Brake energy recovery method and device based on chaotic disturbance adaptive genetic algorithm

Abstract

The application discloses a brake energy recovery method and device based on a chaotic disturbance adaptive genetic algorithm, and the method comprises the following steps: acquiring multi-source state parameters collected by multi-source sensors in a vehicle; constructing a comprehensive fitness objective function of vehicle brake energy recovery according to the collected multi-source state parameters; when the vehicle is in a braking state, encoding a regenerative torque sequence of the vehicle to form an initial population, and performing fitness calculation on individuals in the initial population according to the comprehensive fitness objective function; performing iterative optimization on the initial population based on adaptive crossover and mutation and chaotic disturbance to obtain an optimal individual of a final excellent population; obtaining a regenerative braking torque and a mechanical braking torque based on the optimal individual, forming a brake energy recovery strategy, and recovering energy of the current vehicle based on the brake energy recovery strategy. The application solves the problem that the brake energy recovery control mode in the prior art cannot realize global optimal distribution of brake energy under continuous driving conditions.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and in particular to a braking energy recovery method and device based on a chaotic perturbation adaptive genetic algorithm. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the limited energy density of power batteries and range anxiety remain key pain points restricting the popularization of electric vehicles. Regenerative braking technology can convert the inertial kinetic energy generated during vehicle deceleration into electrical energy stored in the power battery via the drive motor, effectively reducing overall vehicle energy consumption and extending driving range. It is currently an indispensable core energy-saving technology for new energy vehicles. Under driving conditions such as frequent braking, urban congestion, and long downhill slopes, braking energy can account for 15% to 25% of the total energy loss during vehicle operation. Therefore, rationally optimizing regenerative braking distribution strategies and improving energy recovery efficiency are of significant engineering importance for improving the economic performance of the vehicle.

[0003] Currently, there is extensive research on regenerative braking control strategies for electric vehicles both domestically and internationally. The mainstream control schemes are mainly divided into two categories: rule-based allocation methods and intelligent optimization algorithm control methods. Rule-based allocation methods are simple in structure and have low computational complexity, making them widely used in mass-produced vehicles. Intelligent optimization algorithms, with their multi-dimensional optimization capabilities, have become the mainstream direction for current academic research and technological iteration. However, existing technologies generally suffer from problems such as one-sided optimization logic, weak adaptability to operating conditions, and prominent contradictions arising from multi-objective coupling. The specific current status and shortcomings of these technologies are as follows: 1) Traditional rule-based braking allocation scheme: This type of scheme sets a fixed ratio threshold based on calibration experience and uses lookup table logic to simply divide the mechanical braking and regenerative braking torque allocation range according to braking intensity. This method is logically simple and has good real-time performance, but it relies excessively on manual calibration experience and cannot dynamically adjust the allocation ratio based on battery status, road adhesion conditions, and driving conditions. It can only achieve local optima at a single braking moment and cannot perform global optimization for continuous driving conditions, resulting in significant energy waste and low overall recovery efficiency. 2) Conventional intelligent algorithm optimization scheme: Existing research often uses fuzzy control, particle swarm optimization, and basic genetic algorithms to optimize regenerative braking torque allocation. Among them, the basic genetic algorithm has global random search capabilities and performs better than traditional rule-based algorithms, and is widely used in the field of braking energy recovery optimization. However, conventional genetic algorithms have inherent defects such as poor initial randomness of population distribution, reduced population diversity in the later stages of iteration, and fixed crossover mutation probabilities. They are prone to premature convergence, getting trapped in local optima, and iterative oscillations. The algorithm's convergence speed is slow and cannot meet the millisecond-level real-time computing requirements of onboard controllers. 3) Existing multi-objective optimization solutions: Currently, a few studies use multi-objective optimization algorithms to balance energy recovery efficiency and braking smoothness. However, their objective weights are mostly assigned fixed values, making it impossible to dynamically adjust the weight ratio according to the real-time vehicle status. Under harsh conditions such as low temperature, high SOC, and low-adhesion road surfaces, they cannot autonomously bias towards safety constraints or battery protection objectives, easily leading to problems such as battery overcharging, braking jerking, and wheel slippage, resulting in poor robustness under operating conditions. 4) Imperfect constraint mechanisms: Most existing optimization algorithms only consider the motor output torque constraint, ignoring multiple coupled constraints such as battery charging rate, road adhesion limit, and braking impact. The optimal torque generated by the algorithm iteration has the risk of exceeding the limit, resulting in poor engineering feasibility and difficulty in directly adapting to real vehicle control.

