] Figure 2 A schematic diagram of the structure of a comprehensive optimization system for full-link vehicle operation efficiency provided in an embodiment of the present application;

[0061] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0063] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0064] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0065] Figure 1 A flowchart of a comprehensive optimization method for full-link vehicle operation efficiency is provided for an embodiment of the present application, such as Figure 1 As shown, the method includes:

[0066] 101. Obtain multiple operating parameters during vehicle operation;

[0067] Optionally, the obtaining of multiple operating parameters during vehicle operation in step 101 includes: obtaining a parameter list including multiple operating parameters provided by a real-time acquisition system; obtaining multiple operating parameters during vehicle operation from the parameter list; and extracting feature vectors corresponding to the parameters in the multiple types of operating parameter combinations respectively includes: extracting feature vectors corresponding to the parameters in the multiple types of operating parameter combinations respectively using a machine learning model, wherein the machine learning model is a pre-trained model, and during the training process, the machine learning model uses a preset loss function to maximize the correlation between features while minimizing redundancy.

[0068] Wherein, the preset loss function is defined as:

[0069] ;

[0070] in, is the preset loss function, are model parameters, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in represents the number of eigenvectors, and is the weight of the regularization term, is the redundancy penalty term, is the eigenvector and the eigenvector The correlation score between is the eigenvector and the eigenvector The correlation between the features;

[0071] Wherein, the feature correlation is calculated in the manner of Pearson correlation coefficient, ,in, and The eigenvectors are and No. The value of the element, and The eigenvectors are and The average value of .

[0072] In this step:

[0073] Get multiple operating parameters during vehicle operation:

[0074] Meaning: Collect various operating parameters generated by the vehicle during operation. These parameters may include but are not limited to the vehicle's speed, acceleration, fuel consumption, driving distance, load weight, temperature, etc.

[0075] Purpose: To provide basic data for subsequent feature extraction and strategy optimization.

[0076] Get a parameter list including multiple operational parameters provided by the real-time acquisition system:

[0077] Meaning: Get a list of multiple operational parameters from real-time acquisition systems (such as vehicle sensors, GPS systems, etc.).

[0078] Purpose: To ensure the real-time and accuracy of data so as to facilitate timely analysis and decision-making.

[0079] Obtain multiple operating parameters during vehicle operation from the parameter list:

[0080] Meaning: Filter out the operating parameters directly related to vehicle operation from the parameter list provided by the above real-time acquisition system.

[0081] Purpose: To reduce the amount of data processing and retain only the parameters that have a direct impact on operational performance.

[0082] The machine learning model is used to extract the feature vectors corresponding to the parameters in the multiple types of operation parameter combinations:

[0083] Meaning: Use a pre-trained machine learning model to extract feature vectors from multiple types of operational parameter combinations.

[0084] Purpose: To capture the intrinsic relationship between parameters through feature vectors to provide support for subsequent matching and optimization.

[0085] Default loss function:

[0086] Definition: Defines a loss function for training a machine learning model that aims to maximize the correlation between features while minimizing redundancy.

[0087] In the present embodiment of the application, it is assumed that a logistics company has a batch of vehicles and needs to optimize its operational efficiency. The following are the detailed implementation steps:

[0088] Get multiple operating parameters during vehicle operation:

[0089] Example: A logistics company installed vehicle sensors and GPS systems, which collected data such as vehicle speed, acceleration, fuel consumption, travel distance, load weight, etc. in real time.

[0090] Specific operations: Collect data through vehicle-mounted sensors, obtain location information through the GPS system, and transmit this data to the central server in real time.

[0091] Get a parameter list including multiple operational parameters provided by the real-time acquisition system:

[0092] Embodiment: The central server receives data streams from vehicle sensors and GPS systems and organizes them into parameter lists.

[0093] Specific operation: The server receives the data stream through the API interface and stores it as a structured parameter list.

[0094] Obtain multiple operating parameters during vehicle operation from the parameter list:

[0095] Embodiment: Filter out the operating parameters directly related to the vehicle operation from the above parameter list.

[0096] Specific operation: Select vehicle speed, acceleration, fuel consumption and driving distance as key operating parameters and ignore other irrelevant parameters.

[0097] The machine learning model is used to extract the feature vectors corresponding to the parameters in the multiple types of operation parameter combinations:

[0098] Example: Extract feature vectors using a pre-trained machine learning model (e.g., convolutional neural network (CNN)).

[0099] Specific operation: Input the operation parameter list and extract the feature vector through the pre-trained CNN model.

[0100] Pre-trained model: Assume that a pre-trained CNN model is used, which has been trained on a large amount of historical data.

[0101] Feature vector extraction: The model outputs a feature vector for each operating parameter.

