CN114889580B - Hybrid vehicle energy management control method, system, device and storage medium - Google Patents

Abstract

The invention discloses a hybrid vehicle energy management control method, a system, a device and a storage medium, which can improve vehicle driving responsiveness and high-voltage safety. The method comprises the following steps: acquiring historical vehicle speed information and training a vehicle speed prediction model; predicting a predicted speed sequence through a vehicle speed prediction model; predicting a demand power predicted value according to the predicted speed sequence; constructing a dynamic optimization objective function through a range extender power following error according to the demand power predicted value; constructing a safety optimization objective function according to the predicted value of the required power; constructing a dynamic optimization weight coefficient and a safety optimization weight coefficient; a simultaneous dynamic optimization weight coefficient, a safety optimization weight coefficient, a dynamic optimization objective function and a safety optimization objective function are multiple-objective coordinated optimization objective functions; constructing a required power damping coefficient function; constructing a safety constraint condition; and solving the multi-objective coordination optimization objective function and the required power damping coefficient function according to the safety constraint condition to obtain a control sequence matrix.

Description

Hybrid vehicle energy management control method, system, device and storage medium

Technical Field

The present invention relates to the technical field of hybrid vehicle energy management, and in particular to a hybrid vehicle energy management control method, system, device and storage medium.

Background Art

As an inevitable development trend of the future automobile industry, new energy vehicles have broad prospects. As an important part of new energy vehicles, hybrid vehicles have a major advantage of coordinated and complementary multiple power sources. How to coordinate and distribute the energy flow of multiple power sources on the premise of meeting the driver's power needs and optimize the vehicle's performance in terms of power and fuel economy, which puts high demands on the vehicle energy management strategy. In related technologies, energy management strategies are mainly divided into the following four types: rule-based, fuzzy rule-based, global optimization-based, and instantaneous optimization-based. In the research of related technologies, whether it is rule-based or optimization-based control strategies, their control objectives are mainly to improve fuel economy. When dealing with off-road conditions with large longitudinal vehicle speed mutations and strong nonlinear changes in required power, their robustness is generally poor, and it is difficult to achieve global optimization of drive responsiveness and vehicle safety.

Summary of the invention

In order to solve at least one of the above technical problems, the present invention proposes a hybrid vehicle energy management control method, system, device and storage medium, which can improve the driving responsiveness and high-voltage safety of the hybrid vehicle.

In one aspect, an embodiment of the present invention provides a hybrid vehicle energy management control method, comprising the following steps:

Obtaining historical vehicle speed information of the vehicle, and training a vehicle speed prediction model based on the historical vehicle speed information;

Inputting the historical vehicle speed information and the real-time vehicle speed information into the vehicle speed prediction model to predict a predicted speed sequence;

Predicting the required power of the vehicle according to the predicted speed sequence to obtain a required power prediction value;

According to the demand power prediction value, a dynamic optimization objective function is constructed through a range extender power following error;

Constructing a safety optimization objective function according to the demand power prediction value;

Construct dynamic optimization weight coefficient and safety optimization weight coefficient;

By combining the dynamics optimization weight coefficient, the safety optimization weight coefficient, the dynamics optimization objective function and the safety optimization objective function through a cost function, a multi-objective coordinated optimization objective function is obtained;

Constructing a required power damping coefficient function; the required power damping coefficient is positively correlated with the rate of change of the actual required power of the vehicle;

Construct safety constraints based on engine parameters and battery parameters;

The multi-objective coordinated optimization objective function and the required power damping coefficient function are solved according to the safety constraint conditions to obtain a control sequence matrix; the control sequence matrix includes the engine target speed, the engine torque and the required power damping coefficient.

