CN114889581A - Hybrid vehicle control method and device - Google Patents

Abstract

A hybrid vehicle control method and apparatus are provided. The method comprises the steps of obtaining a vehicle travel sample set, wherein the vehicle travel sample set comprises a plurality of vehicle travel samples and label data corresponding to each vehicle travel sample, the vehicle travel samples represent vehicle historical travel data of a vehicle in a historical travel stage, and the label data represent an optimal power distribution strategy of the vehicle in the historical travel stage; training an initial power distribution neural network model by using a vehicle travel sample set to obtain a target power distribution neural network model; establishing a vehicle speed prediction model according to vehicle historical driving data corresponding to an optimal power distribution strategy in a historical driving stage; determining a predicted vehicle speed at the predicted time point according to the predicted time point and the vehicle speed prediction model; and inputting the vehicle speed and the predicted vehicle speed in the current driving stage into a target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, and controlling the vehicle according to the target power distribution strategy.

Description

Hybrid vehicle control method and device

technical field

The present disclosure relates to the technical field of vehicle control, and more particularly, to a hybrid vehicle control method, a hybrid vehicle control device, an electronic device, a computer-readable storage medium, and a computer program product.

Background technique

With the increasingly severe world energy crisis and the aggravation of environmental pollution, people pay more and more attention to efficient and clean power systems, and hybrid vehicles have become a trend in vehicle development. The hybrid vehicle adds an electric drive system to the traditional fuel vehicle power system. The main components of the electric drive include a battery, a generator and a drive motor. Through the speed and torque adjustment function of the motor, the probability of the engine working in the high-efficiency area is increased. Through the energy storage function of the battery, energy recovery can also be carried out for individual working conditions, which greatly increases the fuel economy.

Since the power system of the hybrid vehicle includes not only the electric drive system, but also the power system of the fuel vehicle, in this case, if the power of the engine and the electric motor of the hybrid vehicle cannot be reasonably distributed, the fuel economy of the vehicle will be reduced.

SUMMARY OF THE INVENTION

In view of this, embodiments of the present disclosure provide a hybrid vehicle control method, a hybrid vehicle control device, an electronic device, a computer-readable storage medium, and a computer program product.

One aspect of the embodiments of the present disclosure provides a hybrid vehicle control method, including:

obtaining a vehicle trip sample set, wherein the vehicle trip sample set includes a plurality of vehicle trip samples and label data corresponding to each of the vehicle trip samples, and the vehicle trip samples represent the vehicle historical driving data of the vehicle in the historical driving stage, The above-mentioned vehicle historical driving data includes a speed set corresponding to the above-mentioned vehicle in different time periods in the above-mentioned historical driving stage, and the above-mentioned vehicle speed set includes a first vehicle speed and a second vehicle speed, wherein, the above-mentioned first vehicle speed and the above-mentioned second vehicle speed are in the same time period. The speed of the vehicle with a preset time interval in the middle, and the above-mentioned label data represents the optimal power distribution strategy of the above-mentioned vehicle in the above-mentioned historical driving stage;

The initial power distribution neural network model is trained by using the vehicle trip sample set to obtain the target power distribution neural network model, wherein the initial power distribution neural network model is established according to the test cycle optimal power distribution data corresponding to the test cycle data. The test cycle data represents the vehicle driving data of the above-mentioned vehicle under standard operating conditions;

According to the historical driving data of the vehicle corresponding to the above-mentioned optimal power distribution strategy in the historical driving stage, a vehicle speed prediction model is established;

According to the predicted time point and the above-mentioned vehicle speed prediction model, determine the predicted vehicle speed at the above-mentioned predicted time point;

The vehicle speed in the current driving stage and the predicted vehicle speed are input into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, so as to control the vehicle according to the target power distribution strategy.

According to an embodiment of the present disclosure, the optimal power distribution strategy in the above-mentioned historical driving phase is determined by the following steps:

Use dynamic programming algorithm to process the above-mentioned vehicle historical driving data in the above-mentioned historical driving stage, and obtain the optimal power distribution strategy in the above-mentioned historical driving stage, wherein, the above-mentioned optimal power distribution strategy includes: engine output torque, generator output torque and electric motor Optimal value of output torque.

According to an embodiment of the present disclosure, the above-mentioned initial power distribution neural network model is established according to the test cycle optimal power distribution data corresponding to the test cycle data, including:

Use the dynamic programming algorithm to process the above-mentioned test cycle data, and obtain the above-mentioned test cycle optimal power distribution data;

The above-mentioned initial power distribution neural network model is established according to the above-mentioned test cycle data and the above-mentioned test cycle optimal power distribution data.

According to an embodiment of the present disclosure, the above-mentioned initial power distribution neural network model includes a BP neural network model, and the above-mentioned initial power distribution neural network model includes an input layer, at least one hidden layer, and an output layer.

According to an embodiment of the present disclosure, the historical vehicle driving data in the historical driving stage further includes the total vehicle mileage shown in the first formula, and the vehicle status data in the current driving stage includes the vehicle driving mileage in the current driving stage, Real-time vehicle speed and demand power, the calculation of vehicle mileage in the current driving phase is shown in the second formula:

Among them, Dis _n is the total mileage of the vehicle in the historical driving stage, Dis _n-1 is the total reference mileage of the vehicle in the driving stage before the historical driving stage, Dis _cur is the total reference mileage of the vehicle in the historical driving stage, and m is the historical travel data The update step size of , L _n is the mileage of the vehicle in the current driving stage, and L _n-1 is the mileage of the vehicle in the historical driving stage.

According to an embodiment of the present disclosure, the target power distribution strategy includes: an output torque signal of an engine of the vehicle, an output torque signal of a generator, and an output torque signal of a drive motor, wherein the generator is used to convert mechanical energy into electrical energy;

The above-mentioned historical driving data of the vehicle also includes: the remaining power of the vehicle battery, the gear signal, the accelerator signal and the brake signal.

According to an embodiment of the present disclosure, the above-mentioned vehicle speed prediction model is established according to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the historical driving stage, including:

Determine the vehicle speed state matrix according to the optimal power distribution strategy in the above historical driving stages;

According to the above vehicle speed state matrix and vehicle speed frequency matrix, determine the state transition probability matrix;

The vehicle speed prediction model is determined according to the state transition probability matrix, wherein the vehicle speed prediction model includes a Markov chain vehicle speed prediction model.

According to the embodiment of the present disclosure, the above-mentioned training of the initial power distribution neural network model by using the above-mentioned vehicle trip sample set to obtain the target power distribution neural network model includes:

Inputting the above-mentioned vehicle driving sample into the above-mentioned initial power distribution neural network model, and outputting a predicted power distribution strategy, wherein the above-mentioned predicted power distribution strategy includes the predicted output torque of the engine, the output torque of the generator, and the output torque of the drive motor;

According to the above prediction power allocation strategy and the above label data calculation function, the loss result is obtained;

The parameters of the initial power distribution neural network model are iteratively adjusted according to the loss results to generate the target power distribution neural network model.

According to an embodiment of the present disclosure, the above method further includes:

In the case that the mileage of the vehicle in the current driving stage satisfies the preset update step size, obtain the optimal power distribution strategy for the current driving stage after the update step size;

The optimal power allocation strategy of the current driving stage after the above update step size and the corresponding historical driving data of the vehicle are used as new label data and new vehicle travel samples to train the above target power allocation neural network model to obtain a new target. Power distribution neural network model;

Establish a new vehicle speed prediction model according to the historical vehicle driving data corresponding to the optimal power distribution strategy in the above-mentioned current driving stage;

According to the new predicted time point and the above-mentioned new vehicle speed prediction model, determine the new predicted vehicle speed at the new predicted time point in the future driving stage;

Input the vehicle speed in the above-mentioned future driving stage and the above-mentioned new predicted vehicle speed into the above-mentioned new target power distribution neural network model, and obtain a new target power distribution strategy corresponding to the above-mentioned new predicted vehicle speed. Control the above vehicle.

Another aspect of the embodiments of the present disclosure provides a hybrid vehicle control device, including:

The acquiring module is configured to acquire a vehicle trip sample set, wherein the vehicle trip sample set includes a plurality of vehicle trip samples and label data corresponding to each of the vehicle trip samples, and the vehicle trip samples represent the history of the vehicle in the historical driving stage. Vehicle historical driving data, the above-mentioned vehicle historical driving data includes the vehicle speed set corresponding to different time periods of the above-mentioned vehicle in the above-mentioned historical driving stage, and the above-mentioned vehicle speed set includes a first vehicle speed and a second vehicle speed, wherein, the above-mentioned first vehicle speed and the above-mentioned second vehicle speed The vehicle speed is the vehicle speed at a preset time interval in the same time period, and the above-mentioned tag data represents the optimal power distribution strategy of the above-mentioned vehicle in the above-mentioned historical driving stage;

A training module for training an initial power distribution neural network model using the above-mentioned vehicle trip sample set to obtain a target power distribution neural network model, wherein the above-mentioned initial power distribution neural network model is based on the test cycle optimal power distribution corresponding to the test cycle data data is established, and the above-mentioned test cycle data represents the vehicle driving data of the above-mentioned vehicle under standard operating conditions;

A module is established for establishing a vehicle speed prediction model according to the historical driving data of the vehicle corresponding to the above-mentioned optimal power distribution strategy in the historical driving stage;

a vehicle speed prediction module for determining the predicted vehicle speed at the above-mentioned predicted time point according to the predicted time point and the above-mentioned vehicle speed prediction model; and

The power distribution prediction module is used to input the vehicle speed of the current driving stage and the predicted vehicle speed into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, so as to control the vehicle according to the target power distribution strategy. .

Another aspect of the embodiments of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more programs Multiple processors, when executed, cause the one or more processors to implement the method as described above.

Another aspect of embodiments of the present disclosure provides a computer-readable storage medium storing computer-executable instructions, which when executed, are used to implement the method as described above.

Another aspect of embodiments of the present disclosure provides a computer program product comprising computer-executable instructions that, when executed, are used to implement the method as described above.

According to the embodiments of the present disclosure, the initial power distribution neural network model is trained by the historical vehicle driving data and the optimal power distribution strategy in the historical driving stage, and the target power distribution neural network model is obtained, and the target power distribution neural network model is used to analyze the current power distribution. The vehicle speed and the predicted vehicle speed in the driving stage are processed, so that the target power distribution strategy corresponding to the predicted vehicle speed can be obtained, and then the vehicle is controlled according to the target power distribution strategy. During the training process, the optimal power distribution strategy in the historical data is learned, so that using the predicted target power distribution strategy to control the vehicle can reduce the economy of the vehicle and solve the problem of poor economy in the use of hybrid vehicles.

Description of drawings

The above and other objects, features and advantages of the present disclosure will become more apparent from the following description of embodiments of the present disclosure with reference to the accompanying drawings, in which:

FIG. 1 schematically shows an exemplary system architecture for applying a hybrid vehicle control method according to an embodiment of the present disclosure;

FIG. 2 schematically shows a flowchart of a hybrid vehicle control method according to an embodiment of the present disclosure;

3 schematically shows a training flow chart of a target power distribution neural network model according to an embodiment of the present disclosure;

FIG. 4 schematically shows a block diagram of a hybrid vehicle control apparatus according to an embodiment of the present disclosure; and

FIG. 5 schematically shows a block diagram of an electronic device implementing a hybrid vehicle control method according to an embodiment of the present disclosure.

Detailed ways

Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood, however, that these descriptions are exemplary only, and are not intended to limit the scope of the present disclosure. In the following detailed description, for convenience of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It will be apparent, however, that one or more embodiments may be practiced without these specific details. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concepts of the present disclosure.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the present disclosure. The terms "comprising", "comprising" and the like as used herein indicate the presence of stated features, steps, operations and/or components, but do not preclude the presence or addition of one or more other features, steps, operations or components.

All terms (including technical and scientific terms) used herein have the meaning as commonly understood by one of ordinary skill in the art, unless otherwise defined. It should be noted that terms used herein should be construed to have meanings consistent with the context of the present specification and should not be construed in an idealized or overly rigid manner.

Where expressions like "at least one of A, B, and C, etc.," are used, they should generally be interpreted in accordance with the meaning of the expression as commonly understood by those skilled in the art (eg, "has A, B, and C") At least one of the "systems" shall include, but not be limited to, systems with A alone, B alone, C alone, A and B, A and C, B and C, and/or A, B, C, etc. ).

