Energy efficiency optimization method and device based on vehicle energy consumption prediction, equipment and storage medium

Abstract

The invention relates to an energy efficiency optimization method and device based on vehicle energy consumption prediction, equipment and a storage medium. The method comprises the steps of collecting multi-source driving data of a vehicle in real time; determining multiple types of energy consumption characteristic data based on the multi-source driving data; inputting the energy consumption characteristic data into a trained energy consumption prediction model, predicting the energy consumption of the vehicle in a preset time period, and obtaining an energy consumption prediction result; determining a causal relationship between different types of energy consumption characteristic data and energy consumption data, and the importance of each energy consumption characteristic data; according to the causal relationship between the different types of energy consumption characteristic data and the energy consumption data and the importance degree of each energy consumption characteristic data, determining the contribution degree of the different types of energy consumption characteristic data to the vehicle energy consumption; and determining an energy efficiency optimization suggestion based on the energy consumption prediction result and the contribution degrees of different types of energy consumption characteristic data to the energy consumption. The method is beneficial to improving the effectiveness of the vehicle energy efficiency suggestion.

Description

Technical Field

[0001] The present application relates to the field of new energy technologies, and in particular to an energy efficiency optimization method, apparatus, computer equipment, computer-readable storage medium, and computer program product based on vehicle energy consumption prediction. Background Art

[0002] With the continuous development of electric vehicle technology, electric vehicles have become an essential component of future transportation. However, predicting electric vehicle energy consumption still faces many challenges, particularly in accurately quantifying the contribution of multiple factors to electric vehicle energy consumption. Currently, while machine learning can be used to predict electric vehicle energy consumption by combining extensive real-world driving data with road and traffic information, it is still difficult to accurately quantify the specific impact of each factor on energy consumption, and these factors are intertwined and difficult to separate.

[0003] Traditional electric vehicle energy consumption prediction methods mainly include physical modeling, statistical regression and traditional machine learning. However, traditional electric vehicle energy consumption prediction methods have the problem of low prediction accuracy, which makes the effectiveness of vehicle energy efficiency optimization low. Summary of the Invention

[0004] Based on this, it is necessary to provide an energy efficiency optimization method, device, computer equipment, computer-readable storage medium and computer program product based on vehicle energy consumption prediction that can improve the effectiveness of vehicle energy efficiency optimization in response to the above technical problems.

[0005] In a first aspect, the present application provides an energy efficiency optimization method based on vehicle energy consumption prediction, comprising:

[0006] Real-time collection of multi-source driving data of vehicles;

[0007] Determine various types of energy consumption characteristic data based on multi-source driving data;

[0008] Input the energy consumption feature data into the trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period to obtain the energy consumption prediction result. The energy consumption prediction model is trained based on the historical energy consumption feature data carrying energy consumption labels;

[0009] Determine the causal relationship between different types of energy consumption characteristic data and energy consumption data, as well as the importance of each energy consumption characteristic data;

[0010] Determine the contribution of different types of energy consumption characteristic data to vehicle energy consumption based on the causal relationship between different types of energy consumption characteristic data and energy consumption data, as well as the importance of each type of energy consumption characteristic data;

[0011] Based on the energy consumption forecast results and the contribution of different types of energy consumption characteristic data to energy consumption, energy efficiency optimization suggestions are determined.

[0012] In one embodiment, the multi-source driving data includes at least two of driving behavior data, road condition data, environmental condition data, and vehicle configuration data; and determining multiple types of energy consumption characteristic data based on the multi-source driving data includes at least two of the following methods:

[0013] Determining driving behavior characteristic data based on the driving behavior data;

[0014] Determining road condition characteristic data based on the road condition data;

[0015] determining environmental condition characteristic data based on the environmental condition data;

[0016] Based on the vehicle configuration data, vehicle configuration characteristic data is determined.

[0017] In one embodiment, the driving behavior data includes vehicle speed sequence data;

[0018] Based on the driving behavior data, driving behavior characteristic data is determined, including:

[0019] Based on the vehicle speed sequence data, determine the average acceleration data and the average braking intensity;

[0020] Determine acceleration frequency and braking frequency based on vehicle speed sequence data;

[0021] determining a driving style of a subject driving the vehicle based on the acceleration frequency and the braking frequency;

[0022] Driving behavior characteristic data includes average acceleration data, average braking intensity and driving style.

[0023] In one embodiment, the road condition data includes road surface attribute data;

[0024] Based on the road condition data, road condition characteristic data is determined, including:

[0025] Determine the average road slope and road type based on road surface attribute data;

[0026] Road condition characteristic data include average road slope and road type.

[0027] In one embodiment, the vehicle configuration data includes battery attribute data and electric motor power;

[0028] Based on the vehicle configuration data, vehicle configuration characteristic data is determined, including:

[0029] Assess the vehicle's battery health based on battery attribute data and electric motor power;

[0030] Vehicle configuration characteristic data includes battery health.

[0031] In one embodiment, the environmental condition data includes temperature data and humidity data;

[0032] Based on the environmental condition data, environmental condition characteristic data is determined, including:

[0033] determining an average temperature and an average humidity based on the temperature data and the humidity data, respectively;

[0034] Environmental condition characteristic data include average temperature and average humidity.

[0035] In one embodiment, determining the importance of each energy consumption characteristic data includes:

[0036] For each energy consumption characteristic data, determine the change in the target energy consumption forecast result compared to the energy consumption forecast result after removing the energy consumption characteristic data;

[0037] Determine the importance of energy consumption characteristic data based on the amount of change.

[0038] In one embodiment, determining the causal relationship between different types of energy consumption characteristic data and energy consumption data includes:

[0039] Obtain historical energy consumption data of the vehicle;

[0040] For each type of energy consumption characteristic data, the impact intensity of the energy consumption characteristic data on the energy consumption data is obtained based on the energy consumption characteristic data, historical energy consumption data and the established causal relationship analysis model;

[0041] According to the impact intensity, determine the causal relationship between energy consumption characteristic data and energy consumption data.

[0042] In a second aspect, the present application further provides an energy efficiency optimization device based on vehicle energy consumption prediction, comprising:

[0043] Data acquisition module, used to collect multi-source driving data of the vehicle in real time;

[0044] A feature extraction module for determining various types of energy consumption feature data based on multi-source driving data;

[0045] The energy consumption prediction module is used to input the energy consumption feature data into the trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period and obtain the energy consumption prediction result. The energy consumption prediction model is trained based on the historical energy consumption feature data carrying energy consumption labels;

[0046] An impact analysis module is used to determine the causal relationship between different types of energy consumption characteristic data and energy consumption data, as well as the importance of each energy consumption characteristic data, and to determine the contribution of different types of energy consumption characteristic data to vehicle energy consumption based on the causal relationship between different types of energy consumption characteristic data and energy consumption data, as well as the importance of each energy consumption characteristic data;

[0047] The suggestion generation module is used to determine energy efficiency optimization suggestions based on energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption.

[0048] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the steps in any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction.

[0049] In a fourth aspect, the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction.

[0050] In a fifth aspect, the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps in any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction.

[0051] The above-mentioned energy efficiency optimization method, device, computer equipment, computer-readable storage medium and computer program product based on vehicle energy consumption prediction, first of all, take into account the multi-dimensional factors affecting vehicle energy consumption, obtain multi-source driving data of the vehicle in real time, determine energy consumption characteristic data for vehicle energy consumption prediction based on the multi-source driving data, input the extracted energy consumption characteristic data into a pre-trained energy consumption prediction model, predict the energy consumption of the vehicle in a preset time period, and obtain energy consumption prediction results, thereby improving the accuracy of vehicle energy consumption prediction. Secondly, determine the importance of each energy consumption characteristic data and the causal relationship between different energy consumption characteristic data and energy consumption data, so as to quantify the contribution of each energy consumption influencing factor to energy consumption, and then combine the energy consumption prediction results and the quantified contribution of different energy consumption characteristic data to energy consumption data to determine energy efficiency optimization suggestions, which is conducive to providing accurate data support for driving optimization, charging management and fleet operations, reducing unnecessary energy consumption, extending battery life, reducing operating costs and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 FIG. 1 is an application environment diagram of an energy efficiency optimization method based on vehicle energy consumption prediction in one embodiment;

[0054] Figure 2 1 is a flow chart of an energy efficiency optimization method based on vehicle energy consumption prediction in one embodiment;

[0055] Figure 3 1 is a flow chart of an energy efficiency optimization method based on vehicle energy consumption prediction in another embodiment;

[0056] Figure 4 is a structural block diagram of an energy efficiency optimization device based on vehicle energy consumption prediction in one embodiment;

[0057] Figure 5 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0059] The energy efficiency optimization method based on vehicle energy consumption prediction provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown, the vehicle terminal 102 communicates with the server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104 or placed on the cloud or other network servers.

[0060] Specifically, the vehicle terminal 102 may upload the multi-source driving data collected in real time to the server 104, and the server 104 obtains the multi-source driving data. Secondly, based on the multi-source driving data, multiple types of energy consumption characteristic data are determined, and the energy consumption characteristic data are input into the trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period to obtain an energy consumption prediction result. The energy consumption prediction model is trained based on the energy consumption characteristic historical data carrying energy consumption labels. Then, the causal relationship between different types of energy consumption characteristic data and the energy consumption data, as well as the importance of each energy consumption characteristic data, are determined. According to the causal relationship between different types of energy consumption characteristic data and the energy consumption data, as well as the importance of each energy consumption characteristic data, the contribution of different types of energy consumption characteristic data to vehicle energy consumption is determined. Finally, based on the energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption, energy efficiency optimization suggestions are determined.

[0061] The in-vehicle terminal 102 may be, but is not limited to, an IoT device or a portable wearable device. IoT devices may include smart in-vehicle devices, intelligent central control systems, and in-vehicle locators. Portable wearable devices may include smart watches, smart bracelets, head-mounted devices, and the like. Head-mounted devices may include virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, and the like. A vehicle is a vehicle powered by an onboard power source, including but not limited to pure electric, hybrid, or extended-range vehicles. The server 104 may be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services.

[0062] In an exemplary embodiment, Figure 2 As shown in the figure, an energy efficiency optimization method based on vehicle energy consumption prediction is provided. Figure 1 The server 104 in FIG. 1 is used as an example to illustrate the method, which includes the following steps (hereinafter referred to as S) S100 to S600. Among them:

[0063] S100 collects multi-source driving data of the vehicle in real time.

[0064] Multi-source driving data can include data related to vehicle energy consumption, including but not limited to driving speed, driving path data, driving environment, and vehicle hardware attributes. Driving speed can be acquired through onboard sensors. Driving path data can be acquired through data interfaces provided by navigation systems or third-party mapping services. Driving environment data can be acquired through weather data services, vehicle-to-everything (V2X) technology, or external sensors. Vehicle hardware attributes can be acquired through vehicle diagnostic interfaces.

[0065] In actual applications, the multi-source driving data of the vehicle is collected in real time through the on-board terminal. The on-board terminal may be equipped with a GPS (Global Positioning System) sensor, an IMU (Inertial Measurement Unit) sensor, a CAN (Controller Area Network) bus interface, etc., to collect multi-source driving data in real time and upload it to the server through a wireless communication module. In other embodiments, after collecting the multi-source driving data, the on-board terminal pre-processes the multi-source driving data. The pre-processing includes but is not limited to at least one of removing outliers, filling missing values, and data standardization. Among them, removing outliers can be performing statistical analysis (such as Z-Score) on the multi-source driving data to remove outliers that are beyond a reasonable range. Filling missing values can be filling missing data using interpolation methods (such as linear interpolation, K-nearest neighbor interpolation). Since different data sources have different dimensions, the data needs to be standardized to a unified scale, so data standardization can be standardizing each feature to the interval [0, 1]. The formula is as follows:

[0066]

[0067] in, is the original data, min( ) and max( ) are the minimum and maximum values of the multi-source driving data respectively.

[0068] S200 , determining multiple types of energy consumption characteristic data based on multi-source driving data.

[0069] Among them, the energy consumption characteristic data is used to predict the energy consumption data of the vehicle, including but not limited to average acceleration, road speed limit information, outside temperature and battery capacity. The average acceleration can be calculated based on the driving speed. The road speed limit information may include the speed limit of the road, which can be obtained by combining the GPS position in the driving path data with the speed limit of the road where the vehicle is located provided by a third-party map service. The outside temperature can be determined based on the temperature in the driving environment data. The battery capacity can be determined based on the initial battery specifications and the current available capacity in the vehicle hardware attributes. It can be understood that the energy consumption characteristic data is not limited to this, and is determined according to the type of multi-source driving data obtained, so that for different types of multi-source driving data, the energy consumption characteristic data corresponding to the driving data type is extracted.

[0070] S300, inputting energy consumption feature data into a trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period to obtain an energy consumption prediction result. The energy consumption prediction model is trained based on historical energy consumption feature data carrying energy consumption tags.

[0071] The energy consumption prediction result includes the predicted energy consumption value of the vehicle within a preset time, specifically including the energy consumption value at different time points.

[0072] In practical applications, an initial energy consumption prediction model can be constructed in advance based on supervised learning models such as support vector machines, random forests or deep neural networks. For example, the initial energy consumption prediction model is constructed based on support vector machines, and the energy consumption feature historical data extracted from the historical time period is obtained. The sample data with multiple types of energy consumption feature historical data as multi-dimensional input features is labeled with energy consumption labels, that is, the real historical energy consumption values are labeled, and the labeled sample data is divided into a training sample set and a test sample set. The initial energy consumption prediction model is trained based on the training sample set, and the input feature is set to , the output is the energy consumption value Y, then the goal of the support vector machine is to learn the optimal decision function through the following optimization problem (determine the optimal segmentation hyperplane of the support vector):

[0073] subject to +b) ,

[0074] Among them, x i is the feature vector of the i-th training sample, y i is the corresponding energy consumption label, w and b are the parameters of the support vector machine, is a slack variable.

[0075] The initial energy consumption prediction model is iteratively trained based on the training sample data until a preset training termination condition is met, thereby obtaining a trained energy consumption prediction model. The preset training termination condition may be that the loss function value is less than a preset loss threshold for a predetermined number of consecutive times. Based on the test sample set, the model prediction effect may be evaluated using metrics such as root mean square error, average root mean error, mean absolute error, and root mean square logarithmic error. For example, in this embodiment, cross-validation is used to evaluate model performance, using metrics such as mean square error (MSE) for performance evaluation.

[0076] MSE=

[0077] in, is the actual energy consumption value, is the predicted value and N is the number of samples.

[0078] During specific implementation, the server inputs the energy consumption characteristic data into the trained energy consumption prediction model, predicts the energy consumption of the vehicle within a preset time period, and obtains the energy consumption prediction result.

[0079] S400: Determine the causal relationship between different types of energy consumption characteristic data and energy consumption data, and the importance of each energy consumption characteristic data.

[0080] The causal relationship between energy consumption characteristic data and energy consumption data indicates whether changes in the energy consumption characteristic data lead to changes in the energy consumption data. For example, if average acceleration has a direct impact on energy consumption, then there is a causal relationship between average acceleration and energy consumption data. The causal relationship can be analyzed using causal inference algorithms or correlation analysis algorithms, such as the Pearson correlation coefficient or the Spearman rank correlation coefficient. Causal inference algorithms include, but are not limited to, Granger causal analysis and Do-Calculus.

[0081] In practical applications, the correlation between each energy consumption characteristic data and the energy consumption data can be calculated in advance based on the Pearson correlation coefficient to obtain the correlation analysis results. A causal graph can be constructed based on the correlation analysis results based on the PC algorithm. The server inputs the energy consumption characteristic data and historical energy consumption data and outputs a directed acyclic graph. The directed acyclic graph includes the causal relationship between different types of energy consumption characteristic data and energy consumption data. The causal relationship between different energy consumption characteristic data and energy consumption data is tested through a causal inference algorithm. Based on multiple linear regression or other regression models, the impact of each energy consumption characteristic data on energy consumption is evaluated to obtain the regression coefficient. The importance of each energy consumption characteristic data to the energy consumption data is determined based on the regression coefficient.

[0082] S500 , determining the contribution of different types of energy consumption characteristic data to vehicle energy consumption according to the causal relationship between different types of energy consumption characteristic data and energy consumption data, and the importance of each type of energy consumption characteristic data.

[0083] In practical applications, in order to decouple the impact of different energy consumption characteristic data on vehicle energy consumption, the processor can screen out energy consumption characteristic data that has a direct causal relationship with the energy consumption data from the causal relationship between different types of energy consumption characteristic data and the energy consumption data. According to the importance of the screened energy consumption characteristic data, the importance score of the energy consumption characteristic can be calculated through models such as linear regression, random forest, and gradient boosting decision tree, and the importance score can be standardized to obtain the contribution of the energy consumption characteristic data to vehicle energy consumption.

[0084] S600: Determine energy efficiency optimization suggestions based on the energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption.