[0004] Therefore, a local optimization control method for regenerative braking torque of electric vehicles based on a fixed-weight genetic algorithm has emerged. This scheme uses a basic genetic algorithm as the core optimization architecture, takes a single braking process as the optimization time domain, selects maximizing recovered energy and minimizing braking impact as optimization objectives, and constructs a fitness function by manually setting fixed weight coefficients. The encoding method adopts binary encoding, and sets fixed crossover and mutation probabilities to complete the population iterative optimization. The constraint condition only limits the maximum regenerative torque of the motor, and iteratively outputs the torque distribution ratio of mechanical braking and regenerative braking under a single braking condition to complete the braking energy optimization control.

[0005] While this scheme possesses some multi-objective optimization capabilities and offers improved energy recovery compared to the traditional table lookup method, it still has significant technical shortcomings: 1) Short optimization time domain: It only performs local optimization for a single braking state, without combining the driving conditions before and after, and ignores the coupling effect of battery temperature rise and charge accumulation changes during continuous braking, which cannot achieve the global optimum of the whole vehicle; 2) Poor algorithm performance: It adopts fixed crossover and mutation probabilities, which leads to a lack of population diversity in the later stages of iteration, making it prone to premature convergence. The optimization solution oscillates significantly under complex conditions, and the real-time performance cannot meet the requirements of vehicle; 3) Fixed weights: It adopts fixed weights preset by humans, which cannot adaptively adjust the target bias according to battery SOC, temperature and road adhesion coefficient, resulting in insufficient safety and battery protection under harsh conditions; 4) Single constraint: It only sets a single constraint of motor torque, without considering the battery charging rate, road adhesion and braking smoothness boundaries. The torque command obtained by optimization has safety hazards and poor mass production adaptability.

[0006] Therefore, existing braking energy recovery control methods cannot achieve globally optimal distribution of braking energy under continuous driving conditions. Summary of the Invention

[0007] In view of this, the purpose of the present invention is to provide a braking energy recovery method and device based on a chaotic perturbation adaptive genetic algorithm, which aims to solve the problem that the existing braking energy recovery control methods cannot achieve the globally optimal allocation of braking energy under continuous driving conditions.

[0008] The embodiments of the present invention are implemented as follows: A braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm, applied to a vehicle controller, the method comprising: Acquire multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle; A comprehensive fitness objective function for vehicle braking energy recovery is constructed based on the collected multi-source state parameters. When the vehicle is braking, the regenerative torque sequence of the vehicle is encoded to form an initial population, and the fitness of individuals in the initial population is calculated using the comprehensive fitness objective function. The optimal individuals of the final excellent population are obtained by iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation. Based on the optimal individual, the regenerative braking torque and mechanical braking torque are obtained, a braking energy recovery strategy is formed, and energy is recovered from the current vehicle based on the braking energy recovery strategy.

[0009] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the step of iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population includes: Based on the four boundary constraints of motor torque, battery charging, road surface adhesion, and braking smoothness, all individuals in the population are verified, and individuals that do not meet the constraints are removed: The expression for the motor torque constraint is as follows: ; The expression for the battery charging constraint is: ; The expression for road surface adhesion constraints is: ; The expression for the braking ride comfort constraint is: ; In the formula, For the first The maximum regenerative torque that the motor is allowed to output at each braking sampling time. This is the total braking torque; For the overall vehicle weight; The tire's rolling radius, Provides real-time charging current for the power battery. This is the maximum allowable charging current for the power battery. Remaining battery power The road surface adhesion coefficient, It is the acceleration due to gravity. This refers to the braking impact.

[0010] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the expression for the comprehensive fitness objective function is: ; In the formula: , This refers to the start and end times of a continuous braking process. For instantaneous energy recovery; Braking impact; For battery temperature rise; This refers to the tire adhesion margin; , , , For dynamic weighting coefficients, For time.

[0011] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the formula for calculating the braking impact is: ; In the formula: To slow down the vehicle, For vehicle speed, The formula for calculating instantaneous energy recovery is: ; In the formula: For motor power generation efficiency; This provides regenerated torque for the motor. The angular velocity of the motor. The step size.

