CN112193232B

CN112193232B - Self-adaptive energy management system and method for hybrid electric vehicle

Info

Publication number: CN112193232B
Application number: CN202011013209.3A
Authority: CN
Inventors: 施德华; 刘盛; 容香伟; 汪少华; 蔡英凤; 陈龙
Original assignee: Jiangsu University
Current assignee: Jiangsu University
Priority date: 2020-09-23
Filing date: 2020-09-23
Publication date: 2021-10-12
Anticipated expiration: 2040-09-23
Also published as: CN112193232A

Abstract

The invention discloses a self-adaptive energy management system and a self-adaptive energy management method for a hybrid electric vehicle. The mode judging module judges the working mode of the vehicle, the driving control module and the braking control module adopt corresponding control strategies to solve ideal output torques of different power sources, the torque distribution module further distributes the output torques of the different power sources according to the output quantity of the driving control module or the braking control module, and the driving control module adopts an equivalent fuel consumption minimum control strategy. Aiming at the defect that the working condition adaptability of the fixed equivalent factor in the minimum control strategy of equivalent fuel consumption is poor, the equivalent factor solving module is adopted to carry out fuzzy adaptive solving on the equivalent factor according to the average required power of the road section predicted by the average power prediction module and the current charge state of the battery, so that the whole vehicle fuel economy of the hybrid electric vehicle under the large-range road driving condition is remarkably improved.

Description

Self-adaptive energy management system and method for hybrid electric vehicle

Technical Field

The invention relates to an energy management optimization method for a hybrid electric vehicle, and belongs to the field of overall optimization control of the hybrid electric vehicle.

Background

China has become the largest global single automobile market, the problems of air pollution and energy shortage caused by the traditional fuel oil automobile become more severe, and the hybrid power technology is definitely proposed by the technical route map of energy-saving and new energy automobiles in China as the first automobile energy-saving technology and is worthy of vigorous development. The energy management strategy is the key for improving the fuel economy of the hybrid electric vehicle, and the hybrid electric vehicle can fully exert the advantages of energy conservation and emission reduction by combining with an effective energy management strategy. Currently, energy management strategies, whether rule-based or optimization algorithm-based, are widely studied. The method has the advantages that logic threshold values are predefined based on a rule strategy, the system working mode and the torque distribution are determined according to the running state of the vehicle, the control is simple to realize, the real-time performance is strong, the wide engineering application is obtained, the control effect depends on the accuracy of expert experience, and the high-efficiency control requirement under the complex working condition cannot be met. The control algorithm based on global optimization is heavy in calculation task, the whole cycle working condition needs to be known in advance, and the control algorithm cannot be used for on-line control of the vehicle, but the obtained global optimal result can be used as other bases for judging the quality of other capacity management strategies.

The minimum strategy of equivalent fuel consumption is used as a real-time optimization control method, the electric energy consumption of a battery is converted into corresponding fuel consumption, the total equivalent fuel consumption of the whole vehicle is constructed by combining the actual fuel consumption of an engine, the optimal control quantity is obtained by minimizing the total equivalent fuel consumption, the real-time performance is good, the better application is achieved, but the defect of poor working condition adaptability still exists, namely, the fixed equivalent factor cannot be applied to optimization control under the large-range working conditions. With the development of intelligent traffic technology and wireless communication technology, the road section characteristic information of the vehicle in the future road section or time can be predicted and obtained, the prediction information is combined with the minimum strategy of equivalent fuel consumption, and the equivalent factor self-adaption method based on the prediction information is constructed, so that the method has important significance for improving the self-adaption of the minimum strategy of equivalent fuel consumption under the condition of large-range road conditions, and is also beneficial to further improving the control effect of the minimum strategy of equivalent fuel consumption.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: the battery overcharge or overdischarge caused by the equivalent fuel consumption minimum strategy with a fixed equivalent factor under different working conditions is avoided, and the working condition adaptability of the equivalent fuel consumption minimum strategy is improved, so that the whole vehicle fuel economy of the hybrid electric vehicle under the large-range road driving condition is improved. Aiming at the technical problem, the invention provides a hybrid electric vehicle adaptive energy management system and a hybrid electric vehicle adaptive energy management method, which utilize predicted road section characteristic parameters to carry out fuzzy adaptive adjustment on equivalent factors in an equivalent fuel consumption minimum strategy, thereby effectively maintaining the charge-discharge balance of a battery under complex road conditions and improving the fuel economy of a vehicle. The technical scheme adopted by the invention is as follows:

a self-adaptive energy management system and method for a hybrid electric vehicle are disclosed, the self-adaptive energy management system comprises a mode discrimination module, a driving control module, a braking control module, a torque distribution module, an average power prediction module and an equivalent factor solving module, the mode discrimination module judges whether the vehicle is in a driving or braking mode according to an accelerator pedal, a brake pedal and a vehicle speed signal of a driver, and transmits the obtained mode signal to a driving control module, a braking control module and a torque distribution module, when the vehicle is in a driving mode, the driving control module determines the optimal engine torque and the optimal rotating speed, when the vehicle is in a braking mode, the braking control module determines the regenerative braking torque of the motor on the premise of ensuring the braking safety, and the torque distribution module further solves the output torques of other parts according to the torques of different parts obtained by the driving control module or the braking control module and the vehicle working mode signals determined by the mode discrimination module. The average power prediction module predicts the average required power of the vehicle in a future section of limited time domain according to the road section characteristic parameters and outputs the average required power to the equivalent factor solving module, and the equivalent factor solving module periodically determines the equivalent factor and inputs the equivalent factor to the drive control module.

Preferably, the average power prediction module predicts the average required power of the vehicle in a limited time domain in the future, the required input is 3 road section characteristic parameters of the average speed, the speed standard deviation and the average acceleration of the vehicle in the predicted time domain, the adopted average required power prediction model is a BP neural network model, the model parameters are obtained off line based on sample working conditions, and the sample working conditions comprise international universal standard cycle working conditions and actually acquired vehicle operation working conditions.

Preferably, the drive control module adopts an equivalent fuel consumption minimum strategy to perform energy optimization management, the brake control module adopts an optimal brake energy recovery strategy to determine the distribution of the regenerative brake torque and the friction brake torque, namely when the total brake torque of the vehicle is smaller than the maximum regenerative brake torque, the motor only plays a role in regenerative brake torque, when the total brake torque is larger than the regenerative brake torque, the motor provides the maximum regenerative brake torque, and the rest brake torques are provided by the friction brake torque.

Preferably, the equivalent factor solving module periodically solves the equivalent factor by using time or distance as intervals through a fuzzy control algorithm, the input of the equivalent factor solving module is the predicted average required power and the current state of charge of the battery, and the output is the equivalent factor.

The self-adaptive energy management method provided by the technical scheme of the invention comprises the following steps:

(1) determining the type and the working mode of the hybrid electric vehicle, and establishing a steady-state rotating speed coupling equation and a torque coupling equation of an engine, a motor and required torque in a vehicle driving and braking mode;

(2) determining a hybrid electric vehicle working mode judging method of a mode judging module according to an accelerator pedal, a brake pedal and a vehicle speed signal, and transmitting a determined vehicle driving or braking mode signal to a driving control module, a braking control module and a torque distribution module;

(3) aiming at the driving process of a vehicle, determining an objective function and a constraint condition of a minimum strategy of equivalent fuel consumption, converting the electric energy consumption of a battery into corresponding equivalent fuel consumption by adopting an equivalent factor, and establishing the objective function as

In the formula (I), the compound is shown in the specification,

to include engine oil consumption

Equivalent fuel consumption to battery

The total equivalent fuel consumption is an objective function of a minimum strategy of the equivalent fuel consumption; s_equIs an equivalence factor; h_lhvIs the engine fuel calorific value; eta_batIndicating the working efficiency of the battery; p_LThe load power of the battery is determined by the working power of a motor and an electric accessory of the hybrid electric vehicle;

the constraint conditions comprise the rotating speed and the torque limit value of different power sources such as an engine and a motor, and the working current and the voltage limit value of a battery;

(4) constructing an average required power prediction model based on three road section characteristic parameters of vehicle average speed, speed standard deviation and average acceleration by adopting a BP neural network, training the BP neural network based on sample working conditions, and determining model parameters;

(5) an average power prediction module of the self-adaptive energy management system predicts corresponding average required power based on average speed, speed standard deviation and average acceleration in a future period of time according to an average required power prediction model, and inputs the average required power to an equivalent factor solving module;

(6) an equivalent factor self-adaptive solving method based on fuzzy control is constructed, input variables of a fuzzy controller are predicted average required power and the current state of charge of a battery, output variables are equivalent factors, fuzzy control rules, domains of input and output variables and linguistic variables are determined, and an equivalent factor self-adaptive solving module determines the current equivalent factor according to the fuzzy control and inputs the current equivalent factor to a driving control module;

(7) determining system control variables, wherein the control variables are engine torque and rotating speed in a drive control module, when the vehicle is in a drive mode, the drive control module obtains optimal engine torque and rotating speed according to the objective function and constraint conditions established in the step (3) and the current equivalent factor determined in the step (6), the required input variables comprise a vehicle accelerator pedal signal, actual vehicle speed and battery charge state, and the obtained optimal engine torque and rotating speed are input to a torque distribution module; in the braking control module, control variables are regenerative braking torque and rotating speed of a motor used for recovering braking energy, and when the vehicle is in a braking mode, the braking control module determines the regenerative braking torque and the rotating speed of the motor according to an optimal braking energy recovery strategy and inputs the regenerative braking torque and the rotating speed to the torque distribution module;

(8) when the vehicle is in a driving mode, the torque distribution module determines the output torque of the motor according to the optimal rotating speed and torque of the engine obtained by the driving control module and the steady-state torque coupling equation of the system driving mode established in the step (1), and when the vehicle is in a braking mode, the torque distribution module determines the friction braking torque of the brake according to the regenerative braking torque and rotating speed of the motor obtained by the braking control module and the steady-state torque coupling equation of the system braking mode established in the step (1), so that the self-adaptive optimal distribution of the output torques of different power sources of the hybrid electric vehicle is realized;

preferably, in step (5), the average speed, the speed standard deviation and the average acceleration may be obtained from a traffic information center based on a wireless communication technology, or a future vehicle speed sequence is predicted based on the current vehicle speed and the historical vehicle speed information by using a markov model, an autoregressive model and a BP neural network model, and the average speed, the speed standard deviation and the average acceleration of the future vehicle speed sequence are further obtained.

The method has the advantages that the advantages of the minimum equivalent fuel consumption strategy and the vehicle predicted working condition information are fully combined, the self-adaptive adjustment of the equivalent factor in the minimum equivalent fuel consumption strategy in the vehicle driving process is realized by introducing the predicted road section characteristic parameter information, so that the working condition adaptability of the energy management strategy under the complex working condition is improved, the efficient management of the energy of the whole vehicle under the large-range working condition is realized, and the method has a remarkable effect on further improving the fuel economy of the hybrid electric vehicle.

Drawings

FIG. 1 is a schematic diagram of an adaptive energy management system for a hybrid vehicle.

FIG. 2 is a schematic diagram of a hybrid vehicle powertrain based on a planetary row configuration.

FIG. 3 is a schematic diagram of a BP neural network-based average demand power prediction model.

FIG. 4 is a schematic diagram of a future vehicle speed sequence prediction model based on a BP neural network.

FIG. 5 is a schematic diagram of equivalence factor adaptive solution based on fuzzy control.

Fig. 6 is an engine speed control principle based on PI control.

In fig. 2: s1-front planet row sun gear C1-front planet row planet carrier R1-front planet row gear ring S2-rear planet row sun gear C2-rear planet row planet carrier R2-rear planet row gear ring

In fig. 4: t-current time v (t) -vehicle speed at current time v (t-1) -vehicle speed in one second before v (t-2) -vehicle speed in two seconds before v (t-N)_his+1)-N_hisVehicle speed v (t-N) 1 second ago_his)-N_hisVehicle speed v (t +1) before second, vehicle speed v (t +2) next second, and vehicle speed v (t + N) two seconds after_pre) Future N_preSpeed of the vehicle N in seconds_his-length of historical vehicle speed sequence N_pre-predicting vehicle speed train length

In fig. 6: omega_ErefOptimum engine speed T_Eref-engine optimum torque Δ T_GGenerator dynamic compensation rotation T_GstatSteady state output torque K of the generator₁Front planetary characteristic parameter ω_ESteady engine speed T_GrefGenerator torque capacity

Detailed Description

The invention is further described with reference to the following figures and specific examples.