[0102] Default loss function:

[0103] Embodiment: Define a preset loss function for training a machine learning model.

[0104] Specific operation: Assumption (four operating parameters), (two state features), the loss function can be expressed as:

[0105] ;

[0106] Numerical Example: Assume , , , then the loss function can be expressed as:

[0107] ;

[0108] Relevance score: Assumptions , , and so on.

[0109] Redundancy penalty: Assumption .

[0110] Feature Correlation: Hypothesis .

[0111] Through the above steps, logistics companies can obtain multiple operating parameters of the vehicle in real time, extract feature vectors using machine learning models, and then optimize the model through a preset loss function to achieve comprehensive optimization of vehicle operating efficiency.

[0112] Suppose a logistics company has a fleet of trucks and needs to optimize their operational efficiency. Here are the detailed steps:

[0113] Get multiple operating parameters during vehicle operation:

[0114] Specific operations: Install vehicle-mounted sensors and GPS systems to collect data such as vehicle speed, acceleration, fuel consumption, and driving distance in real time, and transmit the data to the central server through the wireless communication module.

[0115] Example: Assume that a truck generates a data record once a minute, and each record includes:

[0116] Vehicle speed: v = 80 km / h;

[0117] Acceleration: a=2m / s²;

[0118] Fuel consumption: f=10L / 100km;

[0119] Driving distance: d = 120 km;

[0120] Load weight: w=5000kg;

[0121] Get a parameter list including multiple operational parameters provided by the real-time acquisition system:

[0122] Specific operations: The central server receives the above data and organizes it into a parameter list.

[0123] Example: The parameter list is as follows:

[0124] Parameter list = {

[0125] "vehicle speed": [80,81,82,...,85],

[0126] "acceleration": [2,2.1,2.2,...,2.5],

[0127] "Fuel consumption": [10, 10.2, 10.5, ..., 11],

[0128] "Driving distance": [120,121,122,...,125],

[0129] "load weight": [5000,5050,5100,...,5150]

[0130] }

[0131] Obtain multiple operating parameters during vehicle operation from the parameter list:

[0132] Specific operation: Select vehicle speed, acceleration, fuel consumption and driving distance as key operating parameters, and ignore load weight.

[0133] Example: The parameter list is as follows:

[0134] Operation parameter list = {

[0135] "vehicle speed": [80,81,82,...,85],

[0136] "acceleration": [2,2.1,2.2,...,2.5],

[0137] "Fuel consumption": [10, 10.2, 10.5, ..., 11],

[0138] "Driving distance": [120,121,122,...,125]

[0139] }

[0140] The machine learning model is used to extract the feature vectors corresponding to the parameters in the multiple types of operation parameter combinations:

[0141] Specific operation: Use the pre-trained CNN model to extract feature vectors.

[0142] Example: Assume that the feature vector output by the CNN model is as follows:

[0143] Eigenvector list = {

[0144] "Speed ​​feature vector": [0.8, 0.82, 0.84, ..., 0.85],

[0145] "Acceleration eigenvector": [0.2, 0.21, 0.22, ..., 0.25],

[0146] "Fuel consumption feature vector": [0.1, 0.102, 0.105, ..., 0.11],

[0147] "Driving distance feature vector": [1.2,1.21,1.22,...,1.25]

[0148] }

[0149] Default loss function:

[0150] Specific operation: define the preset loss function, assuming , , , then the loss function can be expressed as:

[0151] ;

[0152] Example: Assuming a relevance score , , redundant penalty , feature correlation .

[0153] Through the above embodiments, the logistics company can obtain multiple operating parameters during the operation of the vehicle in real time, and use the machine learning model to extract feature vectors, and then optimize the model through a preset loss function to achieve comprehensive optimization of the vehicle operation efficiency.

[0154] 102. Classify the multiple operating parameters according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations;

[0155] Optionally, the step 102 of classifying the multiple operating parameters according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations includes: identifying the degree of influence of parameters among the multiple operating parameters; and combining parameters with the same degree of influence to obtain at least one type of operating parameter combination.

[0156] Optionally, identifying the degree of influence of a parameter among the multiple operating parameters includes: evaluating the influence factor of the parameter among the multiple operating parameters on the vehicle operating efficiency in a pre-established historical data set; and determining the degree of influence of the parameter according to the size of the influence factor.

[0157] In this step, the multiple operating parameters are classified according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations:

[0158] Meaning: According to the degree of influence of each operating parameter on the vehicle operating efficiency, these parameters are divided into different categories to form a combination of multiple operating parameters.

[0159] Purpose: To facilitate subsequent feature extraction and strategy optimization, and highlight those parameters that have a greater impact on operational efficiency.