[0012] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the step of constructing the comprehensive fitness objective function for vehicle braking energy recovery further includes: Based on the vehicle's current operating conditions, the dynamic weighting coefficients are adaptively adjusted: ; In the formula: The road surface adhesion coefficient; , , for , , The corresponding initial baseline weights, , , This is the proportional adjustment factor for the operating condition. The road surface adhesion coefficient, This is the absolute value of the braking impact. Real-time temperature of the power battery. This indicates the remaining battery power.

[0013] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the step of iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population includes: When the population continuously fails to update its optimal value after a preset number of generations, a Logistic chaotic sequence is introduced to disrupt the distribution of individuals.

[0014] Furthermore, in the above-mentioned braking energy recovery method based on chaotic perturbation adaptive genetic algorithm, the adaptive crossover probability formula is:

[0015] The formula for adaptive mutation probability is:

[0016] The chaotic perturbation update formula is

[0017] In the formula: The adaptive crossover probability for the current individual. To preset the maximum crossover probability, To preset the minimum crossover probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. This represents the adaptive mutation probability of the current individual. To preset the maximum mutation probability, To preset the minimum mutation probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. for The normalized population individual variables of the th generation For the next generation of individual variables updated by chaotic mapping, These are the control parameters for Logistic chaotic bifurcation.

[0018] Another object of the present invention is to provide a braking energy recovery device based on a chaotic perturbation adaptive genetic algorithm, the device comprising: The acquisition module is used to acquire multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle. The module is used to construct a comprehensive fitness objective function for vehicle braking energy recovery based on the collected multi-source state parameters; The component module is used to encode the regenerative torque sequence of the vehicle to form an initial population when the vehicle is in a braking state, and to calculate the fitness of individuals in the initial population using a comprehensive fitness objective function. The iterative module is used to iteratively optimize the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individuals of the final excellent population. The recovery module is used to obtain the regenerative braking torque and mechanical braking torque based on the optimal individual, form a braking energy recovery strategy, and perform energy recovery on the current vehicle based on the braking energy recovery strategy.

[0019] Another object of the present invention is to provide a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0020] Another object of the present invention is to provide a vehicle including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the method described above.

[0021] This invention acquires multi-source state parameters of the vehicle from multiple sensors deployed within the vehicle; based on these parameters, it constructs a comprehensive fitness objective function for vehicle braking energy recovery, incorporating four optimization indicators: instantaneous recovered electrical energy, braking impact, battery temperature rise, and tire adhesion margin; when the vehicle is braking, an initial population is formed using a complete regenerative torque sequence under continuous braking time as coded individuals; fitness calculations are performed on each individual within the initial population based on the comprehensive fitness objective function, distinguishing between superior and inferior individuals based on their fitness; population evolution is completed using an adaptive crossover and mutation probability that automatically changes with fitness; and the population iteration stops when it stagnates. This paper introduces chaotic perturbations to enrich the population diversity, and performs global iterative optimization on the initial population to obtain the optimal individuals of the final excellent population. Finally, based on the optimal individuals, the regenerative braking torque at each time moment is analyzed, and combined with the total braking torque demand of the vehicle, the matching mechanical braking torque is calculated. A global braking energy recovery strategy adapted to the current continuous operating conditions is generated and energy recovery control is executed. It can cover the complete continuous braking time domain, take into account the coupling effects of multiple physical quantities, dynamically balance recovery efficiency, ride smoothness, battery protection and driving safety, and achieve multiple objectives. The algorithm iteration is not prone to premature convergence, and the real-time operation is adapted to the vehicle controller. The output torque command simultaneously meets multiple engineering boundary conditions of motor, battery, road surface, and ride conditions. It solves the problem that the existing braking energy recovery control methods cannot achieve the globally optimal allocation of braking energy under continuous driving conditions.

[0022] The embodiments of the present invention also have at least the following beneficial effects: (1) To address the limitations of traditional strategies in terms of optimization scope and poor adaptability to operating conditions, a multi-constraint global optimization model is constructed under continuous driving conditions to break the limitations of local optimization at a single time step and maximize the cumulative benefits of vehicle braking energy. (2) To address the shortcomings of the basic genetic algorithm, such as premature convergence and slow convergence speed, an improved strategy combining chaotic perturbation and adaptive crossover mutation is proposed to enhance population diversity, improve the algorithm's optimization accuracy and real-time response capability, and meet the computational requirements of the vehicle controller. (3) To address the problem that fixed weights cannot dynamically balance multiple objectives, a dynamic weight allocation mechanism based on vehicle status is designed. The optimization bias is adaptively switched according to SOC, temperature, and road adhesion coefficient, taking into account energy saving, