Extended-range electric vehicle energy management control method based on vehicle speed prediction

Abstract

The invention discloses a control method for energy management of an extended-range electric vehicle based on vehicle speed prediction. The control method comprises the following steps: acquiring historical vehicle speed information based on a vehicle-mounted sensor; a long-short-term memory network vehicle speed prediction model is constructed, and prediction and verification are carried out through the training set and the test set; the obtained historical vehicle speed information is used as input, and the future vehicle speed is predicted; building an engine, a motor, a power battery and a longitudinal dynamic model; the relation between the whole vehicle demand torque and the predicted vehicle speed is constructed; solving a self-adaptive equivalent fuel consumption factor; and solving the minimum amount of equivalent fuel consumption to obtain the torque of the range extender with the optimal control amount, and repeating the steps until the circulation is finished. The future vehicle speed is predicted through the long-short-term memory network, the vehicle running working condition can be predicted in advance, real-time optimization control can be achieved, meanwhile, the equivalent fuel consumption is minimized, the fuel economy and emission performance of the whole vehicle are improved, and the air quality is improved to a certain degree.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy vehicles, and particularly relates to a control method for energy management of a range-extended electric vehicle based on vehicle speed prediction. Background Technique

[0002] The energy management strategy of a range-extended electric vehicle is one of its core research contents, and it plays a crucial role in realizing the energy conservation and emission reduction of the whole vehicle. The energy management strategy is generally divided into two categories: one is the rule-based energy management strategy, and the other is the optimization-based energy management strategy. The rule-based energy management strategy is simple and easy to implement, with strong adaptability and robustness. However, its design depends on the engineering experience of engineers and the steady-state characteristics of each power component, and it is difficult to achieve the optimal control effect, leaving a large room for optimization. The optimization-based energy management strategy can be further divided into the global optimization energy management strategy carried out under specific working conditions or data and the instantaneous optimization energy management strategy based on real-time operating conditions. The global optimization energy management strategy has problems of large computational complexity and poor real-time performance, while the instantaneous optimization energy management strategy has small computational complexity and good real-time performance. Summary of the Invention

[0003] To solve at least one of the problems mentioned in the above background technique, the present invention provides a control method for energy management of a range-extended electric vehicle based on vehicle speed prediction, aiming to improve the fuel economy and emission performance of the whole vehicle and improve the air quality to a certain extent.

[0004] To achieve the above object, the present invention provides the following technical solution: A control method for energy management of a range-extended electric vehicle based on vehicle speed prediction, comprising the following steps; Step 1, construct a long short-term memory network ( ) vehicle speed prediction model, take the vehicle speed under standard working conditions as the standard input, perform online prediction and verify the effectiveness of the model, and select the root mean square error as the standard for measuring the prediction effectiveness; Step 2, obtain the vehicle speed information within a period of time before the current moment based on on-vehicle sensors, perform normalization preprocessing on the obtained historical vehicle speed information to ensure the applicability of the data, and use the preprocessed historical vehicle speed information as the input to predict the vehicle speed in the future for a period of time; Step 3, build vehicle longitudinal dynamics, engine, motor, and power battery models according to the actual parameters of the range-extended electric vehicle; Step 4, construct the relationship between the total vehicle demand torque and the predicted vehicle speed. When the predicted vehicle speed is higher than the vehicle speed at the previous moment, it indicates that the vehicle has an acceleration demand, and at this time the total vehicle demand torque will decrease. When the predicted vehicle speed is lower than the vehicle speed at the previous moment, it indicates that the vehicle has a deceleration demand, and at this time the total vehicle demand torque will increase;​ Step 5: Construct an adaptive equivalent fuel consumption factor based on the relationship between the vehicle demand torque value at the next moment and the vehicle demand torque value at the current moment, and the initial equivalent fuel consumption factor ; Step 6: According to the algorithm, by solving the equivalent fuel consumption , and minimizing the equivalent fuel consumption , obtain the optimal control amount of the range extender torque , and repeat the above steps until the loop ends; In Step 1, the root mean square error calculation formula is as follows: ; where is the root mean square error within the prediction horizon, is the predicted vehicle speed at time , is the actual vehicle speed at time , is the length of the prediction horizon; In Step 4, there is a specific relationship between the vehicle demand torque and the vehicle speed, and the expression is as follows: ; where and are proportionality coefficients, , is the vehicle demand torque at time , is the vehicle demand torque at time , is the vehicle speed at time, is the vehicle speed at time.

[0005] When the predicted vehicle speed is higher than the set threshold, the proportionality coefficient will decrease according to the set ratio. When the predicted vehicle speed is lower than the set threshold, the proportionality coefficient will increase according to the set ratio, and the expression is as follows: ; where is the proportionality coefficient, is the set vehicle speed threshold.

[0006] In Step 5, the adaptive equivalent fuel consumption factor , according to the relationship between the vehicle demand torque value at the next moment and the vehicle demand torque value at the current moment, plus the equivalent fuel consumption factor at the current moment , the adaptive equivalent fuel consumption factor at the next moment is obtained , and the expression is as follows: ; where is the equivalent fuel consumption factor at the current moment, is the vehicle demand torque at the next moment, is the vehicle demand torque at the current moment, is the proportional term coefficient, is the integral term coefficient, and is adjusted through .

[0007] In step 6, the equivalent fuel consumption , and the calculation formula is as follows: ; ; where is the fuel consumption of the engine, is the equivalent fuel consumption rate of the battery, is the adaptive equivalent fuel consumption factor, is the low calorific value of the engine fuel, is the output power of the power battery, is the penalty function, is the vehicle demand torque at time is the vehicle demand torque at time.

[0008] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention predicts the future vehicle speed through a long short-term memory network, can predict the vehicle driving conditions in advance, is conducive to realizing real-time optimal control, and at the same time, minimizes the equivalent fuel consumption, improves the fuel economy and emission performance of the whole vehicle, and improves the air quality to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings: Figure 1 is the control flow chart of a control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to the present invention; Figure 2 is the schematic diagram of the neural unit of the vehicle speed prediction model of the long short-term memory network ( ) of the present invention; Figure 3 is according to the present invention the flow chart of the algorithm. Specific Embodiments

[0010] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Generally, the components of the embodiments of the present invention described and shown in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed present invention, but only represents the selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0011] First, with reference to Figures 1 to 3 A control method for energy management of an extended-range electric vehicle based on vehicle speed prediction involved in the embodiments of the present invention will be specifically described.

[0012] In this embodiment, a control method for energy management of an extended-range electric vehicle based on vehicle speed prediction has a control flow as Figure 1 shown, and specifically includes the following steps: Step 1. Construct a long short-term memory network ( ) vehicle speed prediction model, use the vehicle speed under standard working conditions as the standard input, perform online prediction and verify the effectiveness of the model, and select the root mean square error as the standard for measuring the prediction effectiveness; Step 1.1. The constructed long short-term memory network ( ) is a special recurrent neural network for processing data with time series characteristics. As shown, it generally consists of a forgetting gate, an input gate, and an output gate: Figure 2 ; ; Among them, is the forgetting gate, which determines how much information from the previous time step is forgotten, is the input gate, which is used to control the update degree of the input gate, is used to control the input data, is the output at the current moment, is the output at the previous moment, is the output gate, which determines the output information, is the activation function, is the input value at the current moment, is the state value at the previous moment, is the state value at the current moment, is the weight matrix of the forgetting gate, is the bias vector of the forget gate, is the weight matrix of the input gate, is the bias vector of the input gate, is the weight matrix of the candidate value, is the bias vector of the candidate value, is the weight matrix of the output gate, is the bias vector of the output gate; Step 1.2, The selected root mean square error calculation formula is as follows: ;

[0013] where is the root mean square error within the prediction time domain, is the predicted vehicle speed at time is the actual vehicle speed at time is the prediction time domain length;

[0014] Step 2, Based on in-vehicle sensors, obtain the vehicle speed information within the previous minute of the current moment, perform normalization preprocessing on the obtained historical vehicle speed information to ensure the applicability of the data, and use the preprocessed historical vehicle speed information as the input to predict the vehicle speed for a period of time in the future;

[0015] Step 2.1, The normalization preprocessing adopts the maximum-minimum normalization processing, and the calculation formula is as follows: ;

[0016] where, is the maximum vehicle speed in the obtained historical vehicle speed, is the minimum vehicle speed in the obtained historical vehicle speed, is the normalized vehicle speed;

[0017] Step 3, According to the actual parameters of the range-extended electric vehicle, build vehicle longitudinal dynamics, engine, motor, and power battery models through simulation software;

[0018] Step 4, Construct the relationship between the vehicle's total demand torque and the predicted vehicle speed. When the predicted vehicle speed is higher than the vehicle speed at the previous moment, it indicates that the vehicle has an acceleration demand, and at this time the vehicle's total demand torque will decrease. When the predicted vehicle speed is lower than the vehicle speed at the previous moment, it indicates that the vehicle has a deceleration demand, and at this time the vehicle's total demand torque will increase;

[0019] Step 4.1, There is a certain relationship between the constructed vehicle's total demand torque and the vehicle speed, and the expression is as follows:

[0020] wherein and are proportionality coefficients, , is the vehicle demand torque at time is the vehicle demand torque at time is the vehicle speed at time is the vehicle speed at time;

[0021] Step 4.2: When the predicted vehicle speed is higher than the set threshold, the proportionality coefficient will decrease in a certain proportion. When the predicted vehicle speed is lower than the set threshold, the proportionality coefficient will increase in a certain proportion. The expression is as follows: ; wherein, is the proportionality coefficient, is the set vehicle speed threshold; Step 5: Construct an adaptive equivalent fuel consumption factor according to the relationship between the vehicle demand torque value at the next moment and the vehicle demand torque value at the current moment, and the initial equivalent fuel consumption factor; Step 5.1: The constructed adaptive equivalent fuel consumption factor , according to the relationship between the vehicle demand torque value at the next moment and the vehicle demand torque value at the current moment, plus the equivalent fuel consumption factor at the current moment, to obtain the adaptive equivalent fuel consumption factor at the next moment. The expression is as follows: ; wherein, is the equivalent fuel consumption factor at the current moment, is the vehicle demand torque at the next moment, is the vehicle demand torque at the current moment, is the proportional term coefficient, is the integral term coefficient, which is adjusted through to select an appropriate weight coefficient; Step 6: According to the algorithm, by solving the equivalent fuel consumption , and minimizing the equivalent fuel consumption , to obtain the optimal control amount of the range extender torque , and the constraint condition is: ; Wherein, is the minimum torque of the engine, is the maximum torque of the engine, is the minimum torque of the motor, is the maximum torque of the motor, is the minimum speed of the engine, is the maximum speed of the engine, is the minimum speed of the motor, is the maximum speed of the motor.

[0022] Repeat the above steps until the loop ends; Step 6.1, the equivalent fuel consumption , and the calculation formula is as follows:

[0023] ; Wherein, is the fuel consumption of the engine, is the equivalent fuel consumption rate of the battery, is the adaptive equivalent fuel consumption factor, is the lower calorific value of the engine fuel, is the output power of the power battery, is the penalty function, is the vehicle demand torque at time is the vehicle demand torque at time.

[0024] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the claims and their equivalents.

Claims

1. A control method for energy management of an extended-range electric vehicle based on vehicle speed prediction, characterized in that: The steps include: Step 1: Build a long short-term memory network ( ) vehicle speed prediction model, The vehicle speed under standard working conditions is used as the standard input to make predictions online and verify the effectiveness of the model. The root mean square error is selected as the standard to measure the effectiveness of the prediction. Step 2: Based on the vehicle-mounted sensor, the vehicle speed information in the period before the current moment is obtained, and the obtained historical vehicle speed information is normalized and preprocessed to ensure the applicability of the data. The preprocessed historical vehicle speed information is used as input to predict the vehicle speed in the future period; Step 3: Build the vehicle longitudinal dynamics, engine, motor, and power battery models based on the actual parameters of the extended-range electric vehicle; Step 4: Build the vehicle required torque Relationship with the predicted vehicle speed: when the predicted vehicle speed is higher than the speed at the previous moment, it indicates that the vehicle has a need to accelerate, and the required torque of the vehicle will decrease; when the predicted vehicle speed is lower than the speed at the previous moment, it indicates that the vehicle has a need to decelerate, and the required torque of the vehicle will increase; Step 5: According to the relationship between the vehicle torque value required at the next moment and the vehicle torque value required at the current moment, as well as the initial equivalent fuel consumption factor, an adaptive equivalent fuel consumption factor is constructed. ; Step 6: According to Algorithm, by solving the equivalent fuel consumption , and minimize equivalent fuel consumption , and get the optimal control quantity range extender torque , repeat the above steps until the cycle ends.

2. The control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to claim 1 is characterized in that: In step 1, the root mean square error calculation formula is as follows: ； in is the root mean square error in the prediction time domain, for The predicted speed at the time, for The actual vehicle speed at the moment, is the predicted time domain length.

3. The control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to claim 1 is characterized in that: In step 4, the vehicle required torque There is a specific relationship with the vehicle speed, and the expression is as follows: ； in and is the proportionality coefficient, , for The vehicle torque required at the time, for The vehicle torque required at the time, is the vehicle speed at the time, is the vehicle speed at the time.

4. The control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to claim 3 is characterized by: When the predicted vehicle speed is higher than the set threshold, the proportional coefficient It will decrease according to the set ratio. When the predicted vehicle speed is lower than the set threshold, the proportional coefficient It will increase according to the set ratio, the expression is as follows: ； in, is the proportionality coefficient, is the set vehicle speed threshold.

5. The control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to claim 1 is characterized in that: In step 5, the adaptive equivalent fuel consumption factor , based on the relationship between the vehicle torque requirement at the next moment and the vehicle torque requirement at the current moment, plus the equivalent fuel consumption factor at the current moment , get the adaptive equivalent fuel consumption factor at the next moment , the expression is as follows: ； in is the equivalent fuel consumption factor at the current moment, is the vehicle required torque at the next moment, is the vehicle required torque at the current moment, is the proportional term coefficient, is the integral term coefficient, through adjust.

6. The control method for energy management of an extended-range electric vehicle based on vehicle speed prediction according to claim 1 is characterized by: In step 6, the equivalent fuel consumption , the calculation formula is as follows: ； ； in, is the fuel consumption of the engine, is the equivalent fuel consumption rate of the battery, is the adaptive equivalent fuel consumption factor, The lower calorific value of the engine fuel. Output power for the power battery, is the penalty function, for The vehicle torque required at the time, for The vehicle's required torque at the time.

Images