[0085] Among them, energy efficiency optimization suggestions include optimization suggestions for different types of energy efficiency characteristic data. For example, optimization suggestions for energy efficiency characteristic data for driving speed may be driving optimization suggestions and cruise control mode suggestions; optimization suggestions for energy efficiency characteristic data for driving path data may be low-energy path planning suggestions, etc. It may be that corresponding energy efficiency optimization suggestions are set in advance for different types of energy efficiency characteristic data, or energy efficiency optimization suggestion generation strategies are provided. For example, driving optimization suggestions include reducing sudden acceleration and smooth braking to improve energy efficiency. Cruise control mode suggestions may include smooth acceleration and intelligent deceleration, which are used to feed back to the vehicle's adaptive cruise control system so that the cruise control system dynamically adjusts the acceleration and braking modes to reduce energy consumption. Low-energy path planning suggestions include recommended low-energy paths. The recommended low-energy paths may be obtained by calculating the most energy-efficient driving route based on the acquired real-time traffic information, and are used to reduce unnecessary energy consumption.

[0086] In actual applications, after the server obtains the energy consumption prediction results and the contribution of different energy consumption characteristic data, it can determine the energy consumption level based on the energy consumption prediction results and multiple preset energy consumption thresholds. When the energy consumption level is higher than the preset energy consumption level threshold, it indicates that the current energy consumption is high and there is room for adjustment. Therefore, based on the contribution of different energy consumption characteristic data, the impact of different energy consumption characteristic data on energy consumption is determined. From the contribution of different energy consumption characteristic data, target energy consumption characteristic data with a contribution higher than the preset contribution threshold is screened out, and energy efficiency optimization suggestions corresponding to the target energy consumption characteristic data are obtained, so as to provide targeted energy efficiency optimization suggestions for factors with a greater impact on energy consumption. The energy efficiency optimization suggestions are fed back to the vehicle terminal so that the vehicle terminal can push the energy efficiency optimization suggestions.

[0087] The multiple preset energy consumption thresholds may include energy consumption thresholds corresponding to high, medium, and low energy consumption levels under different operating conditions. The preset energy consumption level threshold is the medium energy consumption level. Different operating conditions may include urban road driving conditions, suburban road driving conditions, and highway driving conditions. It is understood that different operating conditions have different energy consumption levels and thus different corresponding energy consumption level assessments.

[0088] For example, if the energy consumption level is determined to be high based on the energy consumption prediction results, and is higher than a preset energy consumption level threshold, energy efficiency optimization suggestions corresponding to energy consumption characteristic data having a contribution higher than the preset contribution threshold are obtained, and the energy efficiency optimization suggestions are fed back to the vehicle terminal so that the vehicle terminal pushes the energy efficiency optimization suggestions. Methods for pushing energy efficiency optimization suggestions may include displaying the energy efficiency optimization suggestions on a display, providing prompts by emitting specific sounds or vibration patterns, providing visual prompts by flashing indicator lights, etc. It is understood that the method for pushing energy efficiency optimization suggestions may be any one of the aforementioned methods or a combination of any multiple methods.

[0089] In the above energy efficiency optimization method based on vehicle energy consumption prediction, first, the multi-dimensional factors affecting vehicle energy consumption are taken into account. By acquiring multi-source driving data of the vehicle in real time, energy consumption characteristic data for vehicle energy consumption prediction is determined based on the multi-source driving data, and the extracted energy consumption characteristic data is input into a pre-trained energy consumption prediction model to predict the energy consumption of the vehicle in a preset time period to obtain energy consumption prediction results, thereby improving the accuracy of vehicle energy consumption prediction. Secondly, the importance of each energy consumption characteristic data and the causal relationship between different energy consumption characteristic data and energy consumption data are determined, so as to quantify the contribution of each energy consumption influencing factor to energy consumption. Then, the energy consumption prediction results and the quantified contribution of different energy consumption characteristic data to energy consumption data are combined to determine energy efficiency optimization suggestions, which is conducive to providing accurate data support for driving optimization, charging management and fleet operations, reducing unnecessary energy consumption, extending battery life, reducing operating costs, and improving user experience.

[0090] In an exemplary embodiment, the multi-source driving data includes at least two of driving behavior data, road condition data, environmental condition data, and vehicle configuration data; and determining multiple types of energy consumption characteristic data based on the multi-source driving data includes at least two of the following methods:

[0091] Based on the driving behavior data, driving behavior characteristic data is determined.

[0092] Driving behavior data may include, but is not limited to, acceleration data, braking data, throttle opening data, steering angle data, and the like, and may be used to identify the driver's driving style. Driving behavior characteristic data may include, but is not limited to, average acceleration, maximum acceleration, acceleration change rate, average deceleration, braking frequency, average throttle opening, throttle opening change rate, and steering frequency. Average acceleration, maximum acceleration, and acceleration change rate may be derived from acceleration data to identify the driver's acceleration behavior. Average deceleration and braking frequency may be derived from braking data. Average throttle opening and throttle opening change rate may be determined based on throttle opening data. Steering frequency may be determined based on steering angle data.

[0093] Based on the road condition data, road condition characteristic data is determined.

[0094] Road condition data may include, but is not limited to, road surface material and condition, traffic flow information, and more. Road condition feature data may include road surface condition type, traffic density, speed limits, and traffic signal distribution. Road surface condition types may include asphalt, cement, gravel, or slippery surfaces. Different road surface materials and conditions affect tire rolling resistance, thereby affecting energy consumption. This data can be obtained by identifying road surface material and condition using machine vision algorithms. Traffic density can be obtained from navigation systems and traffic flow information. Speed limits and traffic signal distribution can be obtained from traffic flow information.

[0095] Based on the environmental condition data, environmental condition characteristic data is determined.

[0096] The environmental condition data includes, but is not limited to, the temperature, humidity, and wind speed outside the vehicle, etc. The environmental condition characteristic data includes, but is not limited to, the average temperature, average humidity, and maximum wind speed, etc.

[0097] Based on the vehicle configuration data, vehicle configuration characteristic data is determined.

[0098] Vehicle configuration data includes vehicle hardware parameters such as drive mode, electronic device power, tire type, battery capacity, and motor power. Vehicle configuration feature data includes, but is not limited to, battery temperature and drivetrain efficiency. Drivetrain efficiency can be determined based on motor power.

[0099] In this embodiment, energy consumption characteristic data is extracted from four dimensions, namely, driving behavior, road conditions, environmental conditions, and vehicle position, which is conducive to improving the accuracy of energy consumption prediction.

[0100] In an exemplary embodiment, the driving behavior data includes vehicle speed sequence data. Based on the driving behavior data, the driving behavior characteristic data is determined, including:

[0101] Based on the vehicle speed series data, average acceleration data and average braking intensity are determined.

[0102] In practical applications, suppose the vehicle speed data sequence is V={v1,v2,…,v n}.

[0103] Among them, v i represents the vehicle speed (in meters per second) at the i-th time step (usually in seconds).

[0104] Determine the acceleration a i The finite difference method can be used:

[0105]

[0106] in:

[0107] v i+1 and v i are the velocities of adjacent time steps (unit: m / s).

[0108] Δ t is the sampling time interval (unit: s), which is usually assumed to be 1s. It is understandable that in other implementations, the sampling time interval is set according to the actual prediction accuracy requirement.

[0109] Average acceleration is used to describe the overall acceleration level of the vehicle, and only the data of the acceleration phase (i.e. a i>0). Calculation formula:

[0110]

[0111] Where: A is the set of all moments when the acceleration is greater than zero, that is, A= . is the number of acceleration moments.

[0112] The average braking intensity describes the intensity of the braking process, and only the data of the deceleration phase (i.e. ). Calculation formula:

[0113]

[0114] Where: B is the set of all braking moments, that is, B = . is the number of braking moments. Take the absolute value to ensure a positive representation of the braking intensity.

[0115] In actual applications, the server determines the average acceleration data and the average braking intensity according to the above formula.

[0116] Based on the vehicle speed sequence data, the acceleration frequency and braking frequency are determined.

[0117] The acceleration frequency and braking frequency represent the number of accelerations and brakings, respectively, and are determined based on the speed change during the sampling time interval.

[0118] The driving style of the driving subject of the vehicle is determined based on the acceleration frequency and the braking frequency, and the driving behavior characteristic data includes average acceleration data, average braking intensity and driving style.

[0119] Driving style is determined based on the acceleration / braking ratio and pre-set driving style evaluation criteria. The acceleration / braking ratio is the ratio of acceleration frequency to braking frequency and is determined by the following formula:

[0120]

[0121] in: Indicates the number of accelerations (i.e. number of moments). Indicates the number of braking times (i.e. number of moments).

[0122] Pre-set driving style assessment criteria could be:

[0123] like , indicating that the acceleration frequency is greater than the braking frequency, and the driving style may be aggressive.

[0124] like , indicating that the acceleration and braking frequencies are close and the driving style can be smooth.

[0125] like , indicating that the braking frequency is greater than the acceleration frequency, which may indicate that the driver brakes frequently, affecting energy consumption, and the driving style may be conservative.

[0126] In this embodiment, analyzing the driving style of the driver through driving behavior characteristic data is beneficial to improving the accuracy of energy consumption prediction.

[0127] In an exemplary embodiment, the road condition data includes road surface attribute data. Based on the road condition data, determining the road condition characteristic data includes:

[0128] Based on the pavement attribute data, the average pavement slope and pavement type are determined.

[0129] Road condition characteristic data include average road slope and road type.

[0130] The road surface attribute data may include road surface slope, GPS location, etc. The road surface type may include urban roads and highways, etc. The average slope may be determined based on the road surface slope using the following formula:

[0131]

[0132] in, is the slope of the i-th road segment, and L is the total number of road segments. The road type can be matched based on the vehicle's GPS location and a map API (such as Google Maps or Here Maps), with highways being 1 and urban roads being 0.

[0133] In this embodiment, extracting road condition feature data is helpful in helping the model understand the impact of road conditions on vehicle energy consumption, thereby helping to improve the accuracy of energy consumption prediction.

[0134] In an exemplary embodiment, the vehicle configuration data includes battery attribute data and motor power. Based on the vehicle configuration data, determining the vehicle configuration feature data includes:

[0135] The battery health of the vehicle is evaluated based on the battery attribute data and the motor power, and the vehicle configuration characteristic data includes the battery health.

[0136] Battery attribute data may include battery capacity, number of charge and discharge cycles, voltage level, and internal resistance. Battery capacity may include initial capacity specifications and current available capacity. Determining the vehicle's battery health may involve a server inputting the battery attribute data and motor power into a trained battery health prediction model to predict the vehicle's battery health. Alternatively, scoring criteria may be set for different indicator data, and different indicator data scores may be determined based on the corresponding scoring criteria. The battery health is then determined using a comprehensive weighted approach. Exemplary indicator data includes battery capacity (the ratio of initial battery capacity specifications to current available capacity), number of charge and discharge cycles, internal resistance change, and voltage fluctuation. Scoring criteria for battery capacity may include: a battery capacity greater than 90% of the initial value receives a score of 100; a battery capacity between 80% and 90% of the initial value receives a score of 80; and a battery capacity less than 70% of the initial value receives a score of less than 80. Evaluation criteria for the number of charge and discharge cycles may include: a battery with fewer than 500 charge and discharge cycles receives a score of 100; and a battery with between 500 and 1000 cycles receives a score of 90. If the number of cycles reaches 1000 to 1500, the score is 80 points; if the number of cycles exceeds 1500, the score range is below 80 points.

[0137] In this embodiment, evaluating the health of the battery is helpful to improve the accuracy of energy consumption prediction.

[0138] In an exemplary embodiment, the environmental condition data includes temperature data and humidity data. Based on the environmental condition data, determining the environmental condition characteristic data includes:

[0139] An average temperature and an average humidity are determined based on the temperature data and the humidity data, respectively, the environmental condition characteristic data including the average temperature and the average humidity.

[0140] In actual applications, the server determines the average temperature based on the temperature data outside the vehicle, and determines the average humidity based on the humidity data outside the vehicle.

[0141] In this embodiment, considering the influence of external environmental conditions on the working efficiency and energy consumption of the battery, determining characteristic data of environmental conditions as input of the model is helpful to improve the accuracy of energy consumption prediction.

