Self-adaptive fuzzy control method and system for improving braking energy recovery efficiency

Abstract

The invention discloses a self-adaptive fuzzy control method and system for improving the braking energy recovery efficiency, and the method comprises the steps: controlling a motor to generate power and recover energy according to a braking force demand during braking; according to the motor temperature and the motor pull-back braking force, the fuzzy controller calculates the requirement for the refrigerating capacity of the motor of the electric vehicle; the heat management system dynamically adjusts the refrigerating capacity according to the requirement of the motor refrigerating capacity and maintains the optimal working temperature of the motor; and optimizing the thermal management system by using a particle swarm optimization algorithm. By coordinately controlling the motor power generation and heat management system, the energy recovery and heat management system can efficiently operate according to actual requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of brake energy recovery, and in particular to a method for improving the energy recovery efficiency of an electric vehicle. Background Art

[0002] With the rapid development of electric vehicles, the energy recovery technology of electric vehicles has received more and more attention. The existing energy recovery system mainly relies on the motor to generate feedback torque to achieve braking, but due to the inefficiency of the thermal management system, the energy recovery rate is often limited. Especially in high temperature environments, the operating temperature of the motor is too high, which will lead to energy loss and reduced system efficiency.

[0003] During braking, electric vehicles use motors to generate electricity to recover energy. During braking, the motor operates as a generator, converting the vehicle's kinetic energy into electrical energy and storing it in the battery. This method can effectively recover part of the braking energy, thereby improving the vehicle's energy efficiency. However, the motor generates a lot of heat during the power generation process, which needs to be cooled by a thermal management system. Traditional thermal management systems usually use a fixed cooling capacity method to maintain the motor temperature through coolant circulation or fan cooling. This cooling method is inefficient and is not conducive to energy recovery during braking.

[0004] Disadvantages of existing technology:

[0005] 1. Inefficiency caused by fixed cooling capacity: Existing thermal management systems usually adopt a fixed cooling capacity approach and are unable to dynamically adjust the cooling capacity according to actual demand, resulting in low energy efficiency and energy waste in the cooling system under low load conditions.

[0006] 2. Insufficient cooling capacity in high temperature environment: In high temperature environment, the heat generated by the motor increases, and the traditional fixed cooling capacity cannot provide sufficient cooling capacity, resulting in reduced system efficiency and even overheating protection, affecting vehicle performance.

[0007] 3. Lack of coordination with the energy recovery system: In the existing technology, the thermal management system and the energy recovery system operate independently and lack coordinated control, resulting in a low braking energy recovery rate and failure to fully utilize the energy generated during braking.

[0008] In summary, the existing technology urgently needs to solve the following technical problems:

[0009] 1. Improve the efficiency of braking energy recovery.

[0010] 2. Optimize the energy efficiency of the thermal management system to avoid energy waste.

[0011] 3. Improve the cooling capacity of the system, especially the cooling effect in high temperature environment.

[0012] 4. Refrigerant direct cooling motor, through dynamic adjustment of cooling capacity, control of refrigerant cooling capacity, maintain constant temperature of motor, and extend the service life of motor. Summary of the invention

[0013] The present invention aims to overcome the above-mentioned shortcomings of the prior art and provide an adaptive fuzzy control method and system for improving the efficiency of braking energy recovery.

[0014] The present invention provides an adaptive fuzzy control method capable of improving energy recovery rate during braking, and by coordinating and controlling motor power generation and thermal management systems, the energy recovery and thermal management systems can be operated efficiently according to actual needs.

[0015] A first aspect of the present invention provides an adaptive fuzzy control method for improving braking energy recovery efficiency, comprising:

[0016] S1: During braking, the motor is controlled to generate electricity and recover energy according to the braking force requirement;

[0017] S2: Based on the motor temperature and the motor retraction braking force, the fuzzy controller calculates the cooling capacity requirement of the electric vehicle motor.

