Calibration optimization method, device, vehicle and storage medium for vehicle energy management

Abstract

The present application relates to the field of automotive energy management technology, and more particularly to a calibration optimization method, device, vehicle, and storage medium for vehicle energy management. The method comprises: obtaining driving data sets for multiple vehicle models, determining the actual driving data corresponding to each vehicle model to obtain a driving cycle for each vehicle model, determining the optimal power allocation strategy corresponding to each vehicle model's driving cycle based on a preset dynamic programming algorithm, generating a calibrated map for the power allocation strategy, calculating the fuel consumption value of the calibrated map in an energy conservation mode under the driving cycle of each vehicle model, and evaluating the optimal power allocation strategy based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result for the vehicle's energy management. This solves the problems of the energy management methods of related technologies, such as the inability to accurately and quantitatively evaluate the quality of calibration results in actual user driving, the large amount of experimental resources consumed, and the increased workload.

Description

Technical Field

[0001] The present application relates to the technical field of automobile energy management, and in particular to a calibration optimization method, device, vehicle and storage medium for vehicle energy management. Background Art

[0002] In related technologies, the mainstream energy management strategy for mass-produced hybrid vehicles mainly relies on engineers' personal experience, calibration results from previous projects, and desktop calibration to determine the initial version of the MAP.

[0003] However, relying on an engineer's personal engineering experience or drawing on calibration results from past projects will cause calibration results to vary from person to person, and the test + modification cycle requires multiple iterations, which is a huge workload and consumes a lot of experimental resources. The desktop calibration algorithm used requires pre-acquisition of future travel information, and there is a problem that the instantaneous optimal superposition does not equal the global optimal, making it impossible to accurately quantitatively evaluate the pros and cons of the calibration results in the user's actual driving, which urgently needs to be solved. Summary of the Invention

[0004] The present application provides a calibration optimization method, device, vehicle and storage medium for vehicle energy management to solve the problems that the energy management methods in related technologies cannot accurately and quantitatively evaluate the pros and cons of calibration results in actual driving of users, occupy a large amount of test resources and increase workload.

[0005] A first embodiment of the present application provides a calibration optimization method for vehicle energy management, comprising the following steps:

[0006] Acquiring driving data sets of multiple vehicle models, determining actual driving data corresponding to each vehicle model based on the driving data sets, and obtaining a driving cycle for each vehicle model based on the actual driving data corresponding to each vehicle model;

[0007] Determining an optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm, and generating a calibrated MAP of the power allocation strategy using the optimal power allocation strategy;

[0008] The fuel consumption value of the energy conservation mode under the driving cycle of each vehicle type according to the calibrated MAP is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the energy management of the vehicle.

[0009] Optionally, determining actual driving data corresponding to each vehicle type according to the driving data set, and obtaining a driving cycle for each vehicle type according to the actual driving data corresponding to each vehicle type, includes:

[0010] Determining CAN (Controller Area Network) message data of each vehicle type based on the driving data set, and cleaning the CAN message data;

[0011] For each vehicle type, the cleaned CAN message data is segmented to obtain CAN message data for multiple trips, and feature extraction is performed on the CAN message data for each trip to obtain basic statistical data for each trip. The user's driving pattern is then identified based on the basic statistical data.

[0012] For each vehicle type, identifying trips that meet preset similarity conditions among the user's multiple trips, and performing cluster analysis on the trips that meet the preset similarity conditions and the driving patterns using a preset clustering algorithm to obtain the user's typical trips and common driving patterns;

[0013] For each vehicle model, the duration of the driving cycle is determined based on the basic statistical data of the typical trip, and a typical driving cycle is generated in combination with the cluster analysis results. The basic statistical data of the typical driving cycle are compared with the basic statistical data of the original complete data set. When the comparison result meets the preset characterization conditions, the driving cycle of each vehicle model is obtained.

[0014] Optionally, determining the optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm includes:

[0015] Based on the preset dynamic programming algorithm, defining a state space and a dynamic space, and discretizing the state space and the dynamic space respectively;

[0016] The start time and end time of the optimal power allocation strategy are determined, dynamic recursion is performed based on a preset transfer function, a preset objective function and preset constraints, and the optimal execution action at each time and state is determined based on a preset equation to obtain the optimal power allocation strategy corresponding to the driving cycle of each vehicle type.

[0017] Optionally, the generating a calibration MAP of the power allocation strategy by using the optimal power allocation strategy includes:

[0018] Acquire a first coordinate value and a second coordinate value of the calibration MAP, and determine a scale and coordinate points of the calibration MAP based on the first coordinate value and the second coordinate value;

[0019] setting an error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP, and identifying, in the optimal power allocation strategy, a strategy point that falls within the bandwidth corresponding to each coordinate point in the calibration MAP based on the first coordinate quantity and the second coordinate quantity;

[0020] The calibration amount of the coordinate point is determined based on the strategic point, and the calibration MAP is obtained according to the calibration amount of the coordinate point.

[0021] Optionally, determining a calibration amount of a coordinate point based on the strategic point includes:

[0022] Traversing the strategic points within the bandwidth and determining whether there is a strategic point within the bandwidth;

[0023] If the strategic point does not exist within the bandwidth, determining the calibration amount of the coordinate point based on a preset interpolation method; otherwise, determining whether a single strategic point exists within the bandwidth;

[0024] If the single strategy point exists within the bandwidth, the strategy point is used as the calibration quantity of the coordinate point; otherwise, multiple strategy points are grouped, a representative strategy point of each group is determined, and a weight coefficient is assigned to each group. At the same time, a weighted average is performed on the representative strategy points of each group, and the weighted average result is used as the calibration quantity of the coordinate point.