Fig. 1 is a schematic structural diagram of an adaptive energy management system to be used in an embodiment of the present invention. The system comprises a mode discrimination module, a driving control module, a braking control module, a torque distribution module, an average power prediction module and an equivalent factor solving module. The mode judging module judges whether the vehicle is in a driving or braking mode according to the signals of an accelerator pedal, a brake pedal and the vehicle speed of a driver, and transmits the obtained mode signals to the driving control module, the braking control module and the torque distribution module.

When the vehicle is in a driving mode, the driving control module adopts an equivalent fuel consumption minimum strategy to carry out energy optimization management and determine the optimal engine torque and the optimal rotating speed, and the equivalent factor adaptive optimization method based on the road section characteristic parameters is provided aiming at the defect of poor working condition adaptability of the equivalent fuel consumption minimum strategy caused by the fixed equivalent factors. The average power prediction module is used for predicting the average required power of the vehicle in a future limited time domain according to the road section characteristic parameters and outputting the average required power to the equivalent factor solving module, and the equivalent factor solving module is used for periodically solving the equivalent factor according to the predicted average required power and the current state of charge of the battery by adopting a fuzzy control algorithm, so that the control parameters in the drive control module are updated. The input of the average power prediction module is 3 road section characteristic parameters of the average speed, the speed standard deviation and the average acceleration of the vehicle in the prediction time domain, the adopted average demand power prediction model is a BP neural network model, the model parameters are obtained on the basis of sample working conditions in an off-line mode, and the sample working conditions comprise international universal standard cycle working conditions and actually acquired vehicle operation working conditions.

When the vehicle is in a braking mode, the braking control module adopts an optimal braking energy recovery strategy to determine the regenerative braking torque of the motor on the premise of ensuring the braking safety.

The torque distribution module further solves the output torques of other components according to the torques of different components obtained by the driving control module or the braking control module and the vehicle working mode signal determined by the mode discrimination module.

To further illustrate the adaptive energy management method proposed by the embodiment of the present invention, a hybrid electric vehicle based on a planetary row structure is taken as a research object, a schematic diagram of a transmission system of the hybrid electric vehicle is shown in fig. 2, a power coupling mechanism including two planetary rows is used for coupling output power of an engine, a generator and a motor, the engine is connected with a front planetary row carrier C1, the generator is connected with a front planetary row sun gear S1, the motor is connected with a rear planetary row sun gear S2, a front planetary row ring gear R1, a rear planetary row carrier C2 and an output shaft are connected, and the power coupled by the system is output by the output shaft. For the hybrid vehicle transmission system shown in fig. 2, the adaptive energy management method comprises the following steps:

(1) the type of the hybrid electric vehicle is determined to be a series-parallel type adopting a planetary row structure, and the working modes comprise a pure electric mode, a hybrid power mode, a parking charging mode, a mechanical braking mode and a regenerative braking mode. The system comprises a drive control module, a pure electric mode, a hybrid power mode, a parking mode and a parking power generation mode, wherein the pure electric mode, the hybrid power mode, the parking mode and the parking power generation mode can be described as drive modes, and the drive control module is used for solving the optimal engine speed and torque in the drive modes; the mechanical braking mode and the regenerative braking mode are unified into a braking mode, and the braking control module carries out torque distribution solving under the corresponding modes.

When the vehicle is in drive mode, the steady state torque coupling equation for the system is

In the formula, T_E，T_GAnd T_MOutput torques of the engine, the generator, and the motor, respectively; t is_outA load torque required for the output shaft; k₁And K₂Characteristic parameters of the front planet row and the rear planet row are respectively.

When the vehicle is in a braking mode, the motor is adopted to recover regenerative braking energy, the output torque of the engine is zero, and the steady-state torque coupling equation of the system is

T_brk＝T_req-T_{reg_max} (2)

In the formula, T_brkAnd T_reqA total friction braking torque provided to the brakes and a total braking torque required at the wheels, respectively; t is_{reg_max}The maximum regenerative braking torque that can be provided for the motor is the equivalent regenerative braking torque at the wheels. The optimal braking energy recovery strategy is adopted to distribute the regenerative braking torque and the friction braking torque, and when the total braking torque T required at the wheel_reqLess than the maximum equivalent regenerative braking torque T of the motor_{reg_max}In time, only the regenerative braking torque of the motor acts, the frictional braking torque T_brkIs zero; total braking torque T required at the wheels_reqGreater than the maximum equivalent regenerative braking torque T of the motor_{reg_max}At this time, the motor provides the maximum regenerative braking torque, and the remaining braking torque is provided by the friction braking torque system.

The steady state rotational speed coupling model of the system is

In the formula, ω_E，ω_GAnd ω_MSteady state rotational speeds omega of engine, generator and motor, respectively_outIndicating the rotational speed of the output shaft.

(2) The hybrid electric vehicle working mode judging method comprises the steps of determining a mode judging module according to an accelerator pedal, a brake pedal and a vehicle speed signal, and transmitting a determined vehicle driving or braking mode signal to a driving control module, a braking control module and a torque distribution module. If the accelerator pedal signal is larger than zero or the vehicle speed is zero, the mode discrimination module determines that the vehicle works in a driving mode, and if the brake pedal signal is larger than zero, the mode discrimination module determines that the vehicle works in a braking mode.

(3) Aiming at a vehicle driving mode, converting the electric energy consumption of a battery into corresponding equivalent fuel consumption by adopting an equivalent factor, and establishing an objective function of

In the formula (I), the compound is shown in the specification,

to include engine oil consumption

Equivalent fuel consumption to battery

The total equivalent fuel consumption is an objective function of a minimum strategy of the equivalent fuel consumption; s_equIs an equivalence factor; h_lhvIs the engine fuel calorific value; eta_batIndicating the working efficiency of the battery; p_LThe load power of the battery is determined by the working power of the motor and the electric accessories of the hybrid electric vehicle.

The constraint conditions comprise the rotating speed and torque limit values of different power sources such as an engine, a motor and the like, and the working current and voltage limit values of a battery, and are expressed as

In the formula I_batAnd U_batThe working current and voltage of the battery are respectively; (ii) a reaction product of (a) and (b),_maxand (c) a reaction product of (c),_minrespectively, the maximum and minimum values of the corresponding variable.

(4) And (3) constructing an average required power prediction model based on three road section characteristic parameters of the average speed, the standard deviation of the speed and the average acceleration of the vehicle by adopting a BP neural network, wherein the schematic diagram of the prediction model is shown in FIG. 3. And solving the model parameters based on the sample working conditions, wherein the adopted sample working conditions comprise international universal standard cycle working conditions, such as standard cycle working conditions of NEDC, UDDS, WLTC and the like, and the acquisition of corresponding operation working conditions can be carried out according to the actual operation environment of the target vehicle. And aiming at the sample working condition, dividing by time or by taking the length of the road section as an interval, solving the average speed, the standard deviation of the speed, the average acceleration and the average required power of each road section, and forming a sample database of the characteristic parameters of the road section. Training a neural network model based on sample data, wherein training inputs are average vehicle speed, speed standard deviation and average acceleration, training outputs are average required power, and parameter variables such as a training function, a learning function, an activation function and a learning factor are selected through debugging, so that model parameters of the BP neural network are obtained. The solution of the model parameters is completed off line, the constructed average required power prediction model is applied to the average power prediction module, and the model parameters can be periodically updated off line according to the actual operation condition of the vehicle and input to the average power prediction module.

(5) The average speed, the speed standard deviation and the average acceleration in a future time domain can be obtained in various ways, for example, the existing vehicle speed prediction models such as a Markov model, an autoregressive model, a BP neural network model and the like are adopted, a vehicle speed sequence in the future limited time domain is obtained based on the current vehicle speed and historical vehicle speed information, the average speed, the speed standard deviation and the average acceleration are further integrated by 3 road section characteristic parameters, a traffic information center can also perform cloud calculation of the road average speed, the speed standard deviation, the average acceleration and other characteristic parameters according to the collected road traffic big data information, and a vehicle-mounted control module receives corresponding characteristic information in the future road section based on the technologies such as wireless communication and the like. The invention takes the example of adopting a 3-layer BP neural network model to predict the future vehicle speed sequence, the constructed future vehicle speed sequence prediction method based on the BP neural network model is shown in figure 4, and the inputs of the prediction model are the current vehicle speed v (t) and the past N_hisA sequence of historical vehicle speeds in seconds [ v (t-1) v (t-2) ]. v (t-N)_his+1)v(t-N_his)]The output is the future N_preV (t +1) v (t + 2.) a sequence of vehicle speeds in seconds (v)t+N_pre)]. Based on a predicted sequence of vehicle speeds, average speed v_avgStandard deviation of velocity v_stdAnd average acceleration a_avgAre respectively as

(6) And solving the equivalent factor by adopting fuzzy control, wherein a schematic diagram of the fuzzy controller is shown in fig. 5, input variables are predicted average required power and the current state of charge of the battery, output variables are equivalent factors, and the domain and linguistic variables of the input and output variables are determined. According to the actual power demand of the vehicle under different working conditions, defining the domain of average power demand as [ -15kW, 30kW ], and language variables as { negative large, negative small, zero, positive small, positive middle and positive large }; the argument and linguistic variables of the battery state of charge are [0.4, 0.7] and { negative large, negative small, zero, positive small, positive large }, respectively, the output argument of the equivalent factor is [1.3, 3.5], and the linguistic variables are { negative large, negative medium, negative small, zero, positive small, positive medium, positive large }. By formulating a fuzzy rule, an equivalent factor which is suitable for the characteristic parameters of the current prediction time domain section can be obtained according to the average required power and the battery charge state, and the equivalent factor is input into the driving control module, so that the optimal distribution of energy is realized.

(7) In the driving control module, the control variables are engine torque and rotating speed, when the vehicle is in a driving mode, the driving control module obtains optimal engine torque and rotating speed according to the objective function and the constraint condition established in the step (3) and the current equivalent factor determined in the step (6), the required input variables comprise a vehicle accelerator pedal signal, actual vehicle speed and battery state of charge, and the obtained optimal engine torque and rotating speed are input to the torque distribution module; in the braking control module, the control variables are the regenerative braking torque and the rotating speed of a motor used for recovering the braking energy, and when the vehicle is in a braking mode, the braking control module determines the regenerative braking torque and the rotating speed of the motor according to an optimal braking energy recovery strategy and inputs the regenerative braking torque and the rotating speed to the torque distribution module.

(8) When the vehicle is in a driving modeThe torque distribution module is used for calculating the optimal rotating speed omega of the engine according to the driving control module_ErefAnd torque T_ErefAnd (2) determining the steady-state output torques of the generator and the motor by the system driving mode steady-state torque coupling equation established in the step (1), namely

In the formula, T_GstatAnd T_MstatSteady state output torques of the generator and the motor, respectively, wherein the motor steady state output torque T_MstatAs ideal output torque T_MrefMay be directly output by the torque distribution module. Aiming at the planetary-row type hybrid power structure with two degrees of freedom of the embodiment of the invention, in order to enable the actual rotating speed of the engine to track the optimal rotating speed determined by the equivalent fuel consumption minimum strategy, an engine speed regulator based on PI control needs to be designed in the torque distribution module, as shown in FIG. 6. The input of the PI speed regulator is the ideal rotating speed and the actual rotating speed of the engine, and the output is the dynamic compensation torque Delta T of the generator_GTherefore, the torque distribution module is based on the engine optimal speed ω_ErefAnd torque T_ErefDetermined generator ideal output torque T_GrefIs composed of

T_Gref＝T_Gstat+ΔT_G (9)

When the vehicle is in a braking mode, the output torque of the engine is zero, the rotating speed of the engine is idle rotating speed or zero, the torque distribution module determines the friction braking torque of the brake according to the regenerative braking torque of the motor obtained by the braking control module and the system braking mode steady-state torque coupling equation established in the step (1), and in order to regulate the speed of the engine, the torque distribution module determines the friction braking torque of the brake according to the rotating speed omega of the engine_ErefDetermined generator ideal output torque T_GrefIs composed of