A hybrid vehicle energy management control method according to an embodiment of the present invention has at least the following beneficial effects: In this embodiment, the historical vehicle speed information of the hybrid vehicle is first obtained, and then the vehicle speed prediction model is trained according to the historical vehicle speed information, so that the vehicle speed prediction model is used to predict the historical vehicle speed information and the real-time vehicle speed information to obtain a predicted speed sequence. At the same time, the vehicle demand power prediction value is obtained according to the predicted speed sequence prediction. Then, according to the demand power prediction value, the power optimization objective function is constructed by the range extender power following error. In addition, this embodiment also constructs a safety optimization objective function according to the demand power prediction value. In this embodiment, a dynamic optimization weight coefficient and a safety optimization weight coefficient are constructed to adjust the optimization weights of both dynamics and safety through the dynamic optimization weight coefficient and the safety optimization weight coefficient. Further, in this embodiment, the dynamic optimization weight coefficient, the safety optimization weight coefficient, the dynamic optimization objective function and the safety optimization objective function are combined through the cost function to obtain a multi-objective coordinated optimization objective function. Then, in this embodiment, a power damping coefficient function is constructed by constructing a power damping coefficient that is positively correlated with the actual demand power change rate of the wheel to achieve adaptive adjustment of the working condition. At the same time, safety constraints are constructed according to engine parameters and battery parameters. Then, taking the safety constraint condition as the constraint condition, the multi-objective coordinated optimization objective function and the required power damping coefficient function are solved to obtain the control parameters including the engine target speed, engine torque and required power damping coefficient, namely the control sequence matrix. The control sequence matrix is then applied to vehicle control to effectively improve the driving responsiveness and high-voltage safety of hybrid vehicles.

According to some embodiments of the present invention, the training of the vehicle speed prediction model based on the historical vehicle speed information includes:

Normalizing the historical vehicle speed information to obtain normalized historical vehicle speed information;

Dividing the normalized historical vehicle speed information into a training set, a validation set, and a test set;

Construct NAR neural network;

The NAR neural network is trained by the LM algorithm according to the training set, the validation set and the test set to obtain the vehicle speed prediction model.

According to some embodiments of the present invention, constructing a safety optimization objective function according to the required power prediction value includes:

Constructing a high voltage system safety margin coefficient according to the demand power forecast value;

According to the electric power of the range extender, the stall risk safety factor of the range extender system is established;

The safety optimization objective function is constructed based on the high-voltage system safety margin coefficient and the extended-range system stall risk safety factor.

According to some embodiments of the present invention, the safety constraints include: range-extended system constraints and power battery constraints;

The safety constraint conditions are constructed according to the engine parameters and the battery parameters, including:

According to the engine speed, engine torque and engine torque change rate, the range-extending system constraint conditions are established;

The power battery constraint conditions are constructed according to the power battery voltage, power battery current and power battery discharge power.

According to some embodiments of the present invention, solving the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraint condition to obtain a control sequence matrix includes:

The multi-objective coordinated optimization objective function and the required power damping coefficient function are solved according to the safety constraint conditions by using a DP dynamic programming algorithm to obtain a control sequence matrix.

According to some embodiments of the present invention, after executing the step of solving the multi-objective coordinated optimization objective function and the required power damping coefficient function by the DP dynamic programming algorithm according to the safety constraint condition to obtain a control sequence matrix, the method further includes:

The first control vector of the control sequence matrix is applied to the vehicle control system.

According to some embodiments of the present invention, the constructing of a dynamics optimization weight coefficient and a safety optimization weight coefficient comprises:

The power optimization weight coefficient and the safety optimization weight coefficient are constructed according to the accelerator pedal opening.

On the other hand, an embodiment of the present invention further provides a hybrid vehicle energy management control system, comprising:

A vehicle speed prediction model building module is used to obtain historical vehicle speed information of the vehicle and train a vehicle speed prediction model based on the historical vehicle speed information;

A speed sequence prediction module, used for inputting the historical vehicle speed information and the real-time vehicle speed information into the vehicle speed prediction model to predict and obtain a predicted speed sequence;

A power demand prediction module, used to predict the power demand of the vehicle according to the predicted speed sequence to obtain a power demand prediction value;

A dynamic performance optimization module, used to construct a dynamic performance optimization objective function according to the required power prediction value through the range extender power following error;

A safety optimization module, used to construct a safety optimization objective function according to the required power prediction value;

A weight coefficient construction module is used to construct a dynamic optimization weight coefficient and a safety optimization weight coefficient;

A multi-objective coordinated optimization module, used for combining the dynamics optimization weight coefficient, the safety optimization weight coefficient, the dynamics optimization objective function and the safety optimization objective function through a cost function to obtain a multi-objective coordinated optimization objective function;

A demand power damping coefficient building module is used to build a demand power damping coefficient function;

Safety constraint module, used to build safety constraint conditions based on engine parameters and battery parameters;

The control sequence solving module is used to solve the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraint conditions to obtain a control sequence matrix; the control sequence matrix includes the engine target speed, the engine torque and the required power damping coefficient.

