A PHEV energy-saving driving control method based on integrated learning considering battery life

Abstract

The application relates to a PHEV energy-saving driving control method based on integrated learning and considering battery life, which comprises the following steps: S1, constructing a comprehensive driving condition data set; S2, obtaining a global optimal comprehensive energy efficiency cost and an optimal control decision data set; S3, discretely extracting an optimal energy consumption cost, a battery life cost and a current data set; S4, calculating a battery SOC according to a vehicle driving state; S5, establishing a comprehensive energy efficiency cost model considering the energy consumption cost and the battery life cost; S6, designing a hierarchical energy-saving driving control architecture based on integrated learning; S7, performing economic vehicle speed optimization based on integrated learning and considering battery life in the upper layer; and S8, performing power system energy management control according to a planned vehicle speed trajectory in the lower layer. A fusion energy efficiency cost model and a system state transition model considering energy consumption and battery life are established by using a neural network, a complex coupling relationship existing between energy consumption characteristics of a hybrid power system and economic vehicle speed is efficiently decoupled, and the service life of the battery is prolonged.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of plug-in hybrid electric vehicles, and particularly relates to a PHEV energy-saving driving control method based on integrated learning and considering battery life. BACKGROUND

[0002] Automotive electrification provides an effective way to reduce the dependence of vehicle driving on fossil fuels, among which electric vehicles (EV) and hybrid electric vehicles (HEV) occupy a dominant position in various solutions. In addition, PHEV has been widely used due to its advantages of both electric vehicles and fuel vehicles. This development trend makes the design of energy management control algorithm more complex, focusing on effective management of power output distribution between internal combustion engine and driving motor. On the other hand, the rapid development of Internet of Vehicles and automation technology enables vehicles to obtain real-time surrounding driving conditions, thereby providing new opportunities for optimization of economic vehicle speed. The joint optimization of vehicle speed and energy management may provide a new opportunity for further improving the energy-saving driving of PHEV.

[0003] PHEV is more complex than electric vehicles and fuel vehicles due to multiple power sources in the power system, which brings new challenges to the energy-saving driving control of the vehicle. When the power system of the controlled vehicle has multiple control degrees of freedom, there is a complex coupling relationship between the energy consumption characteristics of the vehicle power system and the economic vehicle speed. It is crucial to fully consider the influence of the energy consumption characteristics of the power system and the battery life cost when planning the economic vehicle speed to improve the energy consumption economy of the whole vehicle.

[0004] At present, in the research of PHEV, the energy-saving driving of the powertrain mainly includes two methods of step-by-step optimization and collaborative optimization. However, the step-by-step optimization fails to accurately decouple the influence of speed optimization and energy management, and cannot obtain the optimal driving energy consumption. In contrast, the collaborative optimization overcomes this shortcoming. For example, the intelligent networked plug-in hybrid electric vehicle collaborative optimization energy management method disclosed in Chinese Patent No. CN110667565A points out that the intelligent networked plug-in hybrid electric vehicle obtains the speed limit condition through the intelligent transportation system, uses dynamic programming to perform intelligent networked plug-in hybrid electric vehicle collaborative optimization energy management, and simultaneously optimizes the speed and energy management to maximize the energy consumption of the vehicle powertrain. However, the traditional collaborative optimization has the problems of large calculation burden and difficulty in solving. Moreover, the problem of battery life is not considered in the traditional optimal energy-saving driving control, which may lead to significant reduction of the battery life.

[0005] To this end, an energy-saving driving control method based on ensemble learning considering battery life is proposed, which effectively improves the above problems and provides a comprehensive solution for PHEV energy-saving driving control. SUMMARY

[0006] To solve the above problems, the present application proposes a PHEV energy-saving driving control method based on ensemble learning considering battery life.

[0007] A PHEV energy-saving driving control method based on ensemble learning considering battery life, the specific steps are as follows:

[0008] S1, collect a plurality of actual road working condition data under real driving scene, and construct comprehensive driving condition data set;

[0009] S2, optimize the energy management strategy of plug-in hybrid electric vehicle under comprehensive driving condition by dynamic programming algorithm, obtain global optimal comprehensive energy efficiency cost and optimal control decision data set, and comprehensive energy efficiency cost includes energy consumption cost and battery life cost;

[0010] S3, use the results of global optimization to extract optimal energy consumption cost, battery life cost and current data set;

[0011] S4, use neural network to construct adaptive SOC trajectory model, and calculate battery SOC according to vehicle driving state;

[0012] S5, based on ensemble learning algorithm, establish comprehensive energy efficiency cost model considering energy consumption cost and battery life cost;

[0013] S6, use the established comprehensive energy efficiency cost model and state transition model to design hierarchical energy-saving driving control architecture based on ensemble learning;

[0014] S7, the upper layer optimizes the economic speed based on ensemble learning considering battery life according to the system constraints under specific driving scene;

[0015] S8, the lower layer performs power system energy management control according to the planned speed trajectory.

[0016] Further, the actual road working condition of step S1 includes the driving state of the vehicle in congestion, urban area and high-speed type different road environment.

[0017] Further, step S2 needs to affect the key variables of power system energy consumption characteristics according to the research of existing PHEV energy management strategy, including vehicle speed, acceleration, driving mileage and SOC, and then perform energy management optimization.

[0018] Further, the step S3 obtains the optimal cumulative cost matrix and the optimal control decision matrix by DP algorithm optimization.

[0019] Further, the step S4 establishes a power system adaptive SOC trajectory model by estimating the state transition of the PHEV power system according to the vehicle driving state.

[0020] Further, the step S5 estimates the comprehensive energy efficiency cost of the PHEV power system according to the vehicle driving state, and establishes a comprehensive energy efficiency cost model having two base learners and an integrated network. The base learners include a BP-NN1 neural network for calculating energy consumption cost and a BP-NN2 network for calculating battery life cost, and the integrated network is an RBF-NN neural network, which takes the energy consumption cost and the battery life cost as inputs, integrates the results of the two base learners, and calculates the comprehensive energy efficiency cost as output.

[0021] Further, the step S6 extracts the data set for training the data-driven model from the DP optimization result, and the online module is divided into two parts: 1) economic speed optimization based on integrated learning; and 2) energy management of the power system.

[0022] Further, the step S7 calculates the comprehensive energy efficiency cost of the power system considering energy consumption and battery life according to the system state vector and the control decision vector by the comprehensive energy efficiency model, which is an evaluation model, and the adaptive SOC trajectory model is a system model, and the SOC is calculated according to the vehicle driving state.

[0023] The present application has the following beneficial effects: the data set of power system energy consumption, battery life and current state change is extracted from the optimization result of the DP algorithm, the comprehensive energy efficiency cost model considering energy consumption and battery life and the adaptive SOC trajectory system state transition model are established by using the neural network, which are used to express the coupling relationship of vehicle motion state, power system energy consumption, battery life and state change, efficiently decouple the complex coupling relationship between the energy consumption characteristics of the hybrid power system and the economic speed, and prolong the service life of the battery. BRIEF DESCRIPTION OF DRAWINGS

[0024] The present application will be further described below in combination with the drawings and examples.

[0025] Figure 1 The flowchart of the present application;

[0026] Figure 2 The power system adaptive SOC trajectory model of the present application;

[0027] Figure 3 The comprehensive energy efficiency model structure of the present application fusing energy consumption and battery life;

[0028] Figure 4 For the double-layer energy-saving driving optimal control architecture of the application;

[0029] Figure 5 For the NN-based DP optimization algorithm of the application;

[0030] Figure 6 For the economic vehicle speed trajectory planning result graph of the application. DETAILED DESCRIPTION

[0031] In order to make the technical means, creative features, purposes and effects of the application easy to understand, the application is further described below.

[0032] As shown in Figures 1 to 6 , a PHEV energy-saving driving control method based on integrated learning considering battery life, the specific steps are as follows:

[0033] 1) Collect multiple actual road working condition data under real driving scene, construct comprehensive driving working condition data set; In this embodiment, the driving working condition contains the driving state of the vehicle in congestion, urban area and highway and other different road environments;

[0034] 2) Optimize the energy management strategy of plug-in hybrid electric vehicle under comprehensive driving working condition through dynamic programming algorithm, obtain global optimal comprehensive energy efficiency cost and optimal control decision data set, and comprehensive energy efficiency cost includes energy consumption cost and battery life cost;

[0035] 2-1) According to the existing research on PHEV energy management strategy, the key variables affecting the energy consumption characteristics of power system include vehicle speed, acceleration, driving distance and SOC; The equivalent distance coefficient of driving distance and SOC on energy consumption is calculated according to the following formula:

[0036]

[0037] In the formula, SOC r represents the remaining SOC; S r represents the remaining driving distance, the coefficient L s represents the ratio of the maximum distance to the maximum SOC when fully charged; When S e is greater than or equal to 1, it can be inferred that the power is sufficient to maintain the remaining driving distance, and S e is set to 1 as the upper limit; When S e is within the range of [0, 1], it indicates that the remaining energy in the battery cannot maintain the remaining journey, and the internal combustion engine needs to participate in driving the vehicle to the destination;

[0038] 2-2) Energy management optimization part, according to dynamic programming to obtain economic speed trajectory, with power system energy consumption cost and battery life cost as objective function, power system control decision as control variable, SOC as state variable, to build energy management optimization problem:

[0039]

[0040] u(s) = P m (s)

[0041] x(s) = SOC(s)

[0042] In the formula, J ene is the objective function of energy management optimization, u and x are control variable vector and state variable of energy management optimization respectively;

[0043] 2-3) The optimization problem needs to meet the related constraints of vehicle power system, the instantaneous SOC is greater than the minimum SOC and less than the maximum SOC; the instantaneous battery current I bat (s) is greater than the minimum charging current and less than the minimum discharging current; the instantaneous engine torque T e (s) is greater than the minimum engine torque and less than the maximum engine torque; the instantaneous engine speed n e (s), power P e (s) respectively meet greater than the minimum value and less than the maximum value; the instantaneous motor speed n mot (s), torque T mot (s), power P e (m) respectively meet greater than the minimum value and less than the maximum value;

[0044] 3) The optimal cumulative cost matrix and optimal control decision matrix obtained by using DP algorithm optimization cannot be directly used for training of power system comprehensive energy efficiency model and state transition model, the results of global optimization are converted into optimal energy consumption cost, optimal battery life cost and current, the present application proposes a power system energy consumption and state data set extraction method based on DP optimal solution; the driving cycle and PHEV driving state are discretized, and the equivalent distance coefficient matrix under different battery SOC of each stage can be calculated according to the following formula:

[0045]

[0046] 4) A model of adaptive SOC trajectory of PHEV powertrain is established by estimating the state transition of PHEV powertrain according to the driving state of vehicle. The model covers a back propagation neural network (BP-NN) with three layers to construct the state transition model of adaptive SOC trajectory of PHEV powertrain. Levenberg-Marquardt (L-M) algorithm is used to search the optimal parameters of BP-NN in each iteration, which is an iterative algorithm for finding the extreme value of gradient and has the advantages of gradient method and Newton method. BP-NN is a feedforward neural network with sigmoid function as activation function, which can realize arbitrary nonlinear mapping between input and output. The learning process of BP-NN mainly includes two stages: the first stage is to input learning samples, set the initial weights and thresholds of the network, and calculate the output of each neuron in the first layer of the network. In the second stage, the weights and thresholds are modified, and the influence of each weight and threshold on the gradient (total error) is calculated from the last layer, so as to adjust each weight and threshold accordingly. The activation function is shown in the formula:

[0047]

[0048] where a 1 and a 0 are the neuron outputs of the current layer and the previous layer, respectively; n, W and b represent the cumulative output, weight and bias, respectively.