T_Gref＝ΔT_G (10)。

Claims

1. The management method of the hybrid electric vehicle adaptive energy management system is characterized in that the adaptive energy management system comprises a mode discrimination module, a driving control module, a braking control module, a torque distribution module, an average power prediction module and an equivalent factor solving module, wherein the mode discrimination module judges whether a vehicle is in a driving or braking mode according to signals of an accelerator pedal, a brake pedal and a vehicle speed of a driver, and transmits the obtained mode signals to the driving control module, the braking control module and the torque distribution module; when the vehicle is in a driving mode, the driving control module determines the optimal engine torque and the optimal rotating speed, when the vehicle is in a braking mode, the braking control module determines the regenerative braking torque of the motor on the premise of ensuring the braking safety, and the torque distribution module further solves the output torques of other parts according to the torques of different parts obtained by the driving control module or the braking control module and the vehicle working mode signals determined by the mode discrimination module; the average power prediction module predicts the average required power of the vehicle in a future section of limited time domain according to the road section characteristic parameters and outputs the average required power to the equivalent factor solving module, and the equivalent factor solving module periodically determines an equivalent factor and inputs the equivalent factor to the drive control module;

the adaptive energy management method comprises the following steps:

In the formula (I), the compound is shown in the specification,

to include engine oil consumption

Equivalent fuel consumption to battery

The total equivalent fuel consumption is an objective function of a minimum strategy of the equivalent fuel consumption; s_equIs an equivalence factor; h_lhvIs the engine fuel calorific value; eta_batIndicating the working efficiency of the battery; p_LThe load power of the battery is determined by the working power of a motor and an electric accessory of the hybrid electric vehicle; the constraint conditions comprise the rotating speed and the torque limit value of different power sources such as an engine and a motor, and the working current and the voltage limit value of a battery;

(4) adopting a BP neural network to construct an average required power prediction model based on three road section characteristic parameters of vehicle average speed, speed standard deviation and average acceleration, and training the BP neural network based on sample working conditions to determine model parameters when adopting the BP neural network to construct the prediction model;

(8) when the vehicle is in a driving mode, the torque distribution module determines the output torque of the motor according to the optimal rotating speed and torque of the engine obtained by the driving control module and the steady-state torque coupling equation of the system driving mode established in the step (1), and when the vehicle is in a braking mode, the torque distribution module determines the friction braking torque of the brake according to the regenerative braking torque and rotating speed of the motor obtained by the braking control module and the steady-state torque coupling equation of the system braking mode established in the step (1), so that the self-adaptive optimal distribution of the output torques of different power sources of the hybrid electric vehicle is realized.

2. The management method of the hybrid electric vehicle adaptive energy management system according to claim 1, characterized in that the average power prediction module predicts the average required power of the vehicle in a limited time domain in the future, the required input is 3 road section characteristic parameters of the average speed, the speed standard deviation and the average acceleration of the vehicle in the predicted time domain, the adopted average required power prediction model is a BP neural network model, the model parameters are obtained off-line based on sample working conditions, and the sample working conditions comprise international universal standard cycle working conditions and actually acquired vehicle operation working conditions.

3. The management method of the hybrid electric vehicle adaptive energy management system according to claim 1, characterized in that the drive control module adopts an equivalent fuel consumption minimum strategy to perform energy optimization management, the brake control module adopts an optimal braking energy recovery strategy to determine the distribution of regenerative braking torque and friction braking torque, i.e. when the total braking torque of the vehicle is smaller than the maximum regenerative braking torque, only the regenerative braking torque of the motor acts, when the total braking torque is larger than the regenerative braking torque, the motor provides the maximum regenerative braking torque, and the rest braking torques are provided by the friction braking torque.

4. The management method of the hybrid electric vehicle adaptive energy management system according to claim 1, characterized in that the equivalent factor solving module periodically solves the equivalent factors at intervals of time or distance by adopting a fuzzy control algorithm, the input of the equivalent factor solving module is the predicted average required power and the current state of charge of the battery, and the output is the equivalent factor.

5. The method as claimed in claim 1, wherein in step (5), the average speed, the speed standard deviation and the average acceleration are obtained from a traffic information center based on a wireless communication technology, or a future vehicle speed sequence is predicted based on the current vehicle speed and the historical vehicle speed information by using a BP neural network model, and the average speed, the speed standard deviation and the average acceleration of the future vehicle speed sequence are further obtained.