[0018] S3: The thermal management system dynamically adjusts the cooling capacity according to the motor cooling demand to maintain the optimal operating temperature of the motor.

[0019] S4: Optimize the thermal management system using particle swarm optimization algorithm.

[0020] Preferably, during the braking described in step S1, controlling the motor to generate electricity to recover energy according to the braking force requirement includes:

[0021] S11: The sensor detects the brake pedal position and obtains the driver's braking intention;

[0022] S12: obtaining brake intensity information according to the brake pedal position sensor;

[0023] S13: Calculate the ideal braking force based on the information of the braking intensity, vehicle speed sensor, and front and rear wheel weight;

[0024] S14: Determine the battery SOC value, and if the battery SOC value is less than 0.9, perform brake energy recovery, otherwise perform friction braking deceleration;

[0025] S15: Obtaining the maximum feedback braking force of each wheel according to the driving motor speed information;

[0026] S16: comparing the maximum feedback braking force of each wheel and the ideal braking force of each wheel;

[0027] S17: When the maximum feedback braking force of a wheel is greater than the ideal braking force of the wheel, the feedback braking force of the wheel provides the ideal braking force;

[0028] S18: When the maximum feedback regenerative braking force of a wheel is less than the ideal braking force of the wheel, the wheel provides the maximum feedback regenerative braking force, and at the same time the mechanical braking system provides friction braking force to make the total braking force reach the ideal braking force.

[0029] Furthermore, the characteristic of the ideal braking force in step S13 is that when each wheel reaches the ideal braking force, the front and rear wheels of the electric vehicle are locked at the same time.

[0030] Preferably, the motor temperature described in step S2 is obtained by a sensor, wherein the sensor includes a motor temperature sensor, a motor current and voltage sensor, and an ambient temperature sensor.

[0031] Preferably, the calculation of the return braking force in step S2 includes:

[0032] S21: determining the magnitude of the return braking force according to the composition of the braking force;

[0033] S22: If the wheel is decelerated by the friction braking force, the return braking force is 0;

[0034] S23: If the wheel is decelerated and braked by the motor feedback braking force, the feedback braking force is the ideal braking force of the wheel at this time;

[0035] S24: If the wheel is braked and decelerated by the maximum feedback braking force of the motor and the friction braking force, the return braking force is the maximum feedback braking force of the wheel motor at this time.

[0036] Preferably, the fuzzy controller described in step S2 adopts an adaptive neural fuzzy control method involving a two-input single-output adaptive neural network fuzzy inference system, which is composed of five intermediate layers, namely, fuzzification layer Layer1, fuzzy inference layer Layer2, normalization layer Layer3, rule output layer Layer4 and merged output layer Layer5; fuzzification layer Layer1 and rule output layer Layer4 are nodes with parameters, and fuzzy inference layer Layer2, normalization layer Layer3 and merged output layer Layer5 are nodes without parameters.

[0037] Furthermore, the assumed fuzzy rule base of the adaptive neuro-fuzzy inference system contains two first-order if-then fuzzy rules of the Takagi and Sugeno type:

[0038] Rule 1: (1)

[0039] Rule 2: (2)

[0040] (1) Fuzzy layer Layer 1

[0041] All nodes in the fuzzy layer are square nodes. Nodes have node functions ( )

[0042] (3)

[0043] In the formula is the input value, Is the input value associated with the node function The label, that is for The membership function of Satisfaction quantifier degree, usually It is represented by a bell or Gaussian shape, with a maximum value of 1 and a minimum value of zero, for example:

[0044] (4)

[0045] or

[0046] (5)

[0047] In the formula is the parameter set and all the parameters in this layer are called premise parameters. During the update process, the values ​​of these parameters change, and the membership function will also change accordingly, so that The membership functions of various driving modes are shown above.

[0048] (2) Fuzzy reasoning layer 2

[0049] The output of the nodes in the fuzzy inference layer represents the activation of the rules. Each node in this layer is a A circular node that multiplies the input signals and sends the product out. )

[0050] (6)

[0051] (3) Normalization layer Layer 3

[0052] The normalization layer is used to normalize the rule activation output by the fuzzy inference layer. Each node is a The circular node. The node calculates The ratio of the activation of a rule to the sum of the activations of all rules ( ):

[0053] (7)

[0054] (4) Rule output layer Layer 4

[0055] The rule output layer combines the rule activation output by the normalization layer with rules one and two in formulas 1 and 2 to obtain: )

[0056] (8)

[0057] In the formula is a parameter set, and all parameters in this layer are called subsequent parameters.

[0058] (5) Total output layer Layer 5

[0059] The single node of the total output layer is a The circular node calculates the total output as the sum of all input signals ( ),Right now:

[0060] (9)

[0061] Preferably, the thermal management system is optimized using a particle swarm optimization algorithm as described in step S4;

[0062] The objective function of the particle swarm optimization algorithm is ;in:

[0063] C is the energy consumption rate, which is defined as the ratio of cooling energy consumption to braking energy recovery;

[0064] η is the braking energy recovery efficiency of electric vehicles, which is defined as the ratio of the energy recovered by the system to the energy generated by vehicle deceleration;

[0065] , is the weight coefficient, and .

[0066] A second aspect of the present invention relates to an adaptive fuzzy control system for improving braking energy recovery efficiency, comprising:

[0067] Energy recovery module, used to control the motor to generate electricity and recover energy according to the braking force requirements during braking;

[0068] The motor cooling capacity calculation module is used to calculate the cooling capacity demand of the electric vehicle motor according to the motor temperature, motor retraction braking force, and fuzzy controller.

[0069] The cooling capacity modulation module is used in the thermal management system to dynamically adjust the cooling capacity according to the cooling capacity demand of the motor to maintain the optimal operating temperature of the motor.

[0070] The thermal management system optimization module uses the particle swarm optimization algorithm to optimize the thermal management system.

[0071] The third aspect of the present invention relates to an adaptive fuzzy control device for improving the efficiency of braking energy recovery, comprising a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement an adaptive fuzzy control method for improving the efficiency of braking energy recovery of the present invention.

[0072] The present invention has the following technical advantages:

[0073] 1. Improve energy recovery rate:

[0074] By coordinating the control of the motor power generation and thermal management system, the braking energy can be efficiently recovered, improving the energy recovery rate of the entire system. The dynamic coordination of the generator thermal management ensures efficient use of energy.

[0075] 2. Optimize thermal management system:

[0076] Dynamically adjust the cooling capacity according to actual needs, avoiding energy waste in traditional thermal management systems and improving the energy efficiency of the system. The dynamic adjustment of the thermal management system can be adjusted according to real-time working conditions to ensure efficient operation of the system.

[0077] 3. Extend system life:

[0078] Through effective thermal management, the optimal operating temperature of the motor is maintained, the damage to components caused by high temperature is reduced, and the service life of the system is extended. The temperature is controlled within a reasonable range, which slows down the aging of the motor.

[0079] 4. Improve vehicle performance:

[0080] In high temperature environments, it can provide sufficient cooling capacity to ensure that the motor operates at the optimal temperature, avoiding performance degradation and safety hazards caused by high temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 It is a schematic diagram of the braking force distribution principle of the present invention.

[0082] Figure 2 It is a schematic diagram of the system refrigeration control of the present invention.

[0083] Figure 3 It is a schematic diagram of the thermal management control principle of the present invention.

[0084] Figure 4 It is a structural diagram of the adaptive neural network fuzzy inference system of the present invention.

[0085] Figure 5It is a flow chart of solving the optimization model of the present invention.

[0086] Figure 6 It is a structural diagram of the device of the present invention. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] It should be noted that, in the absence of conflict, the features in the following embodiments and implementations may be combined with each other.

[0089] Example 1

[0090] This embodiment relates to an adaptive fuzzy control method for improving the efficiency of braking energy recovery. Figure 1 and Figure 2 As shown, including:

[0091] 1. When braking, control the motor to generate electricity and recover energy according to the braking force required:

[0092] During the vehicle braking process, the vehicle's kinetic energy is transferred to the motor through the drive device, driving the motor to operate. When the drive device provides force to the motor, the motor provides a reverse regenerative braking torque to the drive device due to the interaction of forces, so that the vehicle gradually slows down and stops. Since the external force drives the motor to operate, the motor is in a power generation state and generates electrical energy. The electrical energy generated by the motor is stored in the energy storage device, thereby realizing the conversion of kinetic energy into electrical energy during the braking process of the electric vehicle.

[0093] The battery needs to retain a certain safety buffer during the charging and discharging process to prevent overcharging and over-discharging, protect the battery and extend its service life. When the SOC value is high, the battery is close to a fully charged state. If energy recovery continues at this time, it may cause the battery to be overcharged, affecting the safety and life of the battery. Therefore, when the battery SOC value reaches 0.9, energy recovery is usually no longer performed to ensure that the battery will not be overcharged. At the same time, the battery's current change ability during discharge is also taken into account to avoid affecting power output and driving experience. Such a setting helps to maximize the efficiency and life of the battery while ensuring battery safety.

[0094] During braking, the braking force is distributed according to the vehicle's braking force requirements in order to obtain maximum energy recovery efficiency. Therefore, the motor regenerative braking force is used to the greatest extent possible during the working process to achieve maximum energy recovery.

[0095] Braking force distribution process:

[0096] (1) The sensor detects the brake pedal position and obtains the driver's braking intention;

[0097] (2) Obtain brake intensity information based on the brake pedal position sensor;

[0098] (3) Calculate the ideal braking force (the braking force required to lock the front and rear wheels at the same time) based on information such as braking intensity, vehicle speed sensor, and front and rear wheel weight;

[0099] (4) Determine the battery SOC value. When the battery SOC is less than 0.9, perform brake energy recovery. Otherwise, perform friction braking to decelerate.

[0100] (5) Obtain the maximum feedback braking force of each wheel based on the driving motor speed information;

[0101] (6) Compare the maximum feedback braking force of each wheel with the ideal braking force of each wheel;

[0102] (7) When the maximum regenerative braking force of a wheel is greater than the ideal braking force of the wheel, the regenerative braking force of the wheel provides the ideal braking force.

[0103] (8) When the maximum regenerative braking force of a wheel is less than the ideal braking force of the wheel, the wheel provides the maximum regenerative braking force, and the mechanical braking system provides friction braking force to make the total braking force reach the ideal braking force.

[0104] 2. Thermal management of brake energy recovery system:

[0105] like Figure 3 , according to the motor temperature and motor retraction braking force, determine the cooling capacity demand of the electric vehicle motor, such as Figure 3 As shown. Through the motor temperature sensor, motor current and voltage sensor, ambient temperature sensor, etc., the sensor data is transmitted to the control unit, and the control unit calculates the cooling capacity demand of the motor, thereby determining the cooling capacity demand of the motor, controlling the compressor and the refrigerant electronic throttling device, thereby controlling the refrigerant capacity for directly cooling the motor and keeping the motor temperature constant.

[0106] Cooling capacity control method of thermal management system. Figure 3As shown, temperature sensing includes motor temperature sensor, motor current and voltage sensor, and ambient temperature sensor. The data acquisition system collects and processes data, and the fuzzy controller calculates the motor cooling demand and actively controls the operation of the motor thermal management system. The control unit sends instructions to the execution unit to adjust the cooling system so that it outputs the corresponding cooling capacity to achieve dynamic cooling. By adjusting the working parameters of the cooling system (such as coolant flow, fan speed, etc.), it is ensured that the system can provide appropriate cooling capacity under different working conditions and maintain the optimal working temperature of the motor, thereby improving the energy recovery efficiency.

[0107] The fuzzy controller adopts the adaptive neuro-fuzzy control method. The adaptive neuro-fuzzy inference system can generate effective fuzzy control rules through data. Even if it is not possible to obtain enough rules through training, fuzzy inference can be performed through the original rules. Fuzzy inference can express the relationship between input and output. By adjusting the weights between neurons in the neural network, function fitting can also express the mapping relationship. The network formed has more perfect performance.

[0108] like Figure 4 As shown in the figure, the structure of a two-input single-output adaptive neural network fuzzy inference system can be presented in the form of neurons. The system consists of five intermediate layers, namely the fuzzification layer, fuzzy inference layer, normalization layer, rule output layer and merged output layer. In order to reflect different adaptive capabilities, nodes with parameters are represented by square nodes, and nodes without parameters are represented by circular nodes. Nodes with parameters will achieve ideal input-output mapping through the update method.

[0109] It is assumed that the fuzzy rule base contains two first-order if-then fuzzy rules of Takagi and Sugeno types.

[0110] Rule 1: (1)

[0111] Rule 2: (2)

[0112] (1) Fuzzy layer (Layer 1)

[0113] All nodes in the fuzzy layer are square nodes. Nodes have node functions ( )

[0114] (3)

[0115] In the formula is the input value, Is the input value associated with the node function In other words, for The membership function of Satisfaction quantifier degree, usually It is represented by a bell or Gaussian shape, with a maximum value of 1 and a minimum value of zero, for example:

[0116] (4)

[0117] or

[0118] (5)

[0119] In the formula is the parameter set and all the parameters in this layer are called premise parameters. During the update process, the values ​​of these parameters change, and the membership function will also change accordingly, so that The membership functions of various driving modes are shown above.

[0120] (2) Fuzzy reasoning layer (Layer 2)

[0121] The output of the nodes in the fuzzy inference layer represents the activation of the rules. Each node in this layer is a A circular node that multiplies the input signals and sends the product out. )

[0122] (6)

[0123] (3) Normalization layer (Layer 3)

[0124] The normalization layer is used to normalize the rule activation output by the fuzzy inference layer. Each node is a The circular node. The node calculates The ratio of the activation of a rule to the sum of the activations of all rules ( ):

[0125] (7)

[0126] (4) Rule output layer (Layer 4)

[0127] The rule output layer combines the rule activation output by the normalization layer with rules one and two in formulas 1 and 2 to obtain: )

[0128] (8)

[0129] In the formula is a parameter set, and all parameters in this layer are called subsequent parameters.

[0130] (5) Total output layer (Layer 5)

[0131] The single node of the total output layer is a The circular node calculates the total output as the sum of all input signals ( ),Right now:

[0132] (9)

[0133] So far, a structural model of a two-input and one-output neural network fuzzy inference system with two rules has been constructed. Its function is equivalent to that of a fuzzy inference system, but it can be updated in an adaptive way and is easy to expand. On this basis, the number of inputs, outputs and rules can be increased.

[0134] Optimize the energy efficiency of the thermal management system to avoid energy waste. While the thermal management system improves the efficiency of brake energy recovery, the refrigerant directly controls the motor temperature and also consumes energy. Therefore, optimize the control system and improve the economic efficiency of the thermal management system.

[0135] The economic efficiency of the thermal management system is expressed by the ratio of cooling energy consumption to braking energy recovery, i.e., energy consumption rate:

[0136] (10)

[0137] Where C is the energy consumption rate; E is the energy consumed by the thermal management system; and D is the energy rate of braking recovery.

[0138] The braking energy recovery efficiency of electric vehicles is the ratio of the energy recovered by the system to the energy generated by the vehicle deceleration, expressed as:

[0139] (11)

[0140] In the formula, Energy recovered by the system, The energy generated by braking the vehicle.

[0141] The objective function of thermal management system optimization can be expressed as follows:

[0142] (12)

[0143] In the formula, , is the weight coefficient, .