[0142] To decouple the impact of different energy consumption characteristic data on energy consumption, in an exemplary embodiment, determining the importance of each energy consumption characteristic data includes S420 to S440:

[0143] S420 , for each energy consumption characteristic data, determining a change in the target energy consumption prediction result after removing the energy consumption characteristic data compared with the energy consumption prediction result.

[0144] S440: Determine the importance of the energy consumption characteristic data according to the change amount.

[0145] In practical applications, an importance evaluation model can be established in advance based on random forests. For each energy consumption feature data, based on the trained energy consumption prediction model, the target energy consumption prediction result after removing the energy consumption feature data is determined, and the change in the energy consumption prediction value is determined by comparing the pre-predicted energy consumption prediction result. Suppose the output of the importance evaluation model is , energy consumption characteristic data The importance of energy consumption is quantified by the following formula:

[0146] Importance )

[0147] in, It is feature removal The change in energy consumption prediction results.

[0148] In this embodiment, quantifying the importance of energy consumption characteristic data to energy consumption is helpful to improving the accuracy of the decoupling result, thereby improving the effectiveness of determining energy efficiency optimization suggestions.

[0149] In an exemplary embodiment, Figure 3 As shown, determining the causal relationship between different types of energy consumption characteristic data and energy consumption data includes S462 to S466:

[0150] S462, obtaining historical energy consumption data of the vehicle.

[0151] S462: For each type of energy consumption characteristic data, the influence intensity of the energy consumption characteristic data on the energy consumption data is obtained based on the energy consumption characteristic data, the historical energy consumption data and the established causal relationship analysis model.

[0152] S462: Determine the causal relationship between the energy consumption characteristic data and the energy consumption data based on the impact intensity.

[0153] In practical applications, the importance of energy consumption characteristic data is considered only by measuring its correlation with energy consumption. Causality is then used to determine whether the energy consumption characteristic data actually affects energy consumption data. In this embodiment, a causal relationship analysis model is pre-built based on Granger causality analysis to determine whether energy consumption characteristic data Xi affects energy consumption Y.

[0154] Obtain the historical energy consumption data of the vehicle through the vehicle terminal, assuming that the energy consumption Yt is composed of the past energy consumption value (historical energy consumption data) Y t−k and energy consumption characteristics X i Historical value X i,t−k Jointly determined, the regression model is:

[0155]

[0156] in, Represents the target variable (explained variable) at the current time t, such as the energy consumption of a vehicle.

[0157] Represents a constant term, which indicates the baseline energy consumption level.

[0158] K represents the maximum time lag order of backtracking, that is, the impact of the data of how many steps in the past have on the current value.

[0159] Indicates the influence coefficient of the past target variable Y on the current Yt, representing the autoregressive part.

[0160] Y t−k Represents the target variable value before time k, that is, the historical energy consumption data.

[0161] It represents the coefficient of influence of explanatory variable X on Y.

[0162] X i,t−k Represents the explanatory variable value before time k, that is, energy consumption characteristic data, such as vehicle speed and ambient temperature.

[0163] represents the error term (noise), represents the random disturbance, and is usually assumed to obey the zero-mean normal distribution.

[0164] Through the above model, determine the influence coefficient (influence intensity) of energy consumption characteristic data on energy consumption data, and conduct hypothesis test on the influence coefficient to determine the significance test P-value (P-value), and judge whether the influence coefficient is statistically different from the prior information (such as P-value is less than 0.05). If the influence coefficient If X≠0 and the p-value is less than the preset significance level (such as 0.05), it is determined that X Granger causally affects Y.

[0165] In this embodiment, by determining the causal relationship between different energy consumption characteristic data and energy consumption data, it is helpful to determine the direct and indirect contributions of different energy consumption characteristic data to energy consumption.

[0166] In other embodiments, the energy efficiency optimization suggestions determined based on the energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption may also include driving style optimization suggestions, intelligent cruise control suggestions, and energy recovery strategies for the vehicle energy consumption management system.

[0167] Among them, the driving style optimization suggestions, intelligent cruise control suggestions and energy recovery strategies are used to feed back to the on-board energy consumption management system, so that the on-board energy consumption management system adjusts the vehicle's management strategy according to the energy efficiency optimization suggestions and reduces energy consumption. Specifically, the driving optimization suggestions can be based on the energy consumption prediction results and the contribution analysis of the energy consumption characteristic data to provide the driver with personalized optimization suggestions and cruise control mode suggestions. Here, reference is made to the determination method of the driving optimization suggestions and cruise control mode suggestions in the above embodiment, which will not be repeated here. The intelligent energy recovery strategy can include optimization suggestions under different driving modes. For example, in the driving mode of urban roads, the energy recovery intensity is increased; on highways, the energy recovery intensity is appropriately reduced to maintain a stable driving speed and reduce the additional energy consumption caused by frequent deceleration. The driving energy recovery strategy can pre-set energy efficiency optimization suggestions for road condition characteristic data, so that when the contribution of the road condition characteristic data is higher than the preset contribution threshold, the energy efficiency optimization suggestions for the road condition characteristic data are determined.

[0168] Energy efficiency optimization suggestions can also be generated based on energy consumption prediction results, combined with intelligent transportation and vehicle-to-everything (V2X) technology. For example, energy efficiency optimization suggestions also include traffic signal optimization suggestions, low-energy consumption route planning, and vehicle-road collaborative energy consumption management suggestions.

[0169] Traffic signal optimization recommendations can be based on energy consumption forecasts and road condition information from intelligent transportation and connected vehicle technologies. These recommendations can be used to adjust vehicle speeds based on traffic light timing to optimize energy efficiency. For example, traffic signal optimization recommendations can include determining the status of traffic lights at intersections based on the road's traffic light cycle and adjusting vehicle speeds based on the intersection's signal status to reduce unnecessary stops and starts, especially during peak hours.