Existing hybrid electric vehicle control methods mainly include rule-based control methods, dynamic programming-based algorithms, model-based prediction algorithms, and equivalent fuel consumption minimization methods. At present, these methods have their own advantages and disadvantages. The dynamic programming algorithm can obtain the optimal result but the calculation amount is large and cannot be used for real-time control. Using a single control method cannot satisfy the optimal control for the changeable actual vehicle driving process, which makes it difficult for hybrid vehicles to achieve the best fuel economy.

In view of this, embodiments of the present disclosure provide a hybrid vehicle control method and apparatus. The method may include obtaining a vehicle trip sample set, wherein the vehicle trip sample set may include a plurality of vehicle trip samples and label data corresponding to each vehicle trip sample, the vehicle trip samples representing vehicle historical travel data of the vehicle during the historical travel phase , the label data represents the optimal power distribution strategy of the vehicle in the historical driving stage; the initial power distribution neural network model is trained by the vehicle trip sample set, and the target power distribution neural network model is obtained, wherein the initial power distribution neural network model is based on and tested The optimal power distribution data of the test cycle corresponding to the cycle data is established. The test cycle data represents the vehicle driving data of the vehicle under standard operating conditions; the vehicle speed is established according to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the historical driving stage. Prediction model: According to the prediction time point and the vehicle speed prediction model, determine the predicted vehicle speed at the predicted time point; input the vehicle speed and the predicted vehicle speed of the current driving stage into the target power distribution neural network model, and obtain the target power distribution strategy corresponding to the predicted vehicle speed. The vehicle is controlled according to the target power distribution strategy.

FIG. 1 schematically illustrates an exemplary system architecture 100 to which a hybrid vehicle control method may be applied, according to an embodiment of the present disclosure. It should be noted that FIG. 1 is only an example of a system architecture to which the embodiments of the present disclosure can be applied, so as to help those skilled in the art to understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used for other A device, system, environment or scene.

As shown in FIG. 1 , the system architecture 100 according to this embodiment may include terminal vehicles 101 , 102 , 103 , a network 104 and a server 105 . The network 104 is the medium used to provide the communication link between the end vehicles 101 , 102 , 103 and the server 105 . The network 104 may include various connection types, such as wired and/or wireless communication links, and the like.

The user may use the terminal vehicles 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages and the like. Various vehicle data collection applications may be installed on the terminal vehicles 101 , 102 and 103 . In an exemplary embodiment, the end vehicle 101 may also acquire the target power distribution neural network model on other end vehicles.

The end vehicles 101, 102, 103 may be hybrid vehicles with independent computing devices or transmitting and receiving devices.

The server 105 may be a server that provides various services, such as a background management server (just an example) that provides support for the vehicle history data transmitted by the terminal vehicles 101 , 102 , 103 . The background management server can analyze and process the received vehicle historical data, and feed back the processing results (for example, the target power distribution neural network model generated according to the vehicle historical data, etc.) to the terminal vehicle.

It should be noted that the hybrid vehicle control method provided by the embodiments of the present disclosure may generally be executed by terminal vehicles 101 , 102 , or 103 , or may also be executed by other terminal vehicles different from terminal vehicles 101 , 102 , or 103 . Correspondingly, the hybrid vehicle control system provided by the embodiments of the present disclosure may generally be provided in the terminal vehicles 101 , 102 , or 103 , or in other terminal vehicles different from the terminal vehicles 101 , 102 , or 103 . Alternatively, the hybrid vehicle control method provided by the embodiment of the present disclosure may also be executed by the server 105 and the final result is transmitted to the terminal vehicle 101 , 102 , or 103 . Correspondingly, the hybrid vehicle control system provided by the embodiments of the present disclosure may generally be provided in the server 105 . The hybrid vehicle control method provided by the embodiments of the present disclosure may also be executed by a server or server cluster different from the server 105 and capable of communicating with the terminal vehicles 101 , 102 , 103 and/or the server 105 . Correspondingly, the hybrid vehicle control system provided by the embodiments of the present disclosure may also be provided in a server or server cluster different from the server 105 and capable of communicating with the terminal vehicles 101 , 102 , 103 and/or the server 105 .

It should be understood that the numbers of terminal vehicles, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.

FIG. 2 schematically shows a flowchart of a hybrid vehicle control method according to an embodiment of the present disclosure.

As shown in FIG. 2 , the hybrid vehicle control method may include operations S201 ˜ S205 .

In operation S201, a vehicle trip sample set is obtained, wherein the vehicle trip sample set may include a plurality of vehicle trip samples and label data corresponding to each vehicle trip sample, and the vehicle trip samples represent the vehicle historical driving data of the vehicle in the historical driving stage , the historical driving data of the vehicle may include vehicle speed sets corresponding to different time periods of the vehicle in the historical driving stage, and the vehicle speed set may include a first vehicle speed and a second vehicle speed, wherein the first vehicle speed and the second vehicle speed are preset intervals in the same time period. The vehicle speed is set for a long time, and the label data represents the optimal power distribution strategy of the vehicle in the historical driving stage.

In operation S202, an initial power distribution neural network model is trained using the vehicle trip sample set to obtain a target power distribution neural network model, wherein the initial power distribution neural network model is established according to the test cycle optimal power distribution data corresponding to the test cycle data , the test cycle data represents the vehicle driving data under standard operating conditions.

In operation S203, a vehicle speed prediction model is established according to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the historical driving phase.

In operation S204, a predicted vehicle speed at the predicted time point is determined according to the predicted time point and the vehicle speed prediction model.

In operation S205, the vehicle speed and the predicted vehicle speed in the current driving stage are input into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, so as to control the vehicle according to the target power distribution strategy.

According to an embodiment of the present disclosure, the test cycle data is factory-configured by the vehicle manufacturer, which is data corresponding to the vehicle type measured under standard operating conditions, and the data may include different vehicle speeds corresponding to different vehicle speeds under standard operating conditions. Optimal power allocation strategy.

According to an embodiment of the present disclosure, the optimal power distribution strategy characterizes the distribution scheme of the output power of the generator and the engine in the hybrid vehicle, which distribution scheme can be used to control the operation of the vehicle.

According to an embodiment of the present disclosure, a hybrid vehicle may include a vehicle that uses both fossil fuel and electric energy as energy sources, such as a gasoline-electric hybrid vehicle that is common in the market.

According to an embodiment of the present disclosure, the historical driving stage is different from the current driving stage. For example, when the total vehicle mileage in the current driving stage is 111 kilometers, the historical driving stage may refer to 0-110 kilometers. Preferably, the historical driving stage of the present disclosure The stage can refer to 100-110 kilometers, which can determine the historical driving stage closest to the current driving stage according to the updated step size. The 11 kilometers in the current stage does not meet the update requirements, therefore, preferably, the historical driving stage may be a total vehicle mileage of 80-100 kilometers.

According to an embodiment of the present disclosure, the predicted time point may be a time point separated from the current time by a preset time. For example, if the current time is 0:00:00, and the preset time is 1 second, the predicted time is 0:00 0 minutes and 1 second. It should be noted that the preset duration can be set by the vehicle engineer according to the actual situation.

In one embodiment, a sample set of vehicle trips in a historical travel phase closest to the current travel phase (eg, 100-110 kilometers or 80-100 kilometers as described above) is used to pair the test corresponding to the test cycle data. The initial power distribution neural network model established by circulating the optimal power distribution data is trained, so that the trained target power distribution neural network model can be obtained.

According to the embodiment of the present disclosure, the data of the closest historical driving stage is used as the vehicle trip sample, which enables the target power distribution neural network model to learn the driving characteristics closest to the current driving state of the driver.

According to the embodiment of the present disclosure, the vehicle speed prediction model is established according to the historical driving data, and the vehicle speed can be predicted according to the predicted time point, thereby obtaining the predicted vehicle speed corresponding to the predicted time point.

According to the embodiments of the present disclosure, the vehicle speed at the current driving stage (for example, the vehicle speed at the current moment) and the predicted vehicle speed are input into the trained target power distribution neural network model, so that the target power distribution strategy corresponding to the predicted vehicle speed can be obtained. For example, the total required power output of the vehicle is 110kw, then the target power distribution strategy can be the engine output 80kw and the motor output 30kw. It should be noted that the above-mentioned examples are only used for illustration, and are not intended to limit the present disclosure.

According to the embodiments of the present disclosure, the initial power distribution neural network model is trained by the historical vehicle driving data and the optimal power distribution strategy in the historical driving stage, and the target power distribution neural network model is obtained, and the target power distribution neural network model is used to analyze the current power distribution. The vehicle speed and predicted vehicle speed in the driving stage are processed, so that the target power distribution strategy corresponding to the predicted vehicle speed can be obtained, and then the vehicle can be controlled according to the target power distribution strategy. Since the target power distribution neural network model learns the historical data during the training process Regarding the optimal power distribution strategy, using the predicted target power distribution strategy to control the vehicle can reduce the economy of the vehicle and solve the problem of poor economy in the use of the hybrid vehicle.

According to an embodiment of the present disclosure, the optimal power distribution strategy in the historical driving phase is determined by the following steps:

The dynamic programming algorithm is used to process the historical driving data of the vehicle in the historical driving stage, and the optimal power distribution strategy in the historical driving stage is obtained. optimal value.

According to the embodiments of the present disclosure, in a multi-stage decision problem, the decisions taken at each stage are generally time-related, the decision depends on the current state, and immediately causes the transition of the state. A decision sequence is the state of change. The process of solving multi-stage decision optimization is called dynamic programming method.

According to the embodiment of the present disclosure, after obtaining the historical driving data of the vehicle, the dynamic programming algorithm is used to perform offline optimization calculation on the above data, so that the optimal power allocation strategy corresponding to the historical driving data can be obtained, for example, in the historical driving data The optimal values of engine output torque, generator output torque and motor output torque corresponding to a certain vehicle speed.

According to an embodiment of the present disclosure, the initial power distribution neural network model is established according to the test cycle optimal power distribution data corresponding to the test cycle data, and may include the following operations:

The dynamic programming algorithm is used to process the test cycle data to obtain the optimal power distribution data of the test cycle. The initial power distribution neural network model is established according to the test cycle data and the test cycle optimal power distribution data.

According to the embodiments of the present disclosure, the test cycle data of the factory configuration of the vehicle is processed by the dynamic programming algorithm, so as to obtain the test cycle optimal power distribution data corresponding to the test cycle data, and the data may include the engine corresponding to a certain vehicle speed under standard operating conditions. Optimal values of output torque, generator output torque and motor output torque. The initial power distribution neural network model is established according to the test cycle data obtained above and the test cycle optimal power distribution data.

According to an embodiment of the present disclosure, the initial power distribution neural network model may include a BP neural network model, and the initial power distribution neural network model may include an input layer, at least one hidden layer, and an output layer.

According to an embodiment of the present disclosure, the input layer is used to perform feature extraction on the input vehicle trip samples, the hidden layer performs model training based on the extracted features, and outputs the corresponding predicted power allocation strategy through the output layer. The number of hidden layers may be specifically set according to actual requirements, for example, it may be one layer or two layers.

According to an embodiment of the present disclosure, the historical driving data of the vehicle in the historical driving stage may further include the total driving mileage of the vehicle as shown in formula (1), and the vehicle state data of the current driving stage may include the driving mileage of the vehicle in the current driving stage, real-time driving distance The vehicle speed and required power, and the vehicle mileage in the current driving stage are calculated as shown in formula (2):

Among them, Dis _n is the total mileage of the vehicle in the historical driving stage, Dis _n-1 is the total reference mileage of the vehicle in the driving stage before the historical driving stage, Dis _cur is the total reference mileage of the vehicle in the historical driving stage, and m is the historical travel data The update step size of , L _n is the mileage of the vehicle in the current driving stage, and L _n-1 is the mileage of the vehicle in the historical driving stage.

According to the embodiment of the present disclosure, the value of the update step size can be set by the vehicle engineer according to the specific situation. For example, the update step size mentioned above is 20 kilometers. The purpose of setting the update step size is to set the update step after the vehicle travels for a certain distance. The internal driving data is used as a training sample to update and train the target power distribution neural network model, so that the target power distribution neural network model can learn the latest driving characteristics of the driver in time, so as to obtain a more economical target power distribution strategy.

According to an embodiment of the present disclosure, the target power distribution strategy may include an output torque signal of an engine of the vehicle, a generator output torque signal, and a drive motor output torque signal, wherein the generator is used to convert mechanical energy into electrical energy.