[0144] The particle swarm optimization algorithm is used to write a program, in which the electric vehicle model is called, a thermal management fuzzy controller is created and parameters are written. At the same time, the system transmits the real-time data required to calculate the fitness value to the particle swarm optimization algorithm program. Figure 5 A flowchart of the process is shown in Figure 1.

[0145] Firstly, the parameters of the membership function of the fuzzy controller are assigned, and the electric vehicle model and NEDC operating conditions are called in. After a NEDC operating condition, the two performance indicators of thermal management energy consumption and brake energy recovery effect are output to determine whether they have reached the optimum. If not, the particle swarm is updated, and the cycle is repeated until the optimum is reached to achieve the best economic goal.

[0146] Example 2

[0147] This embodiment relates to an adaptive fuzzy control system for improving the efficiency of braking energy recovery, to implement an adaptive fuzzy control method for improving the efficiency of braking energy recovery in embodiment 1, comprising:

[0148] Energy recovery module, used to control the motor to generate electricity and recover energy according to the braking force requirements during braking;

[0149] The motor cooling capacity calculation module is used to calculate the cooling capacity demand of the electric vehicle motor according to the motor temperature, motor retraction braking force, and fuzzy controller.

[0150] The cooling capacity modulation module is used in the thermal management system to dynamically adjust the cooling capacity according to the cooling capacity demand of the motor to maintain the optimal operating temperature of the motor.

[0151] The thermal management system optimization module uses the particle swarm optimization algorithm to optimize the thermal management system.

[0152] Example 3

[0153] like Figure 6 As shown, this embodiment relates to an adaptive fuzzy control device for improving the efficiency of braking energy recovery, including a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, it is used to implement an adaptive fuzzy control method for improving the efficiency of braking energy recovery in embodiment 1.

[0154] Each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0155] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. An adaptive fuzzy control method for improving braking energy recovery efficiency, characterized in that: include: S1: During braking, the motor is controlled to generate electricity and recover energy according to the braking force requirement; S2: Based on the motor temperature and the motor retraction braking force, the fuzzy controller calculates the cooling capacity requirement of the electric vehicle motor; S3: The thermal management system dynamically adjusts the cooling capacity according to the motor cooling capacity demand to maintain the optimal operating temperature of the motor; S4: Optimize the thermal management system using particle swarm optimization algorithm.

2. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 1, characterized in that: During braking described in step S1, according to the braking force requirement, the motor is controlled to generate power to recover energy, including: S11: The sensor detects the brake pedal position and obtains the driver's braking intention; S12: obtaining brake intensity information according to the brake pedal position sensor; S13: Calculate the ideal braking force based on the information of the braking intensity, vehicle speed sensor, and front and rear wheel weight; S14: Determine the battery SOC value, and if the battery SOC value is less than 0.9, perform brake energy recovery, otherwise perform friction braking deceleration; S15: Obtaining the maximum feedback braking force of each wheel according to the driving motor speed information; S16: comparing the maximum feedback braking force of each wheel and the ideal braking force of each wheel; S17: When the maximum feedback braking force of a wheel is greater than the ideal braking force of the wheel, the feedback braking force of the wheel provides the ideal braking force; S18: When the maximum feedback regenerative braking force of a wheel is less than the ideal braking force of the wheel, the wheel provides the maximum feedback regenerative braking force, and at the same time the mechanical braking system provides friction braking force to make the total braking force reach the ideal braking force.

3. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 2, characterized in that: The characteristic of the ideal braking force in step S13 is that when each wheel reaches the ideal braking force, the front and rear wheels of the electric vehicle are locked simultaneously.

4. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 1, characterized in that: The motor temperature described in step S2 is obtained by sensors, wherein the sensors include a motor temperature sensor, a motor current and voltage sensor, and an ambient temperature sensor.

5. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 1, characterized in that: The calculation of the return braking force in step S2 includes: S21: determining the magnitude of the return braking force according to the composition of the braking force; S22: If the wheel is decelerated by the friction braking force, the return braking force is 0; S23: If the wheel is decelerated and braked by the motor feedback braking force, the feedback braking force is the ideal braking force of the wheel at this time; S24: If the wheel is braked and decelerated by the maximum feedback braking force of the motor and the friction braking force, the return braking force is the maximum feedback braking force of the wheel motor at this time.

6. The adaptive fuzzy control method for improving the efficiency of braking energy recovery according to claim 1, characterized in that The fuzzy controller described in step S2 adopts an adaptive neural fuzzy control method. The adaptive neural fuzzy control method involves a two-input single-output adaptive neural network fuzzy inference system, which is composed of five intermediate layers, namely, fuzzification layer Layer1, fuzzy inference layer Layer2, normalization layer Layer3, rule output layer Layer4 and merged output layer Layer5; fuzzification layer Layer1 and rule output layer Layer4 are nodes with parameters, and fuzzy inference layer Layer2, normalization layer Layer3 and merged output layer Layer5 are nodes without parameters.

7. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 6, characterized in that: The assumed fuzzy rule base for the adaptive neuro-fuzzy inference system contains two first-order if-then fuzzy rules of the Takagi and Sugeno type: Rule 1: (1) Rule 2: (2) Blurred Layer Layer 1 All nodes in the fuzzy layer are square nodes. The nodes have the node function (3), where , (3) In the formula is the input value, Is the input value associated with the node function The label, that is for The membership function of Satisfaction quantifier degree, usually Represented by a bell or Gaussian shape, with a maximum of 1 and a minimum of zero: (4) or (5) In the formula is the parameter set and all the parameters in this layer are called premise parameters; during the update process, the values ​​of these parameters change, and the membership function will also change accordingly, so that The membership functions of various driving modes are shown above; Fuzzy reasoning layer Layer 2 The output of the nodes in the fuzzy inference layer represents the activation of the rules. Each node in this layer is a A circular node that multiplies the input signals and sends the product out, where , （6） Normalization layer Layer 3 The normalization layer is used to normalize the rule activation output by the fuzzy inference layer; each node is a The circular node of The node calculates The ratio of the activation of a rule to the sum of the activations of all rules, : （7） Rule output layer Layer 4 The rule output layer combines the rule activation output by the normalization layer with rule 1 and rule 2 in formulas (1) and (2) to obtain formula (8), where , （8） In the formula is a parameter set, and all parameters in this layer are called subsequent parameters; Total output layer Layer 5 The single node of the total output layer is a The circular node calculates the total output as the sum of all input signals, ,Right now: （9） 。 8. The adaptive fuzzy control method for improving braking energy recovery efficiency according to claim 1, characterized in that: Optimizing the thermal management system using a particle swarm optimization algorithm as described in step S4; The objective function of the particle swarm optimization algorithm is ;in: C is the energy consumption rate, which is defined as the ratio of cooling energy consumption to braking energy recovery; η is the braking energy recovery efficiency of electric vehicles, which is defined as the ratio of the energy recovered by the system to the energy generated by vehicle deceleration; , is the weight coefficient, and .

9. An adaptive fuzzy control system for improving the efficiency of braking energy recovery, characterized in that: include: Energy recovery module, used to control the motor to generate electricity and recover energy according to the braking force requirements during braking; The motor cooling capacity calculation module is used to calculate the cooling capacity demand of the electric vehicle motor according to the motor temperature, motor retraction braking force, and fuzzy controller; The cooling capacity modulation module is used in the thermal management system to dynamically adjust the cooling capacity according to the cooling capacity demand of the motor to maintain the optimal operating temperature of the motor; The thermal management system optimization module uses the particle swarm optimization algorithm to optimize the thermal management system.

10. An adaptive fuzzy control device for improving braking energy recovery efficiency, characterized in that: It comprises a memory and one or more processors, wherein the memory stores executable codes, and when the one or more processors execute the executable codes, they are used to implement an adaptive fuzzy control method for improving the efficiency of braking energy recovery as described in any one of claims 1-8.

Images