[0025] Optionally, after determining the calibration amount of the coordinate point based on the preset interpolation method, the method further includes:

[0026] The step of setting the error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP is adjusted and executed until the strategic point is within the bandwidth.

[0027] According to the calibration optimization method for vehicle energy management of the embodiment of the present application, driving data sets of multiple vehicle models are obtained, and then the actual driving data corresponding to each vehicle model is determined to obtain the driving cycle of each vehicle model. The optimal power allocation strategy corresponding to the driving cycle of each vehicle model is determined based on a preset dynamic programming algorithm, and a calibration map of the power allocation strategy is generated. The fuel consumption value of the calibration map in the energy conservation mode under the driving cycle of each vehicle model is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the vehicle's energy management. This solves the problems of the energy management methods in the related art that are unable to accurately and quantitatively evaluate the quality of the calibration results in the actual driving of the user, occupy a large amount of experimental resources, and increase the workload. The rule-based energy management architecture utilizes global optimization theory and the user's actual driving data to enable quantitative evaluation of energy management, thereby improving the comprehensive evaluation effect of energy management calibration.

[0028] A second embodiment of the present application provides a calibration optimization device for vehicle energy management, including:

[0029] an acquisition module, configured to acquire driving data sets of multiple vehicle models, determine actual driving data corresponding to each vehicle model based on the driving data sets, and obtain a driving cycle for each vehicle model based on the actual driving data corresponding to each vehicle model;

[0030] a generating module, configured to determine an optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm, and generate a calibrated MAP of the power allocation strategy using the optimal power allocation strategy;

[0031] An evaluation module is configured to calculate the fuel consumption value of the energy conservation mode under the driving cycle of each vehicle type under the calibrated MAP, and evaluate the optimal power allocation strategy based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the energy management of the vehicle.

[0032] Optionally, the acquisition module includes:

[0033] A data cleaning unit, configured to determine CAN message data of each vehicle type based on the driving data set, and clean the CAN message data;

[0034] a data segmentation unit for segmenting the cleaned CAN message data for each vehicle type to obtain CAN message data for multiple trips, performing feature extraction on the CAN message data for each trip to obtain basic statistical data for each trip, and identifying the user's driving pattern based on the basic statistical data;

[0035] a cluster analysis unit, configured to identify, for each vehicle type, trips of the user that meet preset similarity conditions among the multiple trips, and perform cluster analysis on the trips that meet the preset similarity conditions and the driving patterns using a preset clustering algorithm to obtain the user's typical trips and common driving patterns;

[0036] A generating unit is configured to determine, for each vehicle model, a duration of the driving cycle based on basic statistical data of the typical trip, generate a typical driving cycle in combination with a cluster analysis result, compare the basic statistical data of the typical driving cycle with basic statistical data of an original complete data set, and obtain a driving cycle for each vehicle model when the comparison result satisfies a preset characterization condition.

[0037] Optionally, the generating module includes:

[0038] A discrete unit, configured to define a state space and a dynamic space based on the preset dynamic programming algorithm, and discretize the state space and the dynamic space respectively;

[0039] The first determination unit is used to determine the start time and end time of the optimal power allocation strategy, perform dynamic recursion based on a preset transfer function, a preset objective function and preset constraints, and determine the optimal execution action at each time and state based on a preset equation to obtain the optimal power allocation strategy corresponding to the driving cycle of each vehicle type.

[0040] Optionally, the generating module includes:

[0041] a first acquiring unit, configured to acquire a first coordinate value and a second coordinate value of the calibration MAP, and determine a scale and coordinate points of the calibration MAP based on the first coordinate value and the second coordinate value;

[0042] an identification unit, configured to set an error bandwidth of the first coordinate quantity and the second coordinate quantity based on a scale of the calibration MAP, and identify, in the optimal power allocation strategy, a strategy point that falls within the bandwidth corresponding to each coordinate point in the calibration MAP based on the first coordinate quantity and the second coordinate quantity;

[0043] The second acquiring unit is configured to determine a calibration value of a coordinate point based on the strategic point, and obtain the calibration MAP according to the calibration value of the coordinate point.

[0044] Optionally, the second acquiring unit includes:

[0045] A first judgment subunit is configured to traverse the policy points within the bandwidth and determine whether there is a policy point within the bandwidth;

[0046] a second judgment subunit, configured to determine a calibration amount of the coordinate point based on a preset interpolation method if the strategic point does not exist within the bandwidth, and otherwise determine whether a single strategic point exists within the bandwidth;

[0047] The second determination unit is configured to, if the single strategy point exists within the bandwidth, use the strategy point as the calibration quantity of the coordinate point; otherwise, group the multiple strategy points, determine a representative strategy point of each group, assign a weight coefficient to each group, and perform weighted averaging on the representative strategy points of each group, and use the weighted averaging result as the calibration quantity of the coordinate point.

[0048] Optionally, after determining the calibration amount of the coordinate point based on the preset interpolation method, the second judgment subunit further includes:

[0049] The adjustment component is used to adjust and execute the step of setting the error bandwidth of the first coordinate value and the second coordinate value based on the scale of the calibration MAP until the strategic point is within the bandwidth.

[0050] According to the vehicle energy management calibration optimization device of the embodiment of the present application, driving data sets of multiple vehicle models are obtained, and then the actual driving data corresponding to each vehicle model is determined to obtain the driving cycle of each vehicle model. The optimal power allocation strategy corresponding to the driving cycle of each vehicle model is determined based on a preset dynamic programming algorithm, and a calibration map of the power allocation strategy is generated. The fuel consumption value of the calibration map in the energy conservation mode under the driving cycle of each vehicle model is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the vehicle's energy management. This solves the problems of the energy management methods of the related art that cannot accurately and quantitatively evaluate the quality of the calibration results in the actual driving of the user, consumes a large amount of experimental resources, and increases the workload. The rule-based energy management architecture utilizes global optimization theory and the user's actual driving data to enable quantitative evaluation of energy management, thereby improving the comprehensive evaluation effect of energy management calibration.

[0051] The third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the calibration optimization method for vehicle energy management as described in the above embodiment.

[0052] A fourth aspect of the present application provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the calibration optimization method for vehicle energy management as described in the above embodiment.

[0053] The fifth aspect of the present application provides a computer program product, including a computer program, which is executed to implement the calibration optimization method for vehicle energy management described in the above embodiment.

[0054] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0056] Figure 1 A flowchart of a calibration optimization method for vehicle energy management provided according to an embodiment of the present application;

[0057] Figure 2 This is a flow chart of an overall solution according to one embodiment of the present application;

[0058] Figure 3 1 is a block diagram illustrating an exemplary device for calibration and optimization of vehicle energy management according to an embodiment of the present application;

[0059] Figure 4 Schematic diagram of the structure of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0060] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0061] The following describes a calibration optimization method, apparatus, vehicle, and storage medium for vehicle energy management according to an embodiment of the present application with reference to the accompanying drawings. In response to the problems mentioned in the background art above, such as the inability of the energy management methods of the related art to accurately and quantitatively evaluate the quality of the calibration results in actual user driving, the large amount of test resources consumed, and the increased workload, the present application provides a calibration optimization method for vehicle energy management. In this method, a driving data set of multiple vehicle models is obtained, and then the actual driving data corresponding to each vehicle model is determined to obtain a driving cycle for each vehicle model. The optimal power allocation strategy corresponding to the driving cycle of each vehicle model is determined based on a preset dynamic programming algorithm, a calibration map for the power allocation strategy is generated, the fuel consumption value of the calibrated map in the energy conservation mode under the driving cycle of each vehicle model is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the vehicle energy management. Thus, the present application solves the problems mentioned in the background art above, such as the inability of the energy management methods of the related art to accurately and quantitatively evaluate the quality of the calibration results in actual user driving, the large amount of test resources consumed, and the increased workload, by using a rule-based energy management architecture, utilizing global optimization theory and the user's actual driving data to enable quantitative evaluation of energy management, thereby improving the comprehensive evaluation effect of energy management calibration.

[0062] Specifically, before introducing the embodiments of the present application, the mainstream solutions of energy management strategies in related technologies and related problems are first introduced.

[0063] The mainstream energy management strategy for mass-produced hybrid vehicles in the vehicle controller is still rule-based. This means that the rules for the strategy operation and its logical thresholds (calibration quantities) need to be formulated in advance. The most common calibration methods currently include:

[0064] (1) After initially completing the MAP calibration, relying on the engineering experience of relevant technical personnel or drawing on the calibration results of previous projects, problems are found through a large number of real-vehicle tests, and improvements are made by modifying the calibration volume. This method relies too much on personal ability, and the calibration results vary from person to person. The test + modification cycle requires multiple iterations, which is a huge workload and consumes a lot of experimental resources.

[0065] (2) After the initial version of MAP is determined through desktop calibration, it is verified and adjusted with the help of real vehicle testing. Compared with method (1), the calibration workload and experimental test volume of this method are greatly reduced. However, this method still has significant problems. The most common desktop calibration algorithms are to divide the control variables into grids and then traverse the point scanning optimization and ECMS (Equivalent Consumption Minimization Strategy). The former algorithm has the problem of how to compare oil and electric energy. Although the latter solves this problem, it requires obtaining future travel information in advance. In addition, both have the problem that the superposition of instantaneous optimality does not equal the global optimality.

[0066] (3) The current calibration process, whether in the simulation phase or the actual vehicle testing phase, has the problem of "test-oriented development", that is, testing and calibration are performed based on the energy consumption announcement operating cycle, and it is impossible to accurately and quantitatively evaluate the pros and cons of the calibration results in the actual driving of users.

[0067] Therefore, in order to solve the above problems, the embodiments of the present application use users' actual driving big data and global optimization theory to solve the problem of "test development" and the inability to quantitatively evaluate calibration results. The specific implementation methods are as follows.

[0068] Specifically, Figure 1 A flow chart of a calibration optimization method for vehicle energy management provided in an embodiment of the present application.

[0069] like Figure 1 As shown, the calibration optimization method for vehicle energy management includes the following steps:

[0070] In step S101 , driving data sets of multiple vehicle models are acquired, and actual driving data corresponding to each vehicle model is determined based on the driving data sets, and a driving cycle of each vehicle model is obtained based on the actual driving data corresponding to each vehicle model.

[0071] Optionally, actual driving data corresponding to each vehicle model is determined based on the driving data set, and a driving cycle for each vehicle model is obtained based on the actual driving data corresponding to each vehicle model, including: determining CAN message data for each vehicle model based on the driving data set, and cleaning the CAN message data; for each vehicle model, performing data segmentation on the cleaned CAN message data to obtain CAN message data for multiple trips, performing feature extraction on the CAN message data for each trip to obtain basic statistical data for each trip, and identifying the user's driving pattern based on the basic statistical data; for each vehicle model, identifying trips in the user's multiple trips that meet preset similarity conditions, and using a preset clustering algorithm to perform cluster analysis on the trips and driving patterns that meet the preset similarity conditions to obtain the user's typical trips and common driving patterns; for each vehicle model, determining the duration of the driving cycle based on the basic statistical data of the typical trip, generating a typical driving cycle in combination with the clustering analysis results, and comparing the basic statistical data of the typical driving cycle with the basic statistical data of the original complete data set, and obtaining the driving cycle for each vehicle model when the comparison result meets the preset characterization conditions.

[0072] Among them, the preset similarity conditions and the preset characterization conditions can be set by those skilled in the art according to actual test requirements, or can be obtained through computer simulation, and are not specifically limited here.

[0073] Specifically, if Figure 2 As shown, first, a large number of users' driving data sets, that is, the actual driving data of a large number of users, are obtained, and the actual driving data corresponding to each model is determined based on different models. Then, a systematic method is used to convert the actual driving data corresponding to each model into a representative driving cycle for each model.

[0074] Specifically, first, the CAN message data of each vehicle model is obtained through the cloud platform, and the CAN message data is cleaned, which mainly includes signal screening, filtering and smoothing. That is, the vehicle speed and acceleration information are filtered out from the obtained CAN message data, and the data is cleaned by deleting all meaningless erroneous readings or outliers (such as a sudden surge in speed that is physically unreasonable). Then, smoothing techniques (such as moving average or low-pass filter) are applied to reduce the noise and variability in the speed data.

[0075] Secondly, for each vehicle model, the cleaned CAN message data is segmented. The specific segmentation methods mainly include segmentation by trip and segmentation by event. Among them, segmentation by trip mainly divides the CAN message data into a single trip. A trip is defined as the time from when the vehicle starts moving to when the vehicle stops for a considerable period of time. Segmentation by event further divides the data by driving events such as acceleration, deceleration, constant speed and idling.

[0076] Again, the embodiment of the present application performs feature extraction on the segmented CAN message data, which mainly includes statistical analysis and recognition mode. In the statistical analysis, the basic statistical data of each trip are calculated, including the average speed, maximum speed, maximum operating speed, speed variance and speed standard deviation, maximum acceleration, average acceleration in the acceleration section, maximum deceleration, average deceleration in the deceleration section, acceleration standard deviation, acceleration ratio, deceleration ratio, uniform speed ratio, and idle ratio. In the recognition mode, based on the basic statistical data, the driving patterns (i.e., driving behaviors) that frequently appear in the data set are searched.

[0077] Then, based on the feature extraction of CAN message data, it is necessary to perform cluster analysis on it, which mainly includes K-means clustering and hierarchical clustering. Among them, K-means clustering can be understood as using the K-means clustering algorithm to group similar itineraries and driving modes (driving behaviors) of each model for each model, and then statistically analyze the grouping results to identify the typical itineraries and common driving modes (driving behaviors) of different users; for more detailed grouping, hierarchical clustering methods can be used. Hierarchical clustering can be used to determine driving stages at different granularity levels.

[0078] Finally, in an embodiment of the present application, for each vehicle model, the user's typical trip and common driving pattern are standardized. That is, the duration of the driving cycle is determined based on the basic statistical data of the typical trip, while ensuring the manageability of the simulation. Then, a representative typical driving cycle is generated in combination with the cluster analysis results to capture the typical driving cycle reflected in the data. Each cycle should include acceleration, constant speed, deceleration, and idling, as well as changes therebetween, to reflect real-world conditions. The basic statistical data of the typical driving cycle are compared with the basic statistical data of the original complete data set. When the comparison result meets the preset characterization conditions, a driving cycle for each vehicle model is obtained. That is, the comparison result ensures that the cycle can accurately and broadly represent the real driving data set. If the comparison result cannot broadly represent the real driving data set, it can be corrected by redefining the clustering parameters or optimizing the feature value extraction method to ensure that the corrected cycle can accurately and broadly represent the real driving data set.

[0079] It should be noted that, after obtaining the representative driving cycle corresponding to the user data of one vehicle model, the embodiment of the present application can change the vehicle model and repeat the above steps to obtain representative driving cycles corresponding to different vehicle models.

[0080] In step S102 , the optimal power allocation strategy corresponding to the driving cycle of each vehicle type is determined based on a preset dynamic programming algorithm, and a calibrated MAP of the power allocation strategy is generated using the optimal power allocation strategy.

[0081] Optionally, the optimal power allocation strategy corresponding to the driving cycle of each vehicle type is determined based on a preset dynamic programming algorithm, including: defining a state space and a dynamic space based on the preset dynamic programming algorithm, and discretizing the state space and the dynamic space respectively; determining the start time and end time of the optimal power allocation strategy, performing dynamic recursion based on a preset transfer function, a preset objective function and preset constraints, and determining the optimal execution action at each time and state based on a preset equation, so as to obtain the optimal power allocation strategy corresponding to the driving cycle of each vehicle type.

[0082] Optionally, a calibration MAP of a power allocation strategy is generated using an optimal power allocation strategy, including: obtaining a first coordinate quantity and a second coordinate quantity of the calibration MAP, and determining the scale and coordinate points of the calibration MAP based on the first coordinate quantity and the second coordinate quantity; setting an error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP, and identifying a strategy point that falls within the bandwidth corresponding to each coordinate point in the calibration MAP based on the first coordinate quantity and the second coordinate quantity in the optimal power allocation strategy; determining a calibration quantity of the coordinate point based on the strategy point, and obtaining the calibration MAP according to the calibration quantity of the coordinate point.

[0083] Optionally, determining the calibration quantity of the coordinate point based on the strategy point includes: traversing the strategy points within the bandwidth and judging whether there is a strategy point within the bandwidth; if there is no strategy point within the bandwidth, determining the calibration quantity of the coordinate point based on a preset interpolation method, otherwise, judging whether there is a single strategy point within the bandwidth; if there is a single strategy point within the bandwidth, using the strategy point as the calibration quantity of the coordinate point, otherwise, grouping multiple strategy points, determining a representative strategy point of each group, and assigning a weight coefficient to each group, and performing weighted averaging on the representative strategy points of each group, and using the weighted averaging result as the calibration quantity of the coordinate point.

[0084] Among them, the preset dynamic programming algorithm, preset transfer function, preset objective function, preset constraint conditions and preset interpolation method can all be selected by technical personnel in this field according to actual testing requirements and are not specifically limited here.

[0085] Specifically, after obtaining the representative driving cycles of users of each vehicle model, the embodiment of the present application needs to determine the global optimal power allocation strategy for the driving cycles of each vehicle model and common test cycles (such as WLTC (Worldwide Harmonized Light Vehicles Test Procedure, global unified light vehicle test procedure), NEDC (New European Driving Cycle, New European Driving Cycle), etc.) based on a preset dynamic programming algorithm.

[0086] Specifically, if Figure 2As shown, first, based on the preset dynamic programming algorithm, the state space and dynamic space are defined, and the state space and dynamic space are discretized respectively. Among them, the definition of the state space and discretization mainly include (1) state variables: battery state of charge (2) discretization: the state space will be discretized into manageable intervals, because DP (Dynamic Programming) requires a finite state space to be implemented; the definition of the action space and discretization mainly include: (1) action variables: engine power, generator power, (2) discretization: the action space should also be discretized, and the action will involve deciding the power that the engine should generate at each time step.

[0087] Secondly, the start and end time of the optimal power allocation strategy are determined, that is, the time scope of the optimization problem is determined. The optimization problem will be considered within a fixed number of time steps.

[0088] Again, define the preset transfer function and the preset objective function, wherein defining the preset transfer function mainly includes defining how the state variable SOC changes with the engine power (state variable) and the wheel-end power demand (determined by the time step). For example, the SOC (State of Charge) at time t+1 is determined by the SOC, engine power and wheel-end power demand at time t; defining the preset objective function mainly includes minimizing fuel consumption while ensuring battery health and meeting other operating constraints (such as power balance). The preset objective function J can be expressed as a function of the state variable x(t) and the control variable u(t) within the time range T, wherein,

[0089] The preset transfer function can be expressed as:

[0090] SOC(t+1)=SOC(t)+f(P(t))

[0091] The preset objective function can be expressed as:

[0092]

[0093] Wherein, J is the cost that the DP algorithm aims to minimize; t is the discretized time step; x(t) is the state of the system at time t, which is the battery state of charge in the embodiment of the present application; u(t) is the action of the system at time t, which is the engine power and generator power in the embodiment of the present application; f(x(t), u(t)) is the instantaneous cost function at time t, which may include fuel consumption, electricity usage and other operating costs; g(x(t)) is a penalty function used to penalize bad states (i.e., SOC that deviates from the target) to ensure that the SOC at the beginning and end are the same (battery balance); SOC(t) is the SOC value at time t; P(t) is the engine power generated at time t; f(P(t)) is the influence of engine power generated on SOC.

[0094] Furthermore, combining the above formula with the practical application and definition of the embodiments of the present application, we can obtain:

[0095]

[0096] Among them, C fuel (P(t)) is the fuel cost; C battery (SOC(t), P(t)) is the loss (life) cost of the battery; g(SOC(t)) is the penalty function.

[0097] Then, the driving cycle of each vehicle model needs to be constrained based on preset constraints. The main constraints are as follows: (1) Ensure that the output power of the drive motor in each time step can meet the wheel-end demand; (2) Ensure that the sum of the generator's power generation and the drive motor's recovery power in each time step is not greater than the battery's charging capacity; (3) Ensure that the engine's output power in each time step is not greater than the generator's power generation capacity; (4) Ensure that the engine speed does not exceed the set maximum value at different vehicle speeds to avoid NVH (Noise, Vibration, and Harshness) problems; (5) The SOC must be kept within the allowable range to avoid overcharging or over-discharging.

[0098] Finally, dynamic recursion is performed based on the preset transfer function, preset objective function, and preset constraints obtained above, and the optimal execution action at each time and state is determined based on a preset equation (such as the Bellman equation), and the optimal power allocation strategy corresponding to the driving cycle of each vehicle type is obtained. That is, starting from the previous time step at the final moment and moving backward, the Bellman equation is used to calculate the optimal cost and strategy, and the optimal execution action at each time and state is stored for final decision-making. The Bellman equation can be expressed as:

[0099]

[0100] Among them, SOC ′ is the state quantity SOC at the next moment obtained by the transfer function after executing action P(t); V(t+1, SOC ′ ) is the optimal operation at time t+1; C battery (SOC, P(t)) is the battery loss (life) cost; g(SOC) is the penalty function.

[0101] Furthermore, after completing the dynamic recursion, the optimal power allocation strategy corresponding to the driving cycle of each vehicle type is obtained.

[0102] It should be noted that, after obtaining the optimal power allocation strategy corresponding to the driving cycle of a vehicle model, the embodiment of the present application can change the vehicle model and repeat the above steps to obtain the optimal power allocation strategies corresponding to different vehicle models.

[0103] Furthermore, after obtaining the optimal power allocation strategy corresponding to the driving cycle of each vehicle model, the embodiment of the present application uses the optimal power allocation strategy to generate a calibrated MAP in the vehicle controller for determining power allocation, and handles conflicts and missing values ​​through data analysis technology.

[0104] Specifically, if Figure 2 As shown, first, the first coordinate quantity and the second coordinate quantity of the calibration MAP are obtained, and the scale and coordinate points of the calibration MAP are determined based on the first coordinate quantity and the second coordinate quantity. For example, the two coordinate quantities of the calibration MAP (such as vehicle speed and wheel-end power demand) are divided into a grid to determine the scale and coordinate points of the calibration MAP.

[0105] Secondly, based on the scale of the calibration MAP, an error bandwidth of the first coordinate quantity and the second coordinate quantity is set. This bandwidth affects the accuracy and quantity of the optimal power allocation strategy search results. In the optimal power allocation strategy corresponding to different driving cycles, the strategy points that fall within the corresponding bandwidth of each coordinate point in the calibration MAP are identified and recorded based on the first coordinate quantity and the second coordinate quantity. Then, the calibration amount of the coordinate point is determined based on the strategy point, and the calibration MAP is obtained based on the calibration amount of the coordinate point.

[0106] Among them, the determination of the calibration amount of coordinate points can be divided into multiple situations. First, traverse the strategic points within the bandwidth and determine whether there is a strategic point within the bandwidth corresponding to each coordinate point. If there is no strategic point within the bandwidth corresponding to a certain coordinate point, the calibration amount of the coordinate point is determined based on the preset interpolation method.

[0107] Secondly, if a strategy point exists within the bandwidth corresponding to a coordinate point and the strategy point is unique, that is, if a single strategy point exists within the bandwidth corresponding to a coordinate point, then the strategy point is used as the calibration quantity of the coordinate point.

[0108] Thirdly, if a strategy point exists within the bandwidth corresponding to a coordinate point and the strategy point is not unique, that is, if there are multiple strategy points within the bandwidth corresponding to a coordinate point, the strategy points are grouped according to their origin (that is, the global optimal control strategy corresponding to which driving cycle they come from), and the representative strategy point of each group is confirmed. When a strategy point exists within a group and the strategy point is unique, the strategy point is the representative strategy point of the group; when a strategy point exists within a group and the strategy point is not unique, K-means clustering is performed on it, and the centroid of the cluster is used as the representative strategy point of the group. Then, based on the emphasis of the vehicle model on different driving cycles, a weight coefficient is assigned to each group. For example, if a vehicle model attaches importance to the fuel consumption announcement value, a higher weight is given to the WLTC cycle. Then, a weighted average is performed on the representative strategy points of each group. The weighted average result is the calibration value of the coordinate point, and the calibration MAP is obtained based on the calibration value obtained by the above method.

[0109] Optionally, after determining the calibration amount of the coordinate point based on a preset interpolation method, the method further includes adjusting and executing the step of setting the error bandwidth of the first coordinate amount and the second coordinate amount based on the scale of the calibration MAP until a strategic point exists within the bandwidth.

[0110] Specifically, in the embodiment of the present application, if it is found that the number of strategic points in the bandwidth area corresponding to a certain coordinate point is small and there are many areas without strategic points, the process returns to continue executing the above step of adjusting the error bandwidth.

[0111] In step S103 , the fuel consumption value of the calibrated MAP in the energy conservation mode under the driving cycle of each vehicle type is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain the comprehensive energy consumption evaluation result of the vehicle's energy management.

[0112] Specifically, based on the calibrated MAP obtained above, the fuel consumption per 100 kilometers in the energy conservation mode under different test cycles of the calibrated MAP generated in the above steps is further calculated through simulation, and the results of using the DP global optimal strategy are used as the evaluation basis to verify the comprehensive energy consumption performance of the calibration result.

[0113] Specifically, if Figure 2 As shown in the figure, first, the calibrated MAP is placed in the simulation model to calculate its electric balance fuel consumption value under different driving cycles; secondly, the electric balance fuel consumption value under different driving cycles when using the global optimal control strategy is calculated. By comparing the results of the two electric balance fuel consumption values ​​obtained above, the energy consumption performance of the calibrated MAP under different driving cycles can be quantitatively evaluated.

[0114] Therefore, based on the specific discussion of the above embodiment, the following beneficial effects can be obtained:

[0115] (1) The embodiments of the present application are based on the control architecture of current mass-produced models, and innovatively combine global optimal control theory with rule-based strategy calibration. Through data analysis technology, the global optimal control strategy is converted into a rule-based calibration MAP.

[0116] (2) Based on global optimal control theory, the present embodiment provides a method for objectively and quantitatively evaluating calibrated MAP fuel economy without relying on actual testing;

[0117] (3) The embodiment of the present application proposes a systematic method to convert the big data processing of the user's actual driving into a representative driving cycle, which retains the core characteristics of the user's actual driving and avoids the problem that the actual driving is highly random and difficult to repeat.

[0118] According to the calibration optimization method for vehicle energy management of the embodiment of the present application, driving data sets of multiple vehicle models are obtained, and then the actual driving data corresponding to each vehicle model is determined to obtain the driving cycle of each vehicle model. The optimal power allocation strategy corresponding to the driving cycle of each vehicle model is determined based on a preset dynamic programming algorithm, and a calibration map of the power allocation strategy is generated. The fuel consumption value of the calibration map in the energy conservation mode under the driving cycle of each vehicle model is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the vehicle's energy management. This solves the problems of the energy management methods in the related art that are unable to accurately and quantitatively evaluate the quality of the calibration results in the actual driving of the user, occupy a large amount of experimental resources, and increase the workload. The rule-based energy management architecture utilizes global optimization theory and the user's actual driving data to enable quantitative evaluation of energy management, thereby improving the comprehensive evaluation effect of energy management calibration.

[0119] Next, the calibration optimization device for vehicle energy management proposed in accordance with an embodiment of the present application will be described with reference to the accompanying drawings.

[0120] Figure 3 It is a block diagram of a calibration optimization device for vehicle energy management according to an embodiment of the present application.

[0121] like Figure 3 As shown, the calibration optimization device 10 for vehicle energy management includes: an acquisition module 100 , a generation module 200 and an evaluation module 300 .

[0122] The acquisition module 100 is configured to acquire driving data sets of multiple vehicle models, determine actual driving data corresponding to each vehicle model based on the driving data sets, and obtain a driving cycle for each vehicle model based on the actual driving data corresponding to each vehicle model.

[0123] A generation module 200 is configured to determine an optimal power allocation strategy corresponding to a driving cycle of each vehicle type based on a preset dynamic programming algorithm, and to generate a calibrated MAP of the power allocation strategy using the optimal power allocation strategy;

[0124] The evaluation module 300 is used to calculate the fuel consumption value of the calibrated MAP in the energy conservation mode under the driving cycle of each vehicle type, and evaluate the optimal power allocation strategy based on the fuel consumption value to obtain the comprehensive energy consumption evaluation result of the vehicle's energy management.

[0125] Optionally, the acquisition module 100 includes:

[0126] A data cleaning unit is used to determine the CAN message data of each vehicle type based on the driving data set and clean the CAN message data;

[0127] A data segmentation unit is used to segment the cleaned CAN message data for each vehicle type to obtain CAN message data for multiple trips, perform feature extraction on the CAN message data for each trip, obtain basic statistical data for each trip, and identify the user's driving pattern based on the basic statistical data;

[0128] A cluster analysis unit is used to identify, for each vehicle type, trips among multiple trips of a user that meet preset similarity conditions, and to perform cluster analysis on the trips and driving patterns that meet the preset similarity conditions using a preset clustering algorithm to obtain the user's typical trips and common driving patterns;

[0129] A generation unit is used to determine the duration of a driving cycle for each vehicle model based on basic statistical data of a typical trip, generate a typical driving cycle in combination with the cluster analysis result, compare the basic statistical data of the typical driving cycle with the basic statistical data of the original complete data set, and obtain the driving cycle for each vehicle model when the comparison result meets the preset characterization conditions.

[0130] Optionally, the generating module 200 includes:

[0131] Discrete units are used to define the state space and dynamic space based on a preset dynamic programming algorithm, and discretize the state space and dynamic space respectively;

[0132] The first determination unit is used to determine the start time and end time of the optimal power allocation strategy, perform dynamic recursion based on a preset transfer function, a preset objective function and preset constraints, and determine the best execution action at each time and state based on a preset equation to obtain the optimal power allocation strategy corresponding to the driving cycle of each vehicle model.

[0133] Optionally, the generating module 200 includes:

[0134] a first acquiring unit, configured to acquire a first coordinate value and a second coordinate value of the calibration MAP, and determine a scale and coordinate points of the calibration MAP based on the first coordinate value and the second coordinate value;

[0135] an identification unit, configured to set an error bandwidth of the first coordinate quantity and the second coordinate quantity based on a scale of the calibration MAP, and identify, in an optimal power allocation strategy, a strategy point that falls within a bandwidth corresponding to each coordinate point in the calibration MAP based on the first coordinate quantity and the second coordinate quantity;

[0136] The second acquiring unit is configured to determine a calibration quantity of the coordinate point based on the strategic point, and obtain a calibration MAP according to the calibration quantity of the coordinate point.

[0137] Optionally, the second acquiring unit includes:

[0138] The first judgment subunit is used to traverse the policy points in the bandwidth and determine whether there is a policy point in the bandwidth;

[0139] The second judgment subunit is configured to determine the calibration amount of the coordinate point based on a preset interpolation method if no strategic point exists within the bandwidth, and otherwise determine whether a single strategic point exists within the bandwidth;

[0140] The second determination unit is used to use the strategy point as the calibration quantity of the coordinate point if a single strategy point exists within the bandwidth; otherwise, multiple strategy points are grouped, a representative strategy point of each group is determined, and a weight coefficient is assigned to each group. At the same time, a weighted average is performed on the representative strategy points of each group, and the calibration quantity of the coordinate point obtained by the weighted average is obtained.

[0141] Optionally, after determining the calibration amount of the coordinate point based on a preset interpolation method, the second judgment subunit further includes:

[0142] The adjustment subcomponent is used to adjust and execute the step of setting the error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP until a strategic point exists within the bandwidth.

[0143] According to the vehicle energy management calibration optimization device of the embodiment of the present application, driving data sets of multiple vehicle models are obtained, and then the actual driving data corresponding to each vehicle model is determined to obtain the driving cycle of each vehicle model. The optimal power allocation strategy corresponding to the driving cycle of each vehicle model is determined based on a preset dynamic programming algorithm, and a calibration map of the power allocation strategy is generated. The fuel consumption value of the calibration map in the energy conservation mode under the driving cycle of each vehicle model is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the vehicle's energy management. This solves the problems of the energy management methods of the related art that cannot accurately and quantitatively evaluate the quality of the calibration results in the actual driving of the user, consumes a large amount of experimental resources, and increases the workload. The rule-based energy management architecture utilizes global optimization theory and the user's actual driving data to enable quantitative evaluation of energy management, thereby improving the comprehensive evaluation effect of energy management calibration.

[0144] Figure 4 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0145] Memory 401 , processor 402 , and computer programs stored in the memory 401 and executable on the processor 402 .

[0146] When the processor 402 executes the program, the calibration optimization method for vehicle energy management provided in the above embodiment is implemented.

[0147] Furthermore, the vehicle further comprises:

[0148] The communication interface 403 is used for communication between the memory 401 and the processor 402 .

[0149] The memory 401 is used to store computer programs that can be run on the processor 402 .

[0150] The memory 401 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0151] If the memory 401, the processor 402, and the communication interface 403 are implemented independently, the communication interface 403, the memory 401, and the processor 402 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0152] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0153] The processor 402 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0154] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned calibration optimization method for vehicle energy management.

[0155] This embodiment also provides a computer program product, including a computer program, which is executed to implement the calibration optimization method of vehicle energy management of the above embodiment,

[0156] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0157] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature specified as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of this application, "N" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0158] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0159] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or N wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing it in a suitable manner if necessary, and then storing it in a computer memory.