[0160] Identifying the degree of influence of a parameter among the plurality of operating parameters:

[0161] Meaning: Determine the impact of various operating parameters on vehicle operating efficiency.

[0162] Purpose: To provide a basis for classification and ensure the rationality and scientificity of classification.

[0163] Evaluate the influence factors of the parameters in the plurality of operating parameters on the vehicle operating efficiency in a pre-established historical data set:

[0164] Meaning: Use historical data sets to evaluate the impact of various operating parameters on vehicle operating efficiency through statistical analysis or machine learning methods.

[0165] Purpose: To quantify the influence of each parameter and provide data support for subsequent classification.

[0166] According to the magnitude of the influencing factor, the degree of influence of the parameter is determined:

[0167] Meaning: Determine the influence of each parameter based on the size of the assessed influencing factor.

[0168] Purpose: To ensure the accuracy of classification so that parameters with greater impact are classified into more important categories.

[0169] In the embodiment of this application, it is assumed that a logistics company has a batch of vehicles and needs to optimize its operational efficiency. The following are the detailed implementation steps:

[0170] The multiple operating parameters are classified according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations:

[0171] Embodiment: The collected multiple operation parameters are classified according to the degree of influence to form a combination of multiple types of operation parameters.

[0172] Specific operation: Sort the parameters from large to small according to the degree of impact, and group the parameters with similar degree of impact together.

[0173] Identifying the degree of influence of a parameter among the plurality of operating parameters:

[0174] Embodiment: The impact factor of each operating parameter on vehicle operating efficiency is evaluated through historical data sets.

[0175] Specific operations: Use data from historical data sets to evaluate the influencing factors of various operating parameters through statistical analysis or machine learning methods.

[0176] Evaluate the influence factors of the parameters in the plurality of operating parameters on the vehicle operating efficiency in a pre-established historical data set:

[0177] Embodiment: Using data from historical data sets, statistical analysis or machine learning methods are used to evaluate the influencing factors of various operating parameters.

[0178] Specific operation: Assume that the historical data set contains vehicle operation data for the past year, including parameters such as vehicle speed, acceleration, fuel consumption, and driving distance.

[0179] According to the magnitude of the influencing factor, the degree of influence of the parameter is determined:

[0180] Embodiment: According to the magnitude of the evaluated influencing factors, the influence degree of each parameter is determined.

[0181] Specific operation: Assume the evaluation results of the impact factor.

[0182] Suppose a logistics company has a fleet of trucks and needs to optimize their operational efficiency. Here are the detailed steps:

[0183] Get the historical dataset:

[0184] Specific operations: Collect vehicle operation data for the past year, including parameters such as speed, acceleration, fuel consumption, and driving distance.

[0185] Example: Assume the historical data set is as follows:

[0186] Historical Dataset = {

[0187] "vehicle speed": [80,81,82,...,85],

[0188] "acceleration": [2,2.1,2.2,...,2.5],

[0189] "Fuel consumption": [10, 10.2, 10.5, ..., 11],

[0190] "Driving distance": [120,121,122,...,125]

[0191] }

[0192] Evaluate impact factor:

[0193] Specific operations: Evaluate the influencing factors of each parameter through correlation analysis and regression analysis.

[0194] Example: Assume that the evaluation results are as follows:

[0195] The correlation coefficient between vehicle speed and operational efficiency is 0.8;

[0196] The correlation coefficient between fuel consumption and operational efficiency is 0.75;

[0197] The correlation coefficient between acceleration and operational effectiveness is 0.5;

[0198] The correlation coefficient between driving distance and operational efficiency is 0.3;

[0199] Determine the extent of the impact:

[0200] Specific operation: Determine the influence degree of each parameter according to the size of the evaluated influencing factor.

[0201] Example: Determine the influence of each parameter according to the size of the influencing factor:

[0202] Big influence: vehicle speed, fuel consumption

[0203] Medium Impact: Acceleration

[0204] Less impact: driving distance

[0205] Classification combination:

[0206] What it does: Group together parameters of similar impact.

[0207] Example:

[0208] The combination is as follows:

[0209] Multiple types of operation parameter combinations = {

[0210] "Great impact": ["vehicle speed", "fuel consumption"],

[0211] "Medium Impact": ["Acceleration"],

[0212] "less impact": ["driving distance"]

[0213] }

[0214] Through the above embodiments, the logistics company can reasonably classify the operation parameters according to the influencing factors evaluated in the historical data set to form a combination of multiple types of operation parameters, thereby providing support for subsequent feature extraction and strategy optimization.

[0215] 103. Extracting feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations, wherein the plurality of types of operation parameter combinations are arranged in order of influence from greatest to least;

[0216] In this step, we will focus on how to extract the feature vectors from the operating parameter combinations and rank them according to their impact on vehicle operating effectiveness and cost-effectiveness.