The historical driving data of the vehicle may further include: remaining battery power of the vehicle, gear signal, accelerator signal and brake signal.

According to the embodiments of the present disclosure, part of the power source of the vehicle is in the form of converting the mechanical energy generated by the engine into electrical energy, so that the electric motor outputs corresponding kinetic energy to drive the vehicle.

According to the embodiments of the present disclosure, the driver controls the vehicle through multiple components such as gears, accelerators, and brakes according to the actual situation while the vehicle is running. The accelerator signal and brake signal are comprehensively trained, so that a more accurate target power distribution neural network model can be obtained.

According to the embodiments of the present disclosure, when predicting the target power distribution strategy corresponding to the predicted vehicle speed, the vehicle speed at the current stage, the predicted vehicle speed, the remaining battery power of the vehicle, the gear signal, the accelerator signal and the brake signal can be simultaneously input into the target power distribution In the neural network model, the target power allocation strategy is obtained.

According to an embodiment of the present disclosure, establishing a vehicle speed prediction model according to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the historical driving stage may include the following steps:

According to the optimal power distribution strategy in the historical driving stage, the vehicle speed state matrix is determined. According to the vehicle speed state matrix and the vehicle speed frequency matrix, the state transition probability matrix is determined. A vehicle speed prediction model is determined according to the state transition probability matrix, wherein the vehicle speed prediction model may include a Markov chain vehicle speed prediction model.

According to an embodiment of the present disclosure, the Markov chain vehicle speed prediction model is essentially a Markov chain prediction model, and the solution direction of the Markov chain is a probability prediction method about the occurrence of an event, and predicts its future time or time according to the current state. A method of forecasting changes over time.

According to an embodiment of the present disclosure, the vehicle speed state matrix is constructed by the vehicle speed at each historical time point in the historical driving phase. The vehicle speed frequency matrix is composed of the proportion of each vehicle speed in the total number of vehicle speeds. State transition probability matrix: the probability of changing from one vehicle speed to another, a matrix composed of multiple probabilities.

FIG. 3 schematically shows a training flow chart of a target power distribution neural network model according to an embodiment of the present disclosure.

As shown in Figure 3, using the vehicle trip sample set to train the initial power distribution neural network model to obtain the target power distribution neural network model may include steps S301 to S303:

In step S301, the vehicle driving sample is input into the initial power distribution neural network model, and the predicted power distribution strategy is output, wherein the predicted power distribution strategy may include the predicted output torque of the engine, the output torque of the generator and the output torque of the drive motor.

In step S302, the loss result is obtained according to the predicted power allocation strategy and the label data calculation function.

In step S303, the parameters of the initial power distribution neural network model are iteratively adjusted according to the loss result, and the target power distribution neural network model is generated.

According to the embodiments of the present disclosure, in the training process, each vehicle driving sample input into the initial power distribution neural network model can output a predicted power distribution strategy, and the loss results of the two are calculated according to the power allocation strategy and the actual label data. , so that the parameters of the initial power distribution neural network model are adjusted according to the loss results. If the loss result meets the convergence condition, it can be determined as the target power distribution neural network model.

According to an embodiment of the present disclosure, the method may further include the following operations:

In the case that the mileage of the vehicle in the current driving stage satisfies the preset update step size, the optimal power distribution strategy of the current driving stage after the update step size is obtained.

The optimal power allocation strategy of the current driving stage after updating the step size and the corresponding historical driving data of the vehicle are used as new label data and new vehicle travel samples to train the target power allocation neural network model, and a new target power allocation is obtained. Neural network model.

According to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the current driving stage, a new vehicle speed prediction model is established.

According to the new predicted time point and the new vehicle speed prediction model, a new predicted vehicle speed at a new predicted time point in the future driving stage is determined.

The vehicle speed and the new predicted vehicle speed in the future driving stage are input into the new target power distribution neural network model, and a new target power distribution strategy corresponding to the new predicted vehicle speed is obtained, so as to control the vehicle according to the new target power distribution strategy.

In an exemplary embodiment, for example, the preset update step size is 20km, and the total vehicle mileage at the current time t of the vehicle is 111km, so it can be known that the mileage of the current driving stage does not meet the preset update step size. The vehicle continues to drive. At the next time t+1, the total mileage of the vehicle is 120 km, and the mileage at the next time t+1 satisfies the preset update step size. Therefore, the optimal power distribution strategy of the vehicle in the stage of 100km to 120km and the corresponding historical driving data of the vehicle can be used as new label data and new vehicle travel samples to update and train the target power distribution neural network model.

At the same time, the vehicle speed prediction model is re-established according to the vehicle historical driving data corresponding to the optimal power distribution strategy of the vehicle in the stage of 100km-120km, and then the predicted vehicle speed is predicted according to the new prediction time point t+2. Finally, the vehicle speed at time t+1 and the predicted speed at time t+2 are input into the above-mentioned updated and trained target power distribution neural network model, and the target power distribution strategy at time t+2 is obtained.

According to the embodiment of the present disclosure, the above-mentioned update training using the data of the closest historical driving stage as the vehicle itinerary sample can enable the target power distribution neural network model to learn the driving characteristics closest to the current driving state of the driver, so as to be able to Improve the economical use of vehicles.

FIG. 4 schematically shows a block diagram of a hybrid vehicle control apparatus according to an embodiment of the present disclosure.

As shown in FIG. 4 , the hybrid vehicle control apparatus 400 may include an acquisition module 401 , a training module 402 , a establishment module 403 , a vehicle speed prediction module 404 and a power distribution prediction module 405 .

The acquiring module 401 is configured to acquire a vehicle trip sample set, wherein the vehicle trip sample set may include a plurality of vehicle trip samples and label data corresponding to each vehicle trip sample, and the vehicle trip sample represents the vehicle history of the vehicle in the historical driving stage Driving data, the historical driving data of the vehicle may include vehicle speed sets corresponding to different time periods of the vehicle in the historical driving stage, and the vehicle speed set may include the first vehicle speed and the second vehicle speed, wherein the first vehicle speed and the second vehicle speed are the middle of the same time period The label data represents the optimal power distribution strategy of the vehicle in the historical driving stage at the speed of a preset period of time.

The training module 402 is used to train an initial power distribution neural network model by using the vehicle trip sample set to obtain a target power distribution neural network model, wherein the initial power distribution neural network model is based on the test cycle optimal power distribution data corresponding to the test cycle data. Established, test cycle data characterizes vehicle driving data under standard operating conditions.

The establishment module 403 is configured to establish a vehicle speed prediction model according to the historical driving data of the vehicle corresponding to the optimal power distribution strategy in the historical driving stage.

The vehicle speed prediction module 404 is configured to determine the predicted vehicle speed at the predicted time point according to the predicted time point and the vehicle speed prediction model.

The power distribution prediction module 405 is used for inputting the vehicle speed and predicted vehicle speed of the current driving stage into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, so as to control the vehicle according to the target power distribution strategy.

According to the embodiment of the present disclosure, the initial power distribution neural network model is trained through the historical vehicle driving data and the optimal power distribution strategy in the historical driving stage, and the target power distribution neural network model is obtained, and the target power distribution neural network model is used to analyze the current power distribution. The vehicle speed and predicted vehicle speed in the driving phase are processed, so that the target power distribution strategy corresponding to the predicted vehicle speed can be obtained, and then the vehicle can be controlled according to the target power distribution strategy. Since the target power distribution neural network model learns the historical data during the training process Regarding the optimal power distribution strategy, using the predicted target power distribution strategy to control the vehicle can reduce the economy of the vehicle and solve the problem of poor economy in the use of the hybrid vehicle.

According to an embodiment of the present disclosure, the hybrid vehicle control apparatus 400 may further include a computing module.

The calculation module is used to process the historical driving data of the vehicle in the historical driving stage by using the dynamic programming algorithm, and obtain the optimal power distribution strategy in the historical driving stage, wherein the optimal power allocation strategy may include: engine output torque, generator output torque and the optimal value of the motor output torque.

According to an embodiment of the present disclosure, the hybrid vehicle control apparatus 400 may also include processing modules and building modules.

The processing module is used to process the test cycle data by using the dynamic programming algorithm to obtain the optimal power distribution data of the test cycle.

The building block is used to establish an initial power distribution neural network model based on the test cycle data and the test cycle optimal power distribution data.

According to an embodiment of the present disclosure, the initial power distribution neural network model may include a BP neural network model, and the initial power distribution neural network model may include an input layer, at least one hidden layer, and an output layer.

According to an embodiment of the present disclosure, the target power distribution strategy may include an output torque signal of an engine of the vehicle, a generator output torque signal, and a drive motor output torque signal, wherein the generator is used to convert mechanical energy into electrical energy.

The historical driving data of the vehicle may further include: remaining battery power of the vehicle, gear signal, accelerator signal and brake signal.

According to an embodiment of the present disclosure, the establishment module 403 may include a first determination unit, a second determination unit, and a third determination unit.

The first determining unit is configured to determine the vehicle speed state matrix according to the optimal power distribution strategy in the historical driving stage.

The second determining unit is configured to determine the state transition probability matrix according to the vehicle speed state matrix and the vehicle speed frequency matrix.

The third determination unit is configured to determine a vehicle speed prediction model according to the state transition probability matrix, wherein the vehicle speed prediction model may include a Markov chain vehicle speed prediction model.

According to an embodiment of the present disclosure, the training module 402 may include an input unit, a calculation unit, and an iterative unit.

The input unit is used to input the vehicle driving sample into the initial power distribution neural network model, and output the predicted power distribution strategy, wherein the predicted power distribution strategy may include the predicted output torque of the engine, the output torque of the generator and the output torque of the driving motor.

The calculation unit is used to calculate the function according to the predicted power allocation strategy and the label data to obtain the loss result.

The iterative unit is used to iteratively adjust the parameters of the initial power distribution neural network model according to the loss result to generate the target power distribution neural network model.

According to an embodiment of the present disclosure, the hybrid vehicle control apparatus 400 may further include an update module, an update training module, a second establishment module, a second vehicle speed prediction module, and a second power distribution prediction module.

an update module, configured to obtain the optimal power distribution strategy of the current driving stage after the update step when the mileage of the vehicle in the current driving stage meets the preset update step size;

Update the training module, which is used to train the target power distribution neural network model by using the optimal power distribution strategy of the current driving stage after updating the step size and the corresponding historical driving data of the vehicle as new label data and new vehicle travel samples, respectively. A new target power distribution neural network model is obtained.

The second establishment module is used for establishing a new vehicle speed prediction model according to the vehicle historical driving data corresponding to the optimal power distribution strategy in the current driving stage.

The second vehicle speed prediction module is configured to determine a new predicted vehicle speed at a new predicted time point in the future driving stage according to the new predicted time point and the new vehicle speed prediction model.

The second power distribution prediction module is used to input the vehicle speed in the future driving stage and the new predicted vehicle speed into the new target power distribution neural network model, and obtain a new target power distribution strategy corresponding to the new predicted vehicle speed. The target power distribution strategy controls the vehicle.

Any of the modules, units, or at least part of the functions of any of the modules according to the embodiments of the present disclosure may be implemented in one module. Any one or more of the modules and units according to the embodiments of the present disclosure may be divided into multiple modules for implementation. Any one or more of the modules and units according to the embodiments of the present disclosure may be at least partially implemented as hardware circuits, such as Field Programmable Gate Arrays (FPGA), Programmable Logic Arrays (Programmable Logic Arrays, PLA), system-on-chip, system-on-substrate, system-on-package, Application Specific Integrated Circuit (ASIC), or any other reasonable means of hardware or firmware that can integrate or package circuits, Or it can be implemented in any one of the three implementation manners of software, hardware and firmware, or in an appropriate combination of any of them. Alternatively, one or more of the modules and units according to the embodiments of the present disclosure may be implemented at least in part as computer program modules, which, when executed, may perform corresponding functions.

For example, any of the acquisition module 401, the training module 402, the establishment module 403, the vehicle speed prediction module 404, and the power distribution prediction module 405 may be combined in one module/unit for implementation, or any one of the modules/units may be disassembled Divided into multiple modules/units. Alternatively, at least part of the functionality of one or more of these modules/units may be combined with at least part of the functionality of other modules/units/sub-units and implemented in one module/unit. According to an embodiment of the present disclosure, at least one of the acquisition module 401, the training module 402, the establishment module 403, the vehicle speed prediction module 404, and the power distribution prediction module 405 may be implemented at least partially as a hardware circuit, such as a field programmable gate array ( FPGA), Programmable Logic Array (PLA), System-on-Chip, System-on-Substrate, System-on-Package, Application-Specific Integrated Circuit (ASIC), or any other reasonable means by which circuits can be integrated or packaged such as hardware or firmware It can be realized by any one of the three implementation manners of software, hardware and firmware, or by any suitable combination of any of them. Alternatively, at least one of the acquisition module 401, the training module 402, the establishment module 403, the vehicle speed prediction module 404, and the power distribution prediction module 405 may be implemented at least in part as a computer program module that, when executed, may execute corresponding function.

It should be noted that the part of the hybrid vehicle control device in the embodiments of the present disclosure corresponds to the hybrid vehicle control method part in the embodiments of the present disclosure, and the description of the hybrid vehicle control device part refers to the hybrid vehicle control method. part, which will not be repeated here.

Figure 5 schematically shows a block diagram of an electronic device suitable for implementing the method described above according to an embodiment of the present disclosure. The electronic device shown in FIG. 5 is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.

As shown in FIG. 5 , an electronic device 500 according to an embodiment of the present disclosure includes a processor 501 that can be loaded into a random access memory according to a program stored in a read-only memory (Read-Only Memory, ROM) 502 or from a storage part 508 The program in the (Random Access Memory, RAM) 503 executes various appropriate operations and processes. The processor 501 may include, for example, a general-purpose microprocessor (eg, a CPU), an instruction set processor and/or a related chipset, and/or a special-purpose microprocessor (eg, an application-specific integrated circuit (ASIC)), among others. The processor 501 may also include on-board memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to the embodiments of the present disclosure.

In the RAM 503, various programs and data necessary for the operation of the electronic device 500 are stored. The processor 501 , the ROM 502 and the RAM 503 are connected to each other through a bus 504 . The processor 501 performs various operations of the method flow according to the embodiment of the present disclosure by executing the programs in the ROM 502 and/or the RAM 503 . Note that the program may also be stored in one or more memories other than the ROM 502 and the RAM 503 . The processor 501 may also perform various operations of the method flow according to the embodiments of the present disclosure by executing programs stored in the one or more memories.

According to an embodiment of the present disclosure, the electronic device 500 may further include an input/output (I/O) interface 505 which is also connected to the bus 504 . The system 500 may also include one or more of the following components connected to the I/O interface 505: an input portion 506 including a keyboard, mouse, etc.; including components such as a cathode ray tube (CRT), a Liquid Crystal Display (LCD) ), etc., and an output section 507 of speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, and the like. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I/O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is mounted on the drive 510 as needed so that a computer program read therefrom is installed into the storage section 508 as needed.

According to an embodiment of the present disclosure, the method flow according to an embodiment of the present disclosure may be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network via the communication portion 509 and/or installed from the removable medium 511 . When the computer program is executed by the processor 501, the above-described functions defined in the system of the embodiment of the present disclosure are performed. According to embodiments of the present disclosure, the above-described systems, apparatuses, apparatuses, modules, units, etc. can be implemented by computer program modules.

The present disclosure also provides a computer-readable storage medium. The computer-readable storage medium may be included in the device/apparatus/system described in the above embodiments; it may also exist alone without being assembled into the device/system. device/system. The above-mentioned computer-readable storage medium carries one or more programs, and when the above-mentioned one or more programs are executed, implement the method according to the embodiment of the present disclosure.

According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium. For example, it may include but not limited to: portable computer disk, hard disk, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM (Erasable Programmable Read Only Memory, EPROM) or flash memory), Portable compact disk read-only memory (Computer Disc Read-Only Memory, CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

For example, according to embodiments of the present disclosure, a computer-readable storage medium may include one or more memories other than ROM 502 and/or RAM 503 and/or ROM 502 and RAM 503 described above.

Embodiments of the present disclosure also include a computer program product, which includes a computer program, the computer program includes program codes for executing the methods provided by the embodiments of the present disclosure, and when the computer program product runs on an electronic device, the program The code is used to enable the electronic device to implement the hybrid vehicle control method provided by the embodiments of the present disclosure.

When the computer program is executed by the processor 501, the above-mentioned functions defined in the system/device of the embodiment of the present disclosure are performed. According to embodiments of the present disclosure, the systems, apparatuses, modules, units, etc. described above may be implemented by computer program modules.

In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted, distributed in the form of a signal over a network medium, and downloaded and installed through the communication section 509, and/or installed from a removable medium 511. The program code embodied by the computer program may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.

According to the embodiments of the present disclosure, the program code for executing the computer program provided by the embodiments of the present disclosure may be written in any combination of one or more programming languages, and specifically, high-level procedures and/or object-oriented programming may be used. programming language, and/or assembly/machine language to implement these computational programs. Programming languages include, but are not limited to, languages such as Java, C++, python, "C" or similar programming languages. The program code may execute entirely on the user computing device, partly on the user device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (eg, using an Internet service provider business via an Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams or flowchart illustrations, and combinations of blocks in the block diagrams or flowchart illustrations, can be implemented in special purpose hardware-based systems that perform the specified functions or operations, or can be implemented using A combination of dedicated hardware and computer instructions is implemented. Those skilled in the art will appreciate that various combinations and/or combinations of features recited in various embodiments and/or claims of the present disclosure are possible, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments of the present disclosure and/or in the claims may be made without departing from the spirit and teachings of the present disclosure. All such combinations and/or combinations fall within the scope of this disclosure.

Embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only, and are not intended to limit the scope of the present disclosure. Although the various embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the appended claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art can make various substitutions and modifications, and these substitutions and modifications should all fall within the scope of the present disclosure.