[0160] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiment, the N steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0161] Those skilled in the art will appreciate that all or part of the steps in the method for implementing the above-mentioned embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0162] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0163] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A calibration optimization method for vehicle energy management, characterized in that: The following steps are involved: Acquiring driving data sets of multiple vehicle models, determining actual driving data corresponding to each vehicle model based on the driving data sets, and obtaining a driving cycle for each vehicle model based on the actual driving data corresponding to each vehicle model; Determining an optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm, and generating a calibrated MAP of the power allocation strategy using the optimal power allocation strategy; The fuel consumption value of the energy conservation mode under the driving cycle of each vehicle type according to the calibrated MAP is calculated, and the optimal power allocation strategy is evaluated based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the energy management of the vehicle.

2. The method according to claim 1, characterized in that Determining actual driving data corresponding to each vehicle type according to the driving data set, and obtaining a driving cycle of each vehicle type according to the actual driving data corresponding to each vehicle type, includes: Determining CAN message data of each vehicle type based on the driving data set, and cleaning the CAN message data; For each vehicle type, the cleaned CAN message data is segmented to obtain CAN message data for multiple trips, and feature extraction is performed on the CAN message data for each trip to obtain basic statistical data for each trip. The user's driving pattern is then identified based on the basic statistical data. For each vehicle type, identifying trips that meet preset similarity conditions among the user's multiple trips, and performing cluster analysis on the trips that meet the preset similarity conditions and the driving patterns using a preset clustering algorithm to obtain the user's typical trips and common driving patterns; For each vehicle model, the duration of the driving cycle is determined based on the basic statistical data of the typical trip, and a typical driving cycle is generated in combination with the cluster analysis results. The basic statistical data of the typical driving cycle are compared with the basic statistical data of the original complete data set. When the comparison result meets the preset characterization conditions, the driving cycle of each vehicle model is obtained.

3. The method according to claim 1 or 2, characterized in that The determining of the optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm includes: Based on the preset dynamic programming algorithm, defining a state space and a dynamic space, and discretizing the state space and the dynamic space respectively; The start time and end time of the optimal power allocation strategy are determined, dynamic recursion is performed based on a preset transfer function, a preset objective function and preset constraints, and the optimal execution action at each time and state is determined based on a preset equation to obtain the optimal power allocation strategy corresponding to the driving cycle of each vehicle type.

4. The method according to claim 1, wherein The step of generating a calibration MAP of the power allocation strategy by using the optimal power allocation strategy includes: Acquire a first coordinate value and a second coordinate value of the calibration MAP, and determine a scale and coordinate points of the calibration MAP based on the first coordinate value and the second coordinate value; setting an error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP, and identifying, in the optimal power allocation strategy, a strategy point that falls within the bandwidth corresponding to each coordinate point in the calibration MAP based on the first coordinate quantity and the second coordinate quantity; The calibration amount of the coordinate point is determined based on the strategic point, and the calibration MAP is obtained according to the calibration amount of the coordinate point.

5. The method according to claim 4, characterized in that The step of determining a calibration amount of a coordinate point based on the strategic point includes: Traversing the strategic points within the bandwidth and determining whether there is a strategic point within the bandwidth; If the strategic point does not exist within the bandwidth, determining the calibration amount of the coordinate point based on a preset interpolation method; otherwise, determining whether a single strategic point exists within the bandwidth; If the single strategy point exists within the bandwidth, the strategy point is used as the calibration quantity of the coordinate point; otherwise, multiple strategy points are grouped, a representative strategy point of each group is determined, and a weight coefficient is assigned to each group. At the same time, a weighted average is performed on the representative strategy points of each group, and the weighted average result is used as the calibration quantity of the coordinate point.

6. The method according to claim 5, characterized in that After determining the calibration amount of the coordinate point based on the preset interpolation method, the method further includes: The step of setting the error bandwidth of the first coordinate quantity and the second coordinate quantity based on the scale of the calibration MAP is adjusted and executed until the strategic point is within the bandwidth.

7. A calibration optimization device for vehicle energy management, characterized in that: include: an acquisition module, configured to acquire driving data sets of multiple vehicle models, determine actual driving data corresponding to each vehicle model based on the driving data sets, and obtain a driving cycle for each vehicle model based on the actual driving data corresponding to each vehicle model; a generating module, configured to determine an optimal power allocation strategy corresponding to the driving cycle of each vehicle type based on a preset dynamic programming algorithm, and generate a calibrated MAP of the power allocation strategy using the optimal power allocation strategy; An evaluation module is configured to calculate the fuel consumption value of the energy conservation mode under the driving cycle of each vehicle type under the calibrated MAP, and evaluate the optimal power allocation strategy based on the fuel consumption value to obtain a comprehensive energy consumption evaluation result of the energy management of the vehicle.

8. The device according to claim 7, characterized in that The acquisition module includes: A data cleaning unit, configured to determine CAN message data of each vehicle type based on the driving data set, and clean the CAN message data; a data segmentation unit for segmenting the cleaned CAN message data for each vehicle type to obtain CAN message data for multiple trips, performing feature extraction on the CAN message data for each trip to obtain basic statistical data for each trip, and identifying the user's driving pattern based on the basic statistical data; a cluster analysis unit, configured to identify, for each vehicle type, trips of the user that meet preset similarity conditions among the multiple trips, and perform cluster analysis on the trips that meet the preset similarity conditions and the driving patterns using a preset clustering algorithm to obtain the user's typical trips and common driving patterns; A generating unit is configured to determine, for each vehicle model, a duration of the driving cycle based on basic statistical data of the typical trip, generate a typical driving cycle in combination with a cluster analysis result, compare the basic statistical data of the typical driving cycle with basic statistical data of an original complete data set, and obtain a driving cycle for each vehicle model when the comparison result satisfies a preset characterization condition.

9. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the calibration optimization method for vehicle energy management according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the calibration optimization method for vehicle energy management as described in any one of claims 1 to 6.

Images