[0217] Feature vector: Combines multiple operating parameters into a vector to represent the status of vehicle operation.

[0218] Suppose we are developing a comprehensive optimization method for the full-link vehicle operation efficiency for a logistics company. We need to extract useful features from the actual operation data of the vehicle and rank them according to their importance in order to further optimize the operation strategy.

[0219] Collect operational parameters:

[0220] First, we collect the following types of operational parameters: average speed ( ), fuel consumption rate ( ), load ( ), driving distance ( ), maintenance interval ( ).

[0221] Define the eigenvector:

[0222] For each car, we define a feature vector that contains all the above parameters. For example, the feature vector of a car can be expressed as [ ].

[0223] Calculate the impact:

[0224] In order to determine the impact of these parameters on vehicle operating efficiency and cost-effectiveness, we can use a variety of statistical methods or machine learning models. Here we take multivariate linear regression as an example:

[0225] ;

[0226] in, It is a quantitative indicator of cost-effectiveness. is the intercept term, to is the regression coefficient, is the error term.

[0227] Sort the eigenvectors:

[0228] By training the model, we can get the regression coefficient of each parameter The absolute values ​​of these coefficients reflect the degree of influence of the parameters on cost-effectiveness. For example, if , it means that fuel consumption rate has the greatest impact on cost-effectiveness, followed by average speed, and so on.

[0229] Application sorting results:

[0230] Based on the results of the impact ranking, we can reorganize the order of the feature vectors to better understand which factors need the most attention. For example, the new feature vector can represent [ ].

[0231] Example numericalization:

[0232] Suppose we have a set of data as follows:

[0233] Vehicle A: [60km / h, 10L / 100km, 2000kg, 500km, 10000km]

[0234] Vehicle B: [55km / h, 12L / 100km, 1800kg, 450km, 12000km]

[0235] After training the model, the regression coefficients we get may be:

[0236] ;

[0237] From these coefficients, it can be seen that the fuel consumption rate ( ) has the greatest impact on cost-effectiveness because its coefficient has the largest absolute value.

[0238] Through this approach, we can provide logistics companies with a more scientific decision support tool to help them optimize vehicle operation strategies and improve overall cost-effectiveness.

[0239] 104. Based on the feature vector and the acquired state feature corresponding to the vehicle operation state, the plurality of types of operation parameter combinations are matched with the vehicle operation state respectively to determine the optimal strategy configuration for vehicle operation.

[0240] Optionally, the step 104 of matching the plurality of operation parameter combinations with the vehicle operation state based on the feature vector and the acquired state feature corresponding to the vehicle operation state to determine the optimal strategy configuration for vehicle operation includes:

[0241] Based on the feature vector and the state feature corresponding to the acquired vehicle operation state, matching the acquired vehicle operation state with the parameters in the combination of multiple types of operation parameters with the highest degree of influence to obtain a preliminary matching result;

[0242] Starting from the multi-class operating parameter combination of the next impact level of the multi-class operating parameter combination of the highest impact level, determining at least one parameter in the multi-class operating parameter combination of the current impact level that is correlated with the preliminary matching result in the multi-class operating parameter combination of the previous impact level;

[0243] Matching the at least one parameter with the vehicle operation state to obtain a target matching result;

[0244] Using at least one target matching result corresponding to the combination of the multiple types of operating parameters as the optimal strategy configuration for the vehicle operation;

[0245] The matching score corresponding to the target matching result is calculated by the following formula:

[0246] ;

[0247] in, is the eigenvector and state characteristics The matching score between Represents the feature vector and state characteristics The number of elements contained in It is The weight of the parameters, is the eigenvector Middle The value of the element represents the The characteristic value of the operating parameter, It is a state feature Middle The value of an element represents an aspect of the vehicle status. and The eigenvectors are and state characteristics The mean of and The eigenvectors are and state characteristics The standard deviation of is the weight of the bias term, is a predefined constant used to adjust the baseline of the matching score.

[0248] Optionally, the step 104 of matching the plurality of operation parameter combinations with the vehicle operation state based on the feature vector and the acquired state feature corresponding to the vehicle operation state to determine the optimal strategy configuration for vehicle operation includes:

[0249] Calculating feature correlations between feature vectors corresponding to parameters in the plurality of types of operation parameter combinations and the acquired state features of vehicle operation;

[0250] determining at least one optimal strategy configuration in which the feature correlation satisfies a correlation condition;

[0251] The correlation score of the feature correlation is calculated by the following formula:

[0252] ;

[0253] in, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in is the eigenvector Middle The value of the element, It is a state feature Middle The value of the element, and Represents the eigenvectors and state characteristics The average value of is the weight of the adjustment term, is a predefined adjustment factor used to fine-tune the relevance score of feature correlation.