On the other hand, an embodiment of the present invention further provides a hybrid vehicle energy management control device, comprising:

at least one processor;

at least one memory for storing at least one program;

When the at least one program is executed by the at least one processor, the at least one processor implements the hybrid vehicle energy management control method as described in the above embodiment.

On the other hand, an embodiment of the present invention further provides a computer storage medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the hybrid vehicle energy management control method as described in the above embodiment.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a hybrid vehicle energy management control method provided by an embodiment of the present invention;

FIG. 2 is a functional block diagram of a hybrid vehicle energy management control device provided by an embodiment of the present invention.

DETAILED DESCRIPTION

The embodiments described in the embodiments of this application should not be regarded as limitations of this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it can be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

In the future automobile industry, new energy vehicles have broad prospects. As an important part of new energy vehicles, hybrid vehicles have a major advantage of coordinated and complementary multiple power sources. The existing vehicle energy management strategies in the model can be mainly divided into four types: rule-based, fuzzy rule-based, global optimization-based, and instantaneous optimization-based. In existing research and related results, whether it is a rule-based or optimization-based control strategy, its control objectives are mostly based on improving fuel economy. However, due to the relatively single optimization objective, when dealing with off-road conditions with large longitudinal vehicle speed changes and strong nonlinear changes in required power, the robustness of existing control strategies is generally poor, and it is difficult to achieve global optimization of drive responsiveness and vehicle safety.

Based on this, an embodiment of the present invention provides a hybrid vehicle energy management control method, which can improve the driving responsiveness and high-voltage safety of the hybrid vehicle. Referring to FIG1 , the method of the embodiment of the present invention includes but is not limited to step S110, step S120, step S130, step S140, step S150, step S160, step S170, step S180, step S190 and step S1010.

Specifically, the method application process of the embodiment of the present invention includes but is not limited to the following steps:

S110: Obtain historical vehicle speed information of the vehicle, and train a vehicle speed prediction model based on the historical vehicle speed information.

S120: Inputting historical vehicle speed information and real-time vehicle speed information into a vehicle speed prediction model to predict a predicted speed sequence.

S130: Predicting the required power of the vehicle according to the predicted speed sequence to obtain a predicted required power value.

S140: According to the demand power prediction value, a dynamic optimization objective function is constructed through the range extender power following error.

S150: Construct a safety optimization objective function according to the demand power prediction value.

S160: Constructing dynamic optimization weight coefficient and safety optimization weight coefficient.

S170: The multi-objective coordinated optimization objective function is obtained by combining the dynamics optimization weight coefficient, the safety optimization weight coefficient, the dynamics optimization objective function and the safety optimization objective function through the cost function.

S180: Constructing a required power damping coefficient function.

S190: Construct safety constraints based on engine parameters and battery parameters.

S1010: Solve the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraint conditions to obtain a control sequence matrix, wherein the control sequence matrix includes the engine target speed, the engine torque and the required power damping coefficient.

During the operation of this specific embodiment, this embodiment first obtains the historical speed information of the vehicle, and then trains the historical speed information to obtain a speed prediction model. For example, a neural network is trained based on the historical speed information of the vehicle to obtain a speed prediction model that can predict the vehicle speed. Then, this embodiment predicts the speed of the vehicle at the next moment by inputting the historical speed information and the real-time speed information of the vehicle into the trained speed prediction model, and predicts the speed of the vehicle at the next moment by the speed prediction model, thereby obtaining a predicted speed sequence. For example, the speed of the vehicle at the next moment is predicted by the speed prediction model, and the speed prediction within a fixed time domain p in the future is achieved by recursive calculation p times, thereby obtaining a predicted speed sequence. Further, this embodiment predicts the required power of the vehicle by predicting the speed sequence to obtain a required power prediction value. This embodiment calculates the required power prediction value by predicting the relationship between the speed sequence and the required power. Specifically, the calculation method of this embodiment for predicting the required power of the vehicle by predicting the speed sequence is shown in the following formula (1):

Among them, m represents the mass of the vehicle, v represents the real-time speed of the vehicle, represents the predicted value of required power, represents the vehicle speed prediction value in the prediction speed sequence, and _Preq (k-1) represents the actual required power at the previous moment.

Then, according to the obtained demand power forecast value, this embodiment constructs a dynamic optimization objective function through the range extender power following error. Specifically, the dynamic optimization objective function constructed in this embodiment is shown in the following formula (2):

J _p (k)=[P _G (k)-P _{G_d} (k)] ² (2)

Among them, the parameters of formula (2) are shown in formula (3):

Among them, in the above formulas (2) and (3), p is the prediction time domain, _PG (k) is the actual output power of the range extender at the current moment, _{PG_d} (k) is the expected output power of the range extender, _PB is the real-time charge/discharge power of the battery, _IB is the charge/discharge current of the power battery (discharge is a positive value), _VB is the output voltage of the power battery, _ηB is the battery charge/discharge efficiency, _ηB is mainly affected by SOC, current, temperature, temperature difference and pressure difference of each monomer, and can be obtained in real time by table lookup interpolation method. _ηG is the output efficiency of the range extender, which can be calculated in real time by engine speed _nE , generator torque _TG , range extender voltage _VG and current _IG .

Furthermore, the present embodiment constructs a safety optimization objective function through the demand power prediction value. Then, the present embodiment adjusts the optimization weights of both dynamics and safety through the dynamics optimization weight coefficient and the safety optimization weight coefficient. At the same time, the present embodiment combines the dynamics optimization weight coefficient, the safety optimization weight coefficient, the dynamics optimization objective function and the safety optimization objective function through the cost function, thereby obtaining a multi-objective coordinated optimization objective function. Further, the present embodiment constructs a demand power damping coefficient function to achieve adaptive adjustment of operating conditions through a demand power damping coefficient _CP that is positively correlated with the actual rate of change of the vehicle's demand power. For example, under transient conditions, appropriately reducing _CP can effectively constrain the peak discharge rate of the battery and improve safety. Accordingly, under conditions with a higher safety margin, increasing _CP can optimize the corresponding performance of the drive system. Specifically, the demand power damping coefficient function of the present embodiment is shown in the following equations (4) and (5):

P _req (k)＝P _req (k-1)+C _p (α _Acc (k)P _{m_Max} (k)-P _req (k-1)) (4)

Among them, P _{m_Max} in equations (4) and (5) is the maximum required power of the hub motor, C _{P_high} is the maximum value of the damping coefficient C _P , α _ACC is the accelerator pedal opening, w _S is the safety optimization weight coefficient, and _Preq (k) is the actual required power of the vehicle.

Furthermore, the present embodiment constructs safety constraints through engine parameters and battery parameters. The present embodiment uses safety constraints to alleviate the performance deterioration and safety issues of the control system caused by the state variables and control variables being greater than the performance limitations of the power components in the process of solving the optimal control vector. Then, the present embodiment solves the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the constructed safety constraints to obtain a control sequence matrix. Among them, the control sequence matrix includes the engine target speed, the engine torque, and the required power damping coefficient. The present embodiment effectively improves the driving responsiveness and high-voltage safety of the hybrid vehicle by applying the control sequence matrix to vehicle control.

In some embodiments of the present invention, the vehicle speed prediction model is obtained by training according to historical vehicle speed information, including but not limited to the following steps:

The historical vehicle speed information is normalized to obtain normalized historical vehicle speed information.

The normalized historical vehicle speed information is divided into a training set, a validation set, and a test set.

Construct a NAR neural network.

The NAR neural network is trained using the LM algorithm based on the training set, validation set, and test set to obtain a vehicle speed prediction model.

In this specific embodiment, after obtaining the historical speed information of the vehicle, the historical speed information is first normalized to obtain the normalized historical speed information. Specifically, the normalization process of the historical speed information in this embodiment is shown in the following formula (7):

Where v is a single historical vehicle speed information, It is the normalized single historical vehicle speed information, v _max is the maximum historical speed, and v _min is the minimum historical speed.

Then, this embodiment divides the normalized historical vehicle speed information into a training set, a validation set, and a test set. Exemplarily, this embodiment divides the normalized historical vehicle speed information into a training set, a validation set, and a test set at a ratio of 70%:15%:15%. Furthermore, this embodiment constructs a NAR (nonlinear autoregressive) neural network. This embodiment trains the NAR neural network to obtain a vehicle speed prediction model that can predict the vehicle speed and acceleration in a fixed time domain in the future. In this embodiment, the NAR neural network includes an input layer, a hidden layer, and an output layer. Specifically, the hidden layer substitutes the historical state quantity y(ti) of the input layer, i.e., x _i , the weight value w between each hidden layer neuron, and the threshold b into the activation function f() to obtain the output quantity H of the hidden layer neuron as shown in the following formula (8):

Among them, w _ij in the formula represents the weight value of the mapping relationship between the ith input and the jth neuron in the hidden layer, and b represents the threshold of the hidden layer neuron. Selecting the S-type growth function tansig as the activation function of the hidden layer neuron constrains the output range to [-1,1].

Accordingly, the output layer operation method is shown in the following formula (9):

Among them, _wj is the weight value of the mapping relationship between the jth neuron _Hj in the hidden layer and the neurons in the output layer, b is the threshold of the neurons in the output layer, and the output layer uses a linear transmission activation function.

Furthermore, this embodiment trains the NAR neural network through the LM (Levenberg-Marquardt) algorithm according to the training set, the validation set and the test set to obtain a vehicle speed prediction model. Specifically, this embodiment preliminarily determines the number of hidden layer neurons n _h of the NAR neural network through an empirical formula as shown in the following formula (10):

Wherein, m is the number of samples in the data set, α ₁ and α ₂ are constants in the interval [1,10], and n _i and n _o represent the number of neurons in the input layer and the output layer, respectively. In this embodiment, the LM algorithm is used as the training algorithm, and the mean square error value of the test set under different numbers of hidden layer neurons n _h and delay orders k _d (reference test value range 1-30) is obtained through offline training and testing, and the optimal value is obtained.

In some embodiments of the present invention, a safety optimization objective function is constructed according to the demand power prediction value, including but not limited to the following steps:

Construct the safety margin factor of the high voltage system based on the predicted value of demand power.

According to the electric power of the range extender, a safety factor for the stall risk of the range extender system is established.

According to the safety margin coefficient of the high-voltage system and the stall risk safety factor of the extended-range system, a safety optimization objective function is constructed.

In this specific embodiment, this embodiment first constructs the high-voltage system safety margin coefficient according to the demand power prediction value, and then constructs the extended-range system stall risk safety factor through the electric power of the range extender. Further, a safety optimization objective function is constructed according to the high-voltage system safety margin coefficient and the extended-range system stall risk safety factor. Specifically, this embodiment first predicts the discharge load of the battery at a future moment according to the demand power prediction value, and then predicts the degree of over-discharge of the battery at a future moment by constructing the high-voltage system safety margin coefficient. Among them, the high-voltage system safety margin coefficient is shown in the following formula (11):

Wherein, α _S represents the safety margin coefficient of the high-voltage system, η _{B_min} (k-1) is the minimum value of the battery discharge efficiency at the previous moment, _{PB_Max} (k-1) is the battery discharge power safety threshold at the previous moment, and _PG (k-1) is the actual output power of the range extender at the previous moment.

Furthermore, in this embodiment, the stall risk safety factor of the range extender system is constructed by the electric energy power of the range extender. Specifically, the stall risk safety factor of the range extender system Eng _s is shown in the following formula (12):

Wherein, _{PG_Max} ( _nE (k)) represents the electric power that the range extender can stably output at the current speed.

Then, this embodiment constructs a safety optimization objective function through real-time calculation of α _S and Eng _S as shown in the following formula (13):

J _s (k) = [(α _S (k)) ² + (Eng _s (k)) ² ] (13)

It should be noted that in some embodiments of the present invention, the dynamic optimization weight coefficient, the safety optimization weight coefficient, the dynamic optimization objective function and the safety optimization objective function are combined through the cost function to obtain a multi-objective coordinated optimization objective function. Specifically, the constructed multi-objective coordinated optimization objective function is shown in the following formula (14):

J ^* (k)＝w _p (k)J _p (k)+w _S (k)J _S (k) (14)

Among them, w _P represents the power optimization weight coefficient, and w _S represents the safety optimization weight coefficient.

In some embodiments of the present invention, the safety constraints include range-extending system constraints and power battery constraints. Accordingly, in this embodiment, the safety constraints are constructed based on the engine parameters and the battery parameters, including but not limited to the following steps:

The range extender system constraints are constructed based on the engine speed, engine torque and engine torque change rate.

The power battery constraint conditions are constructed according to the power battery voltage, power battery current and power battery discharge power.

In this specific embodiment, the safety constraints constructed in this embodiment include range-extended system constraints and power battery constraints. Accordingly, this embodiment constructs range-extended system constraints through engine speed, engine torque, and engine torque change rate. In addition, this embodiment constructs power battery constraints through power battery voltage, power battery current, and power battery discharge power. Specifically, this embodiment imposes corresponding restrictions on the state variables and control variables to be solved by constructing range-extended system constraints and power battery constraints, thereby alleviating the state variables and control variables being greater than the performance limitations of the power components during the solution of the optimal control vector, thereby causing control system performance deterioration and safety issues. Among them, the range-extended system constraints are shown in the following formula (15):

Among them, n _{E_max} is the upper limit of engine speed, n _{E_min} is the lower limit of engine speed; T _{G_max} is the upper limit of generator torque, T _{G_min} is the lower limit of generator torque; is the rate of change of the engine torque _TG , _{TG_Max} ( _nE ) represents the torque that the engine can stably output at the current speed, and the constraint is constructed so that Do not exceed the current response capability of the engine d( _{TG_Max} ( _nE ))/d( _nE ).

In addition, the power battery constraint condition is shown in the following formula (16):

Among them, in the above formula, _{VB_max} is the maximum terminal voltage of the power battery, _{VB_min} is the minimum terminal voltage of the power battery; _{IB_max} is the upper limit of the real-time charge/discharge current allowed by the power battery, _{IB_min} is the lower limit of the real-time charge/discharge current allowed by the power battery; _{PB_max} is the upper limit of the power battery discharge power, and _{PB_min} is the lower limit of the power battery discharge power.

In some embodiments of the present invention, solving the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraint conditions to obtain the control sequence matrix includes but is not limited to the following steps:

According to the safety constraints, the multi-objective coordinated optimization objective function and the required power damping coefficient function are solved by the DP dynamic programming algorithm to obtain the control sequence matrix.

In this specific embodiment, this embodiment provides a DP dynamic programming algorithm combined with safety constraints to solve the multi-objective coordinated optimization objective function and the required power damping coefficient function to obtain a control sequence matrix. Specifically, under the safety constraints, the solution is performed by the DP dynamic programming algorithm. This embodiment regards the calculation and solution process as several related sub-processes, and sequentially calculates the optimal control variables of each sub-process to obtain the optimal control sequence within a single prediction time domain. Exemplarily, for the multi-objective coordinated optimization objective function, its inverse solution process is shown in the following equations (17) to (19):

J ^* (k+p)＝min{w _p (k)J _p (k+p)+w _S (k)J _s (k+p)} (17)

J ^* (k+p-1)＝min{w _p (k)J _p (k+p-1)+w _S (k)J _S (k+p-1)+J ^* (k+p)} (18)

J ^* (k)＝min{w _p (k)J _p (k)+w _S (k)J _S (k)+J ^* (k+1)} (19)

The optimal solution of the multi-objective coordinated optimization objective function in the prediction time domain p is expressed as follows:

In some embodiments of the present invention, after performing the step of solving the multi-objective coordinated optimization objective function and the required power damping coefficient function by the DP dynamic programming algorithm according to the safety constraints to obtain the control sequence matrix, the hybrid vehicle energy management control method provided by the embodiment of the present invention further includes but is not limited to the following steps:

The first control vector of the control sequence matrix is applied to the vehicle control system.

In this specific embodiment, after solving the multi-objective coordinated optimization objective function and the required power damping coefficient by the DP dynamic programming algorithm under the safety constraint condition, the control sequence matrix u ^* (k) is obtained. The control sequence matrix includes the optimal control vectors at each moment in the prediction time domain p. In this embodiment, the first control vector u ^* (k) = [n _{E_d} T _G C _p ] in the control sequence matrix is used as the optimal control vector at the current moment, and is applied to the vehicle control system to achieve the power distribution at the current moment, thereby achieving the multi-objective optimization of the robustness, power responsiveness and safety of the vehicle power system.

In some embodiments of the present invention, constructing a dynamic optimization weight coefficient and a safety optimization weight coefficient includes but is not limited to the following steps:

According to the accelerator pedal opening, the dynamic optimization weight coefficient and the safety optimization weight coefficient are constructed.

In this specific embodiment, this embodiment constructs a dynamic optimization weight coefficient and a safety optimization weight coefficient through the accelerator pedal opening value. Specifically, this embodiment constructs a dynamic optimization weight coefficient and a safety optimization weight coefficient through the accelerator pedal opening that can more accurately reflect the driving intention, as shown in the following formulas (21), (22) and (23):

Among them, in the above formula (21) to formula (23), α _I and is the accelerator pedal characterization coefficient; are the upper and lower limits of α _I , is the lower limit of α _I ; C ₁ and C ₂ are the fitting coefficients of the accelerator pedal and the accelerator pedal change rate, respectively; α _ACC is the accelerator pedal opening; is the rate of change of accelerator pedal opening.

An embodiment of the present invention further provides a hybrid vehicle capability management control system, the system comprising:

The vehicle speed prediction model building module is used to obtain the vehicle's historical speed information and train a vehicle speed prediction model based on the historical speed information.

The predicted speed sequence module is used to input historical vehicle speed information and real-time vehicle speed information into the vehicle speed prediction model to predict the predicted speed sequence.

The power demand prediction module is used to predict the power demand of the vehicle according to the predicted speed sequence to obtain the power demand prediction value.

The dynamic optimization module is used to construct the dynamic optimization objective function according to the required power prediction value through the range extender power following error.

The safety optimization module is used to construct a safety optimization objective function according to the demand power prediction value.

The weight coefficient construction module is used to construct the dynamic optimization weight coefficient and the safety optimization weight coefficient.

The multi-objective coordinated optimization module is used to obtain the multi-objective coordinated optimization objective function by combining the dynamic optimization weight coefficient, the safety optimization weight coefficient, the dynamic optimization objective function and the safety optimization objective function through the cost function.

The demand power damping coefficient building module is used to build the demand power damping coefficient function.

The safety constraint module is used to construct safety constraint conditions based on engine parameters and battery parameters.

The control sequence solving module is used to solve the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraints to obtain the control sequence matrix. The control sequence matrix includes the engine target speed, engine torque and required power damping coefficient.

2, an embodiment of the present invention further provides a hybrid vehicle capability management control device, the device comprising:

At least one processor 210 .

At least one memory 220 is used to store at least one program.

When the at least one program is executed by the at least one processor 210, the at least one processor 210 implements the hybrid vehicle energy management control method as described in the above embodiment.

An embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions. The computer-executable instructions are executed by one or more control processors, for example, to execute the steps described in the above embodiment.

It will be appreciated by those skilled in the art that all or some of the steps and systems in the disclosed method above may be implemented as software, firmware, hardware and appropriate combinations thereof. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that may be used to store desired information and may be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the above-mentioned implementation mode. Technical personnel familiar with the field can also make various equivalent deformations or substitutions without violating the spirit of the present invention. These equivalent deformations or substitutions are all included in the scope defined by the claims of the present invention.

Claims (10)

1. A hybrid vehicle energy management control method characterized by comprising the steps of:

acquiring historical vehicle speed information of a vehicle, and training according to the historical vehicle speed information to obtain a vehicle speed prediction model;

inputting the historical vehicle speed information and the real-time vehicle speed information into the vehicle speed prediction model, and predicting to obtain a predicted speed sequence;

predicting the required power of the vehicle according to the predicted speed sequence to obtain a predicted value of the required power;

according to the predicted value of the required power, constructing a dynamic optimization objective function through a range extender power following error;

constructing a safety optimization objective function according to the predicted value of the required power;

Constructing a dynamic optimization weight coefficient and a safety optimization weight coefficient;

The dynamic optimization weight coefficient, the safety optimization weight coefficient, the dynamic optimization objective function and the safety optimization objective function are combined through a cost function to obtain a multi-objective coordinated optimization objective function;

Constructing a required power damping coefficient function;

constructing safety constraint conditions according to the engine parameters and the battery parameters;

Solving the multi-objective coordination optimization objective function and the required power damping coefficient function according to the safety constraint condition to obtain a control sequence matrix; the control sequence matrix includes an engine target speed, an engine torque, and a demand power damping coefficient.

2. The hybrid vehicle energy management control method according to claim 1, wherein the training to obtain a vehicle speed prediction model from the historical vehicle speed information includes:

Normalizing the historical vehicle speed information to obtain normalized historical vehicle speed information;

Dividing the normalized historical vehicle speed information into a training set, a verification set and a test set;

constructing an NAR neural network;

and training the NAR neural network through an LM algorithm according to the training set, the verification set and the test set to obtain the vehicle speed prediction model.

3. The hybrid vehicle energy management control method according to claim 1, wherein the constructing a safety optimization objective function from the required power prediction value includes:

Constructing a safety margin coefficient of the high-voltage system according to the predicted value of the required power;

constructing stall risk safety coefficients of the range-increasing system according to the electric energy power of the range-increasing device;

And constructing and obtaining the safety optimization objective function according to the safety margin coefficient of the high-voltage system and the stall risk safety coefficient of the range-extending system.

4. The hybrid vehicle energy management control method according to claim 1, characterized in that the safety constraint condition includes: extended range system constraint conditions and power battery constraint conditions;

The construction of the safety constraint condition according to the engine parameter and the battery parameter comprises the following steps:

constructing constraint conditions of a range-extending system according to the engine speed, the engine torque and the engine torque change rate;

and constructing a power battery constraint condition according to the power battery voltage, the power battery current and the power battery discharge power.

5. The hybrid vehicle energy management control method of claim 1, wherein the solving the multi-objective coordinated optimization objective function and the required power damping coefficient function according to the safety constraint condition results in a control sequence matrix, comprising:

And solving the multi-objective coordination optimization objective function and the required power damping coefficient function through a DP dynamic programming algorithm according to the safety constraint condition to obtain a control sequence matrix.

6. The hybrid vehicle energy management control method according to claim 5, further comprising, after performing the step of solving the multi-objective coordinated optimization objective function and the required power damping coefficient function by a DP dynamic planning algorithm according to the safety constraint condition to obtain a control sequence matrix:

A first control vector of the control sequence matrix is applied to a vehicle control system.

7. The hybrid vehicle energy management control method according to claim 1, characterized in that the constructing a dynamic optimization weight coefficient and a safety optimization weight coefficient includes:

And constructing the dynamic optimization weight coefficient and the safety optimization weight coefficient according to the opening degree of the accelerator pedal.

8. A hybrid vehicle energy management control system, comprising:

the vehicle speed prediction model construction module is used for acquiring historical vehicle speed information of the vehicle and training according to the historical vehicle speed information to obtain a vehicle speed prediction model;

The predicted speed sequence module is used for inputting the historical speed information and the real-time speed information into the speed prediction model and predicting to obtain a predicted speed sequence;

The demand power prediction module is used for predicting the demand power of the vehicle according to the predicted speed sequence to obtain a demand power predicted value;

The dynamic optimization module is used for constructing a dynamic optimization objective function through a range extender power following error according to the demand power predicted value;

The safety optimization module is used for constructing a safety optimization objective function according to the predicted value of the required power;

the weight coefficient construction module is used for constructing a dynamic optimization weight coefficient and a safety optimization weight coefficient;

The multi-objective coordination optimization module is used for obtaining a multi-objective coordination optimization objective function by combining the dynamic optimization weight coefficient, the safety optimization weight coefficient, the dynamic optimization objective function and the safety optimization objective function through a cost function;

the required power damping coefficient construction module is used for constructing a required power damping coefficient function;

The safety constraint module is used for constructing safety constraint conditions according to the engine parameters and the battery parameters;

the control sequence solving module is used for solving the multi-objective coordination optimization objective function and the required power damping coefficient function according to the safety constraint condition to obtain a control sequence matrix; the control sequence matrix includes an engine target speed, an engine torque, and a demand power damping coefficient.

9. A hybrid vehicle energy management control apparatus, characterized by comprising:

At least one processor;

At least one memory for storing at least one program;

the at least one program, when executed by the at least one processor, causes the at least one processor to implement the hybrid vehicle energy management control method of any one of claims 1 to 7.

10. A computer storage medium in which a processor-executable program is stored, characterized in that the processor-executable program is for realizing the hybrid vehicle energy management control method according to any one of claims 1 to 7 when executed by the processor.