[0049] The adaptive SOC trajectory model is established as follows:

[0050]

[0051] where f BP-NN represents a nonlinear mapping function describing the relationship between the driving state of vehicle and the optimal SOC trajectory, as shown in Figure 2

[0052] The proposed DP based on neural network (NN) is used to solve the optimal driving speed curve of energy-saving driving. The powertrain model is combined into the economic speed optimization in the form of data-driven model, without using traditional methods to interpolate the table, derive the powertrain model, which can significantly reduce the calculation load of energy-saving driving control considering the energy consumption characteristics of powertrain, and realize efficient economic speed optimization considering the characteristics of powertrain;

[0053] 5) An integrated energy efficiency cost model considering energy consumption and battery life is established based on ensemble learning, and the overall framework is as follows: Figure 3 ​, the cost calculation module contains two base learners and an integrated network. The base learners include a BP-NN1 neural network for calculating energy consumption cost and a BP-NN2 network for calculating battery life cost, and the integrated network is an RBF-NN neural network that integrates the results of the first two base learners. The integrated network combines energy consumption and battery life costs, reduces the overall computational complexity, and improves the accuracy of overall energy consumption evaluation. RBF-NN is a widely used feedforward network for time series prediction, as it can approximate any continuous function with arbitrary precision; using radial basis functions (RBF) as the "basis" of the hidden layer elements, the hidden layer is constructed, so that the input low-dimensional pattern data can be converted to high-dimensional space, the input vector is transformed, so that the low-dimensional linearly inseparable problem becomes linearly separable in high-dimensional space; by updating the weights, it can be trained to learn highly nonlinear input / output relationships; compared with other neural network models, it has faster convergence speed and lower computational complexity; generally, Gaussian functions are used as RBF in the hidden layer to activate neurons, and the approximation function of RBF-NN is represented as:

[0054]

[0055] where x is the input vector, the weight vector is represented by ω j , and its dimension is K, j is the jth variance of the Gaussian center on x j ;

[0056] 5-1) Establish energy consumption NN1 cost model

[0057] The input vector x is composed of v, a and S e , and the output is the estimated powertrain energy consumption cost The model can be expressed as:

[0058] x = [v, a, S e ] T

[0059]

[0060] In the formula, f BP-NN is a nonlinear mapping function that describes the coupling relationship between vehicle driving state and optimal energy consumption of power system;

[0061] 5-2) Establish battery life NN2 cost model:

[0062] The input vector is I c , SOC, I, and the output is the estimated battery life cost cost bat , then the battery life cost model can be expressed as:

[0063]

[0064] 5-3) Integrated energy consumption cost and battery life cost to establish a comprehensive energy efficiency cost model, the input vector is cost bat 、 The output is the estimated comprehensive energy efficiency cost The comprehensive energy efficiency cost model can be expressed as:

[0065]

[0066] 6) Design a hierarchical energy-saving driving control architecture based on ensemble learning for hybrid system vehicles, as shown in Figure 4 The data set used to train the ensemble learning model is extracted from the DP optimization results in the offline module. The online module is divided into two parts:

[0067] 6-1) Economic vehicle speed optimization based on ensemble learning;

[0068] 6-2) Energy management of power system;

[0069] 7) Establish the upper layer of the energy-saving driving hierarchical control architecture: economic vehicle speed optimization based on ensemble learning. The standard DP algorithm cannot solve the optimal economic vehicle speed problem in the proposed control architecture, because the traditional power system model is not applied to economic vehicle speed optimization. A DP algorithm based on ensemble learning is established to solve the optimal driving speed curve of the energy-saving driving comprehensive energy efficiency model of the power system energy consumption cost and battery life. The evaluation model is defined as the system model, and the adaptive SOC trajectory model is calculated according to the vehicle driving state. As shown in Figure 5 The SOC state transition is calculated according to the following formula:

[0070]

[0071] In this way, without using traditional methods to interpolate the power system model table, derivative, the calculation load of energy-saving driving control considering the energy consumption characteristics of the power system can be significantly reduced;

[0072] Based on the optimization results of DP (Dynamic Programming, DP) algorithm, the data set of power system energy consumption and battery life state change is extracted, and the fusion energy efficiency cost model considering energy consumption and battery life is established using neural network, which is used to express the coupling relationship between vehicle motion state and power system energy consumption and state change. The complex coupling relationship between energy consumption characteristics of hybrid power system and economic vehicle speed is decoupled efficiently, and the service life of the battery is prolonged;

[0073] 7-1) Establish the target function of speed optimization:

[0074]

[0075] In the formula, J is the speed optimization target function of energy-saving driving control, wherein L1, L2, L3 are defined by the following equations:

[0076] L1=L c (x k ,u k )

[0077] L2=L t (x k ,u k )+L(x N )

[0078]

[0079] Wherein, L c represents the energy consumption cost of each discrete stage, L t indicates the driving time cost of each discrete stage, L(x N ) is the terminal cost of completing the driving task, L3 is the calculation of battery effective Ah-flux, which is a cost function to measure the battery aging, k represents the battery life cost converted into money cost coefficient, represents the weight coefficient, σ represents the severity factor of the battery, I C represents the charge and discharge rate of the battery;

[0080] The integrated learning model is introduced as the evaluation model and system model of speed optimization, which optimizes the energy consumption economy and battery life, and improves the calculation efficiency, and finally realizes the optimal control of PHEV energy-saving driving;

[0081] 7-2) In order to further improve the calculation efficiency, the driving time is eliminated from the state variable of optimal control problem by using the driving time penalty factor β, and the optimal control cost function is established:

[0082] L(x s ,u s )=L com +L t

[0083]

[0084] In the formula, L com represents the comprehensive energy efficiency cost of each discrete stage, L t indicates the driving time cost of each discrete stage, and the expected driving time can be adjusted by the weight coefficient through bisection method;

[0085] 7-3) The objective function of vehicle speed optimization can be transformed into:

[0086]

[0087] Where, L(x N ) is the terminal cost of completing the driving task, which can be determined based on the terminal constraints of vehicle speed and travel time;

[0088] 7-4) Establish control variables, state variables, and state transition equations based on ensemble learning speed optimization:

[0089] u=T

[0090] x=[a SOC] T

[0091]

[0092] SOC(s+1)=SOC(s)+f BP-NN

[0093] Where T represents the total output torque of the vehicle; M0 represents the vehicle mass; r w represents the tire radius; g represents the acceleration due to gravity; f represents the friction coefficient; C d Indicates the drag coefficient; A f Indicates the frontal area; F r Indicates driving resistance, F r The calculation formula is as follows:

[0094]

[0095] 7-5) Establish the optimal control speed, acceleration, duration, and SOC constraint conditions. The initial SOC is equal to SOC0, the instantaneous SOC(s) satisfies the conditions of being greater than the instantaneous minimum SOC and less than the instantaneous maximum SOC; the instantaneous acceleration a(s) is greater than the minimum instantaneous acceleration and less than the maximum instantaneous acceleration; the instantaneous speed v(s) is greater than the minimum instantaneous speed and less than the maximum instantaneous speed, and the starting speed v(0) is equal to v0; the time t(s k ) is less than the expected time for the vehicle to reach the destination;

[0096] 8) Lower layer of the energy-saving driving hierarchical control architecture: Power system energy management control is performed according to the planned vehicle speed trajectory, and torque control commands are sent to the power system engine, motor and other components.

[0097] In order to verify the proposed energy-saving driving control method based on integrated learning considering battery life, the present application compares and analyzes two energy-saving driving control strategies through simulation experiments, including: data-driven energy-saving driving control strategy without considering battery life, referred to as D-opt, and energy-saving driving control strategy based on integrated learning considering battery life, referred to as B-opt.

[0098] The comprehensive energy efficiency cost model established based on integrated learning can enable PHEV to cope with the complex relationship between the two in real environment, and find a reasonable balance between battery life and energy consumption, thereby improving the overall performance of PHEV and prolonging the battery usage time.

[0099] The economic speed trajectory planning results of the two energy-saving driving control strategies are shown in Table 1. Figure 6

[0100] Table 1 Comparison of energy consumption and battery life of UUDS working conditions

[0101]

[0102] As can be seen from Table 1, compared with the traditional D-opt strategy, the B-opt strategy proposed only increases the fuel consumption cost by 1.75%; while the effective ampere-hour flux and battery capacity recession amount are reduced by 22.4% and 12.8% respectively, which is a larger decrease than the fuel consumption.

[0103] Although the energy-saving driving control strategy considering battery aging increases fuel consumption to some extent, it significantly slows down the decline of battery capacity, thereby effectively reducing the degree of battery aging.

[0104] The terms involved in the present application are explained as follows:

[0105] Data-driven: A decision-making and operational method that emphasizes the use of data analysis and data insight to guide decision-making, optimize processes and develop strategies.

[0106] Dynamic programming algorithm: An optimization method for solving complex problems, particularly suitable for problems with overlapping subproblems and optimal substructure properties. Dynamic programming divides the problem into smaller subproblems, stores the solutions to these subproblems to avoid repeated calculations, thereby improving efficiency.

[0107] Neural network: A computational model that mimics the biological nervous system, such as the brain, for processing various types of complex tasks and data. They play a central role in machine learning and artificial intelligence. The basic unit of a neural network is a neuron, similar to a nerve cell in the biological nervous system. Each neuron receives input, performs weighted calculation, and generates output.​

[0108] Integrated Energy Efficiency Model: A framework used to assess and optimize the energy consumption efficiency of systems. It combines multiple factors and variables to provide a comprehensive analysis of energy usage efficiency, particularly in the context of electric vehicles and plug-in hybrid electric vehicles.

[0109] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A PHEV energy-saving driving control method based on ensemble learning and considering battery life, characterized by: The specific steps are as follows: S1. Collect multiple actual road condition data under real driving scenarios and build a comprehensive driving condition dataset; S2. Optimize the energy management strategy of the plug-in hybrid electric vehicle under comprehensive driving conditions through a dynamic programming algorithm to obtain the global optimal comprehensive energy efficiency cost and the optimal control decision data set. The comprehensive energy efficiency cost includes energy consumption cost and battery life cost. S3, using the results of global optimization to discretely extract the optimal energy consumption cost, battery life cost and current data set; S4. Using a neural network, an adaptive SOC trajectory model is constructed to calculate the battery SOC based on the vehicle's driving status; S5. Establish a comprehensive energy efficiency cost model that considers energy consumption cost and battery life cost based on an integrated learning algorithm; S6. Using the established comprehensive energy efficiency cost model and state transition model, design a hierarchical energy-saving driving control architecture based on ensemble learning; S7, the upper layer performs economic vehicle speed optimization based on ensemble learning and considering battery life according to system constraints in specific driving scenarios; S8, the lower layer performs power system energy management control according to the planned vehicle speed trajectory; The step S7 calculates the comprehensive energy efficiency cost of the integrated power system energy consumption cost and the battery life cost according to the system state vector and the control decision vector, the adaptive SOC trajectory model is a system model, and the SOC is calculated according to the vehicle driving state; 7-1) Establish the objective function for vehicle speed optimization: Where, is the vehicle speed optimization objective function for energy-saving driving control, where It is defined by the following equation: ； in, represents the energy consumption cost of each discrete stage, represents the travel time cost of each discrete stage, is the terminal cost of completing the driving task, The calculation of the battery's effective Ah-flux is a cost function to measure battery aging. Indicates the conversion coefficient of battery life cost into monetary cost, represents the weight coefficient, Indicates the severity factor of the battery, Indicates the charge and discharge rate of the battery; 7-2) In order to further improve the computational efficiency, the travel time penalty factor is used Eliminate travel time from the state variables of the optimal control problem and establish the optimal control cost function: Where, represents the comprehensive energy efficiency cost of each discrete stage, It represents the travel time cost of each discrete stage, and the expected travel time is achieved by adjusting the weight coefficient by dichotomy; 7-3) The objective function of vehicle speed optimization is transformed into: Where, is the terminal cost of completing the driving task, which is determined based on the terminal constraints of vehicle speed and travel time; 7-4) Establish control variables, state variables, and state transition equations based on ensemble learning speed optimization: in, Indicates the total output torque of the vehicle; Indicates the vehicle mass; Indicates tire radius; represents the acceleration due to gravity; represents the friction coefficient; Indicates the drag coefficient; represents the frontal area; Indicates the driving resistance, The calculation formula is as follows: ; 7-5) Establish optimal control speed, acceleration, duration, and SOC constraints. The initial SOC is equal to , instantaneous Satisfy the requirement of greater than the instantaneous minimum SOC and less than the instantaneous maximum SOC; instantaneous acceleration Greater than the minimum instantaneous acceleration and less than the maximum instantaneous acceleration; instantaneous speed Greater than the minimum instantaneous speed, less than the maximum instantaneous speed, starting speed equal ;often Less than the expected time for the vehicle to reach the destination.

2. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The actual road conditions in step S1 include the driving conditions of the vehicle in different road environments such as congested roads, urban areas, and highways.

3. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The step S2 needs to be based on the research of existing PHEV energy management strategies. The key variables that affect the energy consumption characteristics of the power system include vehicle speed, acceleration, mileage and SOC. 2-1) Calculate the equivalent distance coefficient of the impact of mileage and SOC on energy consumption; 2-2) Constructing the optimization problem of energy management: Where, is the objective function for energy management optimization, and are the control variable vector and state variables for energy management optimization, respectively.

4. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The step S3 obtains the optimal cumulative cost matrix and the optimal control decision matrix through DP algorithm optimization.

5. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The step S4 estimates the energy consumption and state transition of the PHEV power system based on the vehicle driving state, and establishes a power system adaptive SOC trajectory model using a BP neural network; Build an adaptive SOC trajectory model whose input is acceleration ,speed and the equivalent distance coefficient , the output is the trajectory of SOC, as shown in the following formula: 。 6. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The comprehensive energy efficiency cost model of step S5 includes two base learners and an integrated network. The base learners include a BP-NN1 neural network for calculating energy consumption cost and a BP-NN2 network for calculating battery life cost. The integrated network is an RBF-NN neural network that integrates the results of the first two base learners. 6-1) Establish the energy consumption NN1 cost model, input is and The output is the estimated powertrain energy consumption cost , the model can be expressed as: Where, A nonlinear mapping function that describes the coupling relationship between the vehicle's driving state and the optimal energy consumption of the power system; 6-2) Establish the battery life NN2 cost model, the input vector is , the output is the estimated battery life cost , then the battery life cost model is expressed as: ; 6-3) Integrate energy consumption cost and battery life cost to establish a comprehensive energy efficiency cost model. The input vector is The output is the estimated comprehensive energy efficiency cost , then the comprehensive energy efficiency cost model is expressed as: 。 7. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The DP optimization results in step S6 are used to extract a data set for training the ensemble learning model. The online module is divided into two parts: 1) economical vehicle speed optimization based on ensemble learning; 2) energy management of the power system.

8. The PHEV energy-saving driving control method based on ensemble learning and considering battery life according to claim 1, characterized in that: The step S8 performs power system energy management control according to the planned vehicle speed trajectory and sends torque control instructions to the power system engine and motor components.

Images