[0254] In this step, "based on the feature vector and the state characteristics corresponding to the acquired vehicle operation state, the multiple types of operation parameter combinations are matched with the vehicle operation state to determine the best strategy configuration for vehicle operation" means that by analyzing the state characteristics and feature vectors of vehicle operation, the operation parameter combination that best matches the current state of the vehicle is found, and then the best vehicle operation strategy configuration is determined. This process includes matching operation parameter combinations with different degrees of influence, calculating matching scores or correlation scores, and finally determining the best strategy configuration.

[0255] In the embodiment of the present application, it is assumed that a logistics company needs to develop an optimal operation strategy for its fleet to improve operational efficiency and reduce costs. The system has obtained multiple combinations of operation parameters and state characteristics of vehicle operation, and will determine the optimal strategy configuration based on this information.

[0256] The specific embodiments are as follows:

[0257] Get feature vectors and vehicle operation status features

[0258] Eigenvector: Assume that the eigenvector Contains the following operating parameters:

[0259] : Vehicle load capacity;

[0260] : Vehicle fuel efficiency;

[0261] : Vehicle maintenance cycle;

[0262] : Driver working hours;

[0263] : route busyness;

[0264] State characteristics of vehicle operation: Assumed state characteristics It includes the following aspects:

[0265] : Current cargo capacity;

[0266] : Current remaining fuel amount;

[0267] : The time of the most recent maintenance;

[0268] : The driver’s current working hours;

[0269] : Current route busyness;

[0270] Calculate the matching score:

[0271] Matching score calculation formula:

[0272] ;

[0273] in, is the eigenvector and state characteristics The matching score between Represents the feature vector and state characteristics The number of elements contained in It is The weight of the parameters, is the eigenvector Middle The value of the element represents the The characteristic value of the operating parameter, It is a state feature Middle The value of an element represents an aspect of the vehicle status. and The eigenvectors are and state characteristics The mean of and The eigenvectors are and state characteristics The standard deviation of is the weight of the bias term, is a predefined constant used to adjust the baseline of the matching score.

[0274] Assume that the feature vector obtained by the system is and state characteristics as follows:

[0275] (Vehicle load capacity, vehicle fuel efficiency, vehicle maintenance cycle, driver working hours, route busyness)

[0276] (Current cargo volume, current remaining fuel, last maintenance time, driver’s current working hours, current route busyness)

[0277] Assume that the calculated mean and standard deviation are as follows:

[0278] ;

[0279] Assumption Weight and the weight of the bias term as follows:

[0280] ;

[0281] Calculate the matching score:

[0282] ;

[0283] Calculate feature correlations:

[0284] Feature correlation calculation formula:

[0285] ;

[0286] in, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in is the eigenvector Middle The value of the element, It is a state feature Middle The value of the element, and Represents the eigenvectors and state characteristics The average value of is the weight of the adjustment term, is a predefined adjustment factor used to fine-tune the relevance score of feature correlation.

[0287] Assume that the feature vector obtained by the system is and state characteristics As mentioned above.

[0288] Assume that the calculated average is as follows:

[0289] ;

[0290] Assume the weight of the adjustment and adjustment factor as follows:

[0291] ;

[0292] Calculate feature correlations:

[0293] ;

[0294] Determine the optimal policy configuration:

[0295] Based on the matching score and feature relevance score, the system can determine the operating parameter combination that best matches the vehicle's operating status. Assuming that the system has calculated the matching scores and relevance scores of multiple operating parameter combinations, the system will select the combination with the highest score as the best strategy configuration.

[0296] Through the above steps, the system determines the optimal strategy configuration for vehicle operation based on the feature vector and the state characteristics of vehicle operation through matching scores and feature correlation scores.

[0297] Optionally, the vehicle operation status is vehicle position information that changes in real time; the method further includes:

[0298] Based on the optimal strategy configuration, adjusting the location scheduling of the vehicle; or, based on the optimal strategy configuration, predicting the energy consumption of the vehicle; or, based on the optimal strategy configuration, optimizing the maintenance plan of the vehicle, and determining the optimal maintenance strategy by calculating the ratio of maintenance cost to expected benefit;

[0299] The energy consumption is calculated by the following formula:

[0300] ;

[0301] in, is the predicted time point Energy consumption, It is a recurrent neural network (RNN) that is particularly suitable for processing sequence data and can capture long-term dependencies in time series. It's time point The characteristic vector of It's time point Historical energy consumption data, is the weight of the drift term, is the drift coefficient used to capture long-term trends;

[0302] Among them, the optimal maintenance strategy is calculated by the following formula:

[0303] ;

[0304] in, is the optimal maintenance strategy. is the maintenance cost, Given a maintenance strategy The expected return under is the weight of the penalty term, is a penalty factor used to prevent premature or late maintenance.

[0305] In this step, "adjusting the location scheduling of the vehicle based on the optimal strategy configuration; or, predicting the energy consumption of the vehicle based on the optimal strategy configuration; or, optimizing the maintenance plan of the vehicle based on the optimal strategy configuration" means that one of the three operations can be performed according to the determined optimal strategy configuration: adjusting the location scheduling of the vehicle, predicting the energy consumption of the vehicle, and optimizing the maintenance plan of the vehicle. These operations are intended to improve the efficiency of vehicle operation, reduce energy consumption, and optimize maintenance plans through the optimal strategy configuration.

[0306] In the embodiment of this application, it is assumed that a logistics company needs to develop an optimal operation strategy for its fleet to improve operational efficiency and reduce costs. The system has obtained multiple combinations of operation parameters and the state characteristics of vehicle operation, and will determine the optimal strategy configuration based on this information. Specifically, we will show how to predict the energy consumption of the vehicle based on the optimal strategy configuration.

[0307] Predict vehicle energy consumption:

[0308] Energy consumption prediction formula:

[0309] ;

[0310] in, is the predicted time point Energy consumption, It is a recurrent neural network (RNN) that is particularly suitable for processing sequence data and can capture long-term dependencies in time series. It's time point The characteristic vector of It's time point Historical energy consumption data, is the weight of the drift term, is the drift coefficient used to capture long-term trends;

[0311] Specific numerical calculation:

[0312] Assume that the feature vector obtained by the system is and historical energy consumption data as follows:

[0313] (Vehicle load capacity, vehicle fuel efficiency, vehicle maintenance cycle, driver working hours, route busyness)

[0314] (Energy consumption data at several points in the past)

[0315] Assume that the calculated drift term weight is And the drift coefficient is as follows:

[0316] ;

[0317] use Model predictions Energy consumption at each moment:

[0318] ;

[0319] Assumptions The model predicts the following results:

[0320] ;

[0321] Therefore, the predicted energy consumption is:

[0322] ;

[0323] Optimize vehicle maintenance schedule:

[0324] Optimal maintenance strategy calculation formula:

[0325] ;

[0326] in, is the optimal maintenance strategy. is the maintenance cost, Given a maintenance strategy The expected return under is the weight of the penalty term, is a penalty factor used to prevent premature or late maintenance.

[0327] Assume that the system obtains the maintenance cost and expected returns as follows:

[0328] (Maintenance cost)

[0329] (Expected Return)

[0330] Assume that the calculated penalty weight is And the penalty coefficient is as follows:

[0331] ;

[0332] Calculate the optimal maintenance strategy:

[0333] ;

[0334] Since this is a minimization problem, we need to choose the expression that minimizes the above expression. In this simple example, we assume that It has been determined that .

[0335] Through the above steps, the system predicts the energy consumption of the vehicle based on the optimal strategy configuration, and determines the optimal maintenance strategy by calculating the ratio of maintenance cost to expected benefit. This can effectively improve the efficiency of vehicle operation and reduce costs.

[0336] Assume that a logistics company has a transport vehicle and the system has determined the optimal policy configuration. Now, we will show how to predict the energy consumption of the vehicle based on the optimal policy configuration and optimize the maintenance plan.

[0337] Get feature vectors and vehicle operation status features

[0338] Eigenvector

[0339] Historical energy consumption data

[0340] Predict vehicle energy consumption:

[0341] use Model predictions Energy consumption at each moment:

[0342] ;

[0343] Optimize vehicle maintenance schedule:

[0344] Maintenance costs

[0345] Expected Returns

[0346] Calculate the optimal maintenance strategy:

[0347] ;

[0348] Through the above steps, the system predicts the energy consumption of the vehicle based on the optimal strategy configuration, and determines the optimal maintenance strategy by calculating the ratio of maintenance cost to expected benefit. This helps logistics companies better manage and dispatch vehicles, improve operational efficiency and reduce costs.

[0349] Figure 2 A structural diagram of a comprehensive optimization system for full-link vehicle operation efficiency is provided for the embodiment of the present application, such as Figure 2 As shown, the device comprises:

[0350] An acquisition module 21 is used to acquire a plurality of operation parameters during the operation of the vehicle;

[0351] A classification module 22, configured to classify the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency, so as to obtain a plurality of types of operating parameter combinations;

[0352] An extraction module 23 is used to extract the feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations; wherein the plurality of types of operation parameter combinations are used in the order of influence from large to small;

[0353] The matching module 24 is used to match the multiple types of operating parameter combinations with the vehicle operating state in order of influence from large to small, based on the feature vector and the state characteristics corresponding to the acquired vehicle operating state, to determine the optimal strategy configuration for vehicle operation.