[0170] Low-energy route planning recommendations can combine energy consumption prediction models with real-time traffic information from intelligent transportation and connected vehicle technologies to recommend the most energy-efficient driving routes. Specifically, these recommendations can include the lowest-energy routes recommended by the road navigation system based on vehicle type, slope, traffic conditions, and other factors. They can also include multiple alternative routes and their corresponding estimated energy consumption for the driver to choose from.

[0171] Vehicle-road cooperative energy management can optimize electric vehicle energy management by integrating road gradient and traffic flow data from the vehicle-to-everything (V2X) system. For example, real-time road gradient and traffic flow data provided by onboard sensors and roadside equipment can be used to dynamically adjust the vehicle's energy management strategy. For example, in congested roads or areas with complex terrain, vehicles can be notified in advance to adjust their speed or switch to energy-saving mode to reduce energy waste.

[0172] In other embodiments, the server calculates and displays the vehicle's energy consumption per unit distance based on multi-source driving data and energy consumption prediction models, and feeds back the energy consumption per unit distance to the on-board terminal. The on-board terminal pushes the energy consumption per unit distance to the on-board display screen to display the impact of current driving behavior on energy consumption in real time, helping the driver understand the impact of his driving habits on energy consumption.

[0173] In order to more clearly illustrate the energy efficiency optimization method based on vehicle energy consumption prediction provided by this application, a specific embodiment is described below. The specific embodiment includes the following steps:

[0174] S1, real-time collection of multi-source driving data of the vehicle, which includes driving behavior data, road condition data, environmental condition data and vehicle configuration data.

[0175] S2, based on the vehicle speed sequence data, determining the average acceleration data and the average braking intensity, based on the vehicle speed sequence data, determining the acceleration frequency and the braking frequency, and based on the acceleration frequency and the braking frequency, determining the driving style of the driving object of the vehicle.

[0176] S3, determining the average road slope and road type based on the road surface attribute data.

[0177] S4, assesses the battery health of the vehicle based on the battery attribute data and the motor power.

[0178] S5 , determining an average temperature and an average humidity based on the temperature data and the humidity data, respectively.

[0179] S6, inputting the energy consumption feature data into a trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period to obtain an energy consumption prediction result. The energy consumption prediction model is trained based on historical energy consumption feature data carrying energy consumption tags.

[0180] S7, for each energy consumption characteristic data, determining a change in the target energy consumption prediction result after removing the energy consumption characteristic data compared with the energy consumption prediction result, and determining the importance of the energy consumption characteristic data according to the change.

[0181] S8, obtaining the historical energy consumption data of the vehicle, and for each type of energy consumption characteristic data, obtaining the impact intensity of the energy consumption characteristic data on the energy consumption data based on the energy consumption characteristic data, historical energy consumption data and the established causal relationship analysis model, and determining the causal relationship between the energy consumption characteristic data and the energy consumption data based on the impact intensity.

[0182] S9 , determining the contribution of different types of energy consumption characteristic data to vehicle energy consumption according to the causal relationship between different types of energy consumption characteristic data and the energy consumption data, and the importance of each type of energy consumption characteristic data.

[0183] S10, determining energy efficiency optimization suggestions based on the energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption.

[0184] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed 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 portion of steps or stages in other steps.

[0185] In an exemplary embodiment, Figure 5 As shown, an energy efficiency optimization device 600 based on vehicle energy consumption prediction is provided, comprising: a data acquisition module 610, a feature extraction module 620, an energy consumption prediction module 630, an impact analysis module 640 and a suggestion generation module 650, wherein:

[0186] A data acquisition module 610 is used to collect multi-source driving data of the vehicle in real time;

[0187] A feature extraction module 620 is used to determine multiple types of energy consumption feature data based on multi-source driving data;

[0188] Energy consumption prediction module 630, used to input energy consumption feature data into a trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period and obtain an energy consumption prediction result. The energy consumption prediction model is trained based on historical energy consumption feature data carrying energy consumption tags;

[0189] Impact analysis module 640, configured to determine the causal relationship between different types of energy consumption characteristic data and energy consumption data, as well as the importance of each type of energy consumption characteristic data, and determine the contribution of each type of energy consumption characteristic data to vehicle energy consumption based on the causal relationship between the different types of energy consumption characteristic data and the importance of each type of energy consumption characteristic data;

[0190] The suggestion generation module 650 is configured to determine energy efficiency optimization suggestions based on the energy consumption prediction results and the contribution of different types of energy consumption characteristic data to energy consumption.

[0191] In an exemplary embodiment, the feature extraction module 620 is further used to determine driving behavior feature data based on driving behavior data; determine road condition feature data based on road condition data; determine environmental condition feature data based on environmental condition data; and determine vehicle configuration feature data based on vehicle configuration data.

[0192] In an exemplary embodiment, the feature extraction module 620 is also used to determine the average acceleration data and the average braking intensity based on the vehicle speed sequence data; determine the acceleration frequency and the braking frequency based on the vehicle speed sequence data; determine the driving style of the vehicle driver based on the acceleration frequency and the braking frequency; the driving behavior feature data includes the average acceleration data, the average braking intensity and the driving style.

[0193] In an exemplary embodiment, the feature extraction module 620 is further configured to determine an average road slope and a road type based on the road attribute data; the road condition feature data includes the average road slope and the road type.

[0194] In an exemplary embodiment, the feature extraction module 620 is further configured to evaluate the battery health of the vehicle based on the battery attribute data and the motor power; the vehicle configuration feature data includes the battery health.

[0195] In an exemplary embodiment, the feature extraction module 620 is further configured to determine an average temperature and an average humidity based on the temperature data and the humidity data, respectively; the environmental condition feature data includes the average temperature and the average humidity.

[0196] In an exemplary embodiment, the impact analysis module 640 is further configured to determine, for each energy consumption characteristic data, a change in the target energy consumption prediction result compared to the energy consumption prediction result after removing the energy consumption characteristic data;

[0197] Determine the importance of energy consumption characteristic data based on the amount of change.

[0198] In an exemplary embodiment, the data acquisition module 610 is further configured to acquire historical energy consumption data of the vehicle;

[0199] The impact analysis module 640 is also used to obtain the impact intensity of the energy consumption characteristic data on the energy consumption data for each type of energy consumption characteristic data based on the energy consumption characteristic data, historical energy consumption data and the constructed causal relationship analysis model; and determine the causal relationship between the energy consumption characteristic data and the energy consumption data based on the impact intensity.

[0200] Each module in the aforementioned energy efficiency optimization device 600 based on vehicle energy consumption prediction can be implemented in whole or in part through software, hardware, or a combination thereof. Each module 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 the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0201] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. 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 computer program in the non-volatile storage medium. The database of the computer device is used to store data. 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 energy efficiency optimization method based on vehicle energy consumption prediction is implemented.

[0202] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the 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 component arrangement.

[0203] In an exemplary 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 steps in any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction are implemented.

[0204] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction are implemented.

[0205] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned energy efficiency optimization method embodiments based on vehicle energy consumption prediction.

[0206] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0207] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. 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. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory 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), magnetic 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. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0208] The technical features of the above embodiments can be combined arbitrarily. In order 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 application.

[0209] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. An energy efficiency optimization method based on vehicle energy consumption prediction, characterized in that: The method comprises: Real-time collection of multi-source driving data of vehicles; determining multiple types of energy consumption characteristic data based on the multi-source driving data; Inputting the energy consumption characteristic data into a trained energy consumption prediction model to predict the energy consumption of the vehicle within a preset time period to obtain an energy consumption prediction result, wherein the energy consumption prediction model is trained based on historical energy consumption characteristic data carrying energy consumption tags; Determining the causal relationship between different types of energy consumption characteristic data and energy consumption data, and the importance of each of the energy consumption characteristic data; Determining the contribution of different types of energy consumption characteristic data to vehicle energy consumption according to the causal relationship between the different types of energy consumption characteristic data and the energy consumption data, and the importance of each type of energy consumption characteristic data; Based on the energy consumption prediction result and the contribution of the different types of energy consumption characteristic data to energy consumption, energy efficiency optimization suggestions are determined.

2. The method according to claim 1, characterized in that The multi-source driving data includes at least two of driving behavior data, road condition data, environmental condition data, and vehicle configuration data; and determining multiple types of energy consumption characteristic data based on the multi-source driving data includes at least two of the following methods: determining driving behavior characteristic data based on the driving behavior data; Determining road condition characteristic data based on the road condition data; determining environmental condition characteristic data based on the environmental condition data; Based on the vehicle configuration data, vehicle configuration characteristic data is determined.

3. The method according to claim 2, characterized in that The driving behavior data includes vehicle speed sequence data; The determining of driving behavior characteristic data based on the driving behavior data includes: determining average acceleration data and average braking intensity based on the vehicle speed sequence data; determining an acceleration frequency and a braking frequency based on the vehicle speed sequence data; determining a driving style of a subject driving the vehicle based on the acceleration frequency and the braking frequency; The driving behavior characteristic data includes average acceleration data, average braking intensity and driving style.

4. The method according to claim 2, characterized in that The road condition data includes road surface attribute data; The determining of road condition characteristic data based on the road condition data includes: Determining an average road surface slope and a road surface type based on the road surface attribute data; The road condition characteristic data includes the average road slope and road type.

5. The method according to claim 2, characterized in that The vehicle configuration data includes battery attribute data and motor power; The determining of vehicle configuration characteristic data based on the vehicle configuration data includes: evaluating a battery health of the vehicle based on the battery attribute data and the motor power; The vehicle configuration characteristic data includes battery health.

6. The method according to claim 2, characterized in that The environmental condition data includes temperature data and humidity data; The determining of environmental condition characteristic data based on the environmental condition data includes: determining an average temperature and an average humidity based on the temperature data and the humidity data, respectively; The environmental condition characteristic data includes average temperature and average humidity.

7. The method according to any one of claims 1 to 6, characterized in that Determining the importance of each of the energy consumption characteristic data includes: For each energy consumption characteristic data, determining a change in the target energy consumption prediction result after removing the energy consumption characteristic data compared with the energy consumption prediction result; The importance of the energy consumption characteristic data is determined according to the change amount.

8. The method according to claim 7, characterized in that Determine the causal relationship between different types of energy consumption signature data and energy consumption data, including: Obtaining historical energy consumption data of the vehicle; For each type of energy consumption characteristic data, the influence intensity of the energy consumption characteristic data and the energy consumption data is obtained based on the energy consumption characteristic data, the historical energy consumption data and the established causal relationship analysis model; According to the impact intensity, a causal relationship between the energy consumption characteristic data and the energy consumption data is determined.

9. An energy efficiency optimization device based on vehicle energy consumption prediction, characterized in that: The device comprises: Data acquisition module, used to collect multi-source driving data of the vehicle in real time; a feature extraction module, configured to determine multiple types of energy consumption feature data based on the multi-source driving data; An energy consumption prediction module, configured to input the energy consumption characteristic data into a trained energy consumption prediction model, predict the energy consumption of the vehicle within a preset time period, and obtain an energy consumption prediction result, wherein the energy consumption prediction model is trained based on historical energy consumption characteristic data carrying energy consumption tags; an impact analysis module, configured to determine the causal relationship between different types of energy consumption characteristic data and the energy consumption data, as well as the importance of each of the energy consumption characteristic data, and determine the contribution of the different types of energy consumption characteristic data to vehicle energy consumption based on the causal relationship between the different types of energy consumption characteristic data and the energy consumption data, as well as the importance of each of the energy consumption characteristic data; The suggestion generating module is used to determine energy efficiency optimization suggestions based on the energy consumption prediction result and the contribution of the different types of energy consumption characteristic data to energy consumption.

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

Images