Claims (10)

1. A hybrid vehicle control method comprising:

obtaining a vehicle travel sample set, wherein the vehicle travel sample set comprises a plurality of vehicle travel samples and tag data corresponding to each vehicle travel sample, the vehicle travel samples represent vehicle historical travel data of the vehicle in a historical travel stage, the vehicle historical travel data comprise vehicle speed sets respectively corresponding to different time periods of the vehicle in the historical travel stage, the vehicle speed sets comprise a first vehicle speed and a second vehicle speed, the first vehicle speed and the second vehicle speed are vehicle speeds separated by a preset time length in the same time period, and the tag data represent an optimal power distribution strategy of the vehicle in the historical travel stage;

training an initial power distribution neural network model by using the vehicle travel sample set to obtain a target power distribution neural network model, wherein the initial power distribution neural network model is established according to test cycle optimal power distribution data corresponding to test cycle data, and the test cycle data represent vehicle running data of the vehicle under a standard working condition;

establishing a vehicle speed prediction model according to vehicle historical driving data corresponding to the optimal power distribution strategy in a historical driving stage;

determining a predicted vehicle speed at the predicted time point according to the predicted time point and the vehicle speed prediction model;

and inputting the vehicle speed of the current driving stage and the predicted vehicle speed into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed, and controlling the vehicle according to the target power distribution strategy.

2. The method of claim 1, wherein the optimal power allocation strategy within the historical driving phase is determined by:

processing the historical driving data of the vehicle in the historical driving stage by using a dynamic programming algorithm to obtain an optimal power distribution strategy in the historical driving stage, wherein the optimal power distribution strategy comprises the following steps: and optimal values of the engine output torque, the generator output torque and the motor output torque.

3. The method of claim 1, wherein the initial power allocation neural network model is built from test cycle optimal power allocation data corresponding to test cycle data, comprising:

processing the test cycle data by using a dynamic programming algorithm to obtain test cycle optimal power distribution data;

and establishing the initial power distribution neural network model according to the test cycle data and the test cycle optimal power distribution data.

4. The method of claim 3, wherein the initial power distribution neural network model comprises a BP neural network model, the initial power distribution neural network model comprising an input layer, at least one hidden layer, and an output layer.

5. The method of claim 1, wherein the historical driving data of the vehicle in the historical driving stage further comprises a total driving range of the vehicle as shown in formula (1), the vehicle state data of the current driving stage comprises the driving range of the vehicle in the current driving stage, a real-time vehicle speed and a required power, and the driving range of the vehicle in the current driving stage is calculated as shown in formula (2):

therein, Dis _n Is the total driving distance, Dis, of the vehicle in the historical driving stage _n-1 The total driving reference mileage, Dis, of the vehicle in a driving stage before the historical driving stage _cur The total driving reference mileage of the vehicle in the historical driving stage, m is the updating step length of the historical travel data, L _n For the mileage of the vehicle in the current driving stage, L _n-1 The driving distance of the vehicle in the historical driving stage is obtained.

6. The method of claim 5, wherein the target power allocation policy comprises: an output torque signal of an engine of the vehicle, a generator output torque signal, and a drive motor output torque signal, wherein the generator is configured to convert mechanical energy into electrical energy;

the vehicle history traveling data further includes: the vehicle battery residual capacity, the gear signal, the accelerator signal and the brake signal.

7. The method of claim 1, wherein the building a vehicle speed prediction model from historical vehicle travel data corresponding to an optimal power distribution strategy over historical travel phases comprises:

determining a vehicle speed state matrix according to the optimal power distribution strategy in the historical driving stage;

determining a state transition probability matrix according to the vehicle speed state matrix and the vehicle speed frequency matrix;

and determining the vehicle speed prediction model according to the state transition probability matrix, wherein the vehicle speed prediction model comprises a Markov chain vehicle speed prediction model.

8. The method of claim 1, wherein the training an initial power distribution neural network model using the vehicle trip sample set to obtain a target power distribution neural network model comprises:

inputting the vehicle running sample into the initial power distribution neural network model, and outputting a predicted power distribution strategy, wherein the predicted power distribution strategy comprises predicted output torque of an engine, output torque of a generator and output torque of a driving motor;

calculating a function according to the predicted power distribution strategy and the tag data to obtain a loss result;

and iteratively adjusting parameters of the initial power distribution neural network model according to the loss result to generate the target power distribution neural network model.

9. The method of any of claims 1-8, further comprising:

under the condition that the driving mileage of the vehicle in the current driving stage meets a preset updating step length, acquiring an optimal power distribution strategy of the current driving stage after the updating step length;

respectively taking the optimal power distribution strategy of the current driving stage after the step length is updated and the corresponding vehicle historical driving data as new label data and a new vehicle travel sample to train the target power distribution neural network model to obtain a new target power distribution neural network model;

establishing a new vehicle speed prediction model according to the vehicle historical driving data corresponding to the optimal power distribution strategy in the current driving stage;

determining a new predicted vehicle speed at a new predicted time point of a future driving stage according to the new predicted time point and the new vehicle speed prediction model;

and inputting the vehicle speed in the future driving stage and the new predicted vehicle speed into the new target power distribution neural network model to obtain a new target power distribution strategy corresponding to the new predicted vehicle speed, and controlling the vehicle according to the new target power distribution strategy.

10. A hybrid vehicle control apparatus comprising:

the vehicle travel sample set comprises a plurality of vehicle travel samples and tag data corresponding to each vehicle travel sample, wherein the vehicle travel samples represent vehicle historical travel data of the vehicle in a historical travel stage, the vehicle historical travel data comprise vehicle speed sets corresponding to the vehicle in different time periods in the historical travel stage respectively, the vehicle speed sets comprise a first vehicle speed and a second vehicle speed, the first vehicle speed and the second vehicle speed are vehicle speeds separated by a preset time period in the same time period, and the tag data represent an optimal power distribution strategy of the vehicle in the historical travel stage;

the training module is used for training an initial power distribution neural network model by using the vehicle travel sample set to obtain a target power distribution neural network model, wherein the initial power distribution neural network model is established according to test cycle optimal power distribution data corresponding to test cycle data, and the test cycle data represents vehicle running data of the vehicle under a standard working condition;

the establishing module is used for establishing a vehicle speed prediction model according to vehicle historical driving data corresponding to the optimal power distribution strategy in a historical driving stage;

the vehicle speed prediction module is used for determining the predicted vehicle speed at the predicted time point according to the predicted time point and the vehicle speed prediction model; and

and the power distribution prediction module is used for inputting the vehicle speed of the current driving stage and the predicted vehicle speed into the target power distribution neural network model to obtain a target power distribution strategy corresponding to the predicted vehicle speed so as to control the vehicle according to the target power distribution strategy.

Images