[0354] Figure 2 The comprehensive optimization system for the whole-link vehicle operation efficiency can be implemented Figure 1 The implementation principle and technical effects of the comprehensive optimization method for the full-link vehicle operation efficiency described in the illustrated embodiment will not be described in detail. The specific manner in which each module and unit performs operations in the comprehensive optimization system for the full-link vehicle operation efficiency in the above embodiment has been described in detail in the embodiment of the method, and will not be elaborated here.

[0355] In one possible design, Figure 2 The comprehensive optimization system for the full-link vehicle operation efficiency of the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0356] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0357] The processing component 32 is used to: obtain multiple operating parameters during vehicle operation; classify the multiple operating parameters according to the degree of influence on the vehicle operating efficiency to obtain multiple types of operating parameter combinations; extract the characteristic vectors corresponding to the parameters in the multiple types of operating parameter combinations; wherein the multiple types of operating parameter combinations are used in order from large to small in terms of influence; based on the characteristic vectors and the state characteristics corresponding to the obtained vehicle operating state, the multiple types of operating parameter combinations are matched with the vehicle operating state to determine the optimal strategy configuration for vehicle operation.

[0358] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0359] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0360] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0361] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0362] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0363] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0364] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The comprehensive optimization method of the full-link vehicle operation efficiency of the illustrated embodiment.

[0365] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0366] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0367] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0368] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A comprehensive optimization method for full-link vehicle operation efficiency, characterized in that: include: Acquire multiple operating parameters during the operation of the vehicle, wherein the multiple operating parameters include at least the speed, acceleration, fuel consumption, travel distance, load weight, and temperature of the vehicle; Classifying the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency to obtain a plurality of operating parameter combinations; Extracting feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations, wherein the plurality of types of operation parameter combinations are arranged in descending order of influence; Based on the feature vector and the state feature corresponding to the acquired vehicle operation state, the plurality of operation parameter combinations are matched with the vehicle operation state respectively to determine the optimal strategy configuration for vehicle operation; wherein the vehicle operation state is the vehicle position information that changes in real time; The step of matching the plurality of operation parameter combinations with the vehicle operation state to determine the optimal strategy configuration for vehicle operation based on the feature vector and the acquired state feature corresponding to the vehicle operation state includes: Based on the feature vector and the state feature corresponding to the acquired vehicle operation state, matching the acquired vehicle operation state with the parameters in the combination of multiple types of operation parameters with the highest degree of influence to obtain a preliminary matching result; Starting from the multi-class operating parameter combination of the next impact level of the multi-class operating parameter combination of the highest impact level, determining at least one parameter in the multi-class operating parameter combination of the current impact level that is correlated with the preliminary matching result in the multi-class operating parameter combination of the previous impact level; Matching the at least one parameter with the vehicle operation state to obtain a target matching result; Using at least one target matching result corresponding to the combination of the multiple types of operating parameters as the optimal strategy configuration for the vehicle operation; The matching score corresponding to the target matching result is calculated by the following formula: ； in, is the eigenvector and state characteristics The matching score between Represents the feature vector and state characteristics The number of elements contained in It is The weight of the parameters, is the eigenvector Middle The value of the element represents the The characteristic value of the operating parameter, It is a state feature Middle The value of an element represents an aspect of the vehicle status. and The eigenvectors are and state characteristics The mean of and The eigenvectors are and state characteristics The standard deviation of is the weight of the bias term, is a predefined constant used to adjust the baseline of the matching score.

2. The method according to claim 1, characterized in that The step of classifying the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency to obtain a plurality of operating parameter combinations includes: identifying a degree of influence of a parameter among the plurality of operating parameters; Parameters with the same impact degree are combined to obtain at least one type of operation parameter combination.

3. The method according to claim 2, characterized in that The identifying the influence degree of a parameter among the plurality of operating parameters comprises: Evaluating the influence factors of the parameters among the plurality of operating parameters on the vehicle operating efficiency in a pre-established historical data set; The influence degree of the parameter is determined according to the size of the influence factor.

4. The method according to claim 1, characterized in that The step of matching the plurality of operation parameter combinations with the vehicle operation state to determine the optimal strategy configuration for vehicle operation based on the feature vector and the acquired state feature corresponding to the vehicle operation state includes: Calculating feature correlations between feature vectors corresponding to parameters in the plurality of types of operation parameter combinations and the acquired state features of vehicle operation; determining at least one optimal strategy configuration in which the feature correlation satisfies a correlation condition; The correlation score of the feature correlation is calculated by the following formula: ； in, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in is the eigenvector Middle The value of the element, It is a state feature Middle The value of the element, and Represents the eigenvectors and state characteristics The average value of is the weight of the adjustment term, is a predefined adjustment factor used to fine-tune the relevance score of feature correlation.

5. The method according to claim 1, characterized in that The acquisition of multiple operating parameters during vehicle operation includes: Obtain a parameter list including multiple operation parameters provided by a real-time acquisition system; Acquire multiple operating parameters during vehicle operation from the parameter list; The step of extracting the feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations comprises: A machine learning model is used to extract feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations, wherein the machine learning model is a pre-trained model, and during the training process, the machine learning model uses a preset loss function to maximize the correlation between features while minimizing redundancy. Wherein, the preset loss function is defined as: ； in, is the preset loss function, are model parameters, is the eigenvector and state characteristics The correlation score between Represents the feature vector and state characteristics The number of elements contained in represents the number of eigenvectors, and is the weight of the regularization term, is the redundancy penalty term, is the eigenvector and the eigenvector The correlation score between is the eigenvector and the eigenvector The correlation between the features; Wherein, the feature correlation is calculated in the manner of Pearson correlation coefficient, ,in, and The eigenvectors are and No. The value of the element, and The eigenvectors are and The average value of .

6. The method according to claim 1, characterized in that The vehicle operation status is the vehicle position information that changes in real time; The method further comprises: Based on the optimal strategy configuration, adjusting the location scheduling of the vehicle; or, based on the optimal strategy configuration, predicting the energy consumption of the vehicle; or, based on the optimal strategy configuration, optimizing the maintenance plan of the vehicle, and determining the optimal maintenance strategy by calculating the ratio of maintenance cost to expected benefit; The energy consumption is calculated by the following formula: ； in, is the predicted time point Energy consumption, It is a recurrent neural network (RNN) that is particularly suitable for processing sequence data and can capture long-term dependencies in time series. It's time point The characteristic vector of It's time point Historical energy consumption data, is the weight of the drift term, is the drift coefficient used to capture long-term trends; Among them, the optimal maintenance strategy is calculated by the following formula: ； in, is the optimal maintenance strategy. is the maintenance cost, Given a maintenance strategy The expected return under is the weight of the penalty term, is a penalty factor used to prevent premature or late maintenance.

7. A comprehensive optimization system for full-link vehicle operation efficiency, characterized in that: include: An acquisition module, used to acquire multiple operating parameters during the operation of the vehicle; A classification module, used for classifying the plurality of operating parameters according to the degree of influence on the vehicle operating efficiency, so as to obtain a plurality of types of operating parameter combinations; An extraction module, used to extract the feature vectors corresponding to the parameters in the plurality of types of operation parameter combinations; wherein the plurality of types of operation parameter combinations are used in the order of influence from large to small; A matching module, configured to match the plurality of types of operation parameter combinations with the vehicle operation state in order of influence from large to small, based on the feature vector and the acquired state characteristics corresponding to the vehicle operation state, respectively, to determine the optimal strategy configuration for vehicle operation; The step of matching the plurality of operation parameter combinations with the vehicle operation state to determine the optimal strategy configuration for vehicle operation based on the feature vector and the acquired state feature corresponding to the vehicle operation state includes: Based on the feature vector and the state feature corresponding to the acquired vehicle operation state, matching the acquired vehicle operation state with the parameters in the combination of multiple types of operation parameters with the highest degree of influence to obtain a preliminary matching result; Starting from the multi-class operating parameter combination of the next impact level of the multi-class operating parameter combination of the highest impact level, determining at least one parameter in the multi-class operating parameter combination of the current impact level that is correlated with the preliminary matching result in the multi-class operating parameter combination of the previous impact level; Matching the at least one parameter with the vehicle operation state to obtain a target matching result; Using at least one target matching result corresponding to the combination of the multiple types of operating parameters as the optimal strategy configuration for the vehicle operation; The matching score corresponding to the target matching result is calculated by the following formula: ； in, is the eigenvector and state characteristics The matching score between Represents the feature vector and state characteristics The number of elements contained in It is The weight of the parameters, is the eigenvector Middle The value of the element represents the The characteristic value of the operating parameter, It is a state feature Middle The value of an element represents an aspect of the vehicle status. and The eigenvectors are and state characteristics The mean of and The eigenvectors are and state characteristics The standard deviation of is the weight of the bias term, is a predefined constant used to adjust the baseline of the matching score.

8. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a comprehensive optimization method for full-link vehicle operation efficiency as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a comprehensive optimization method for full-link vehicle operation efficiency as described in any one of claims 1 to 6 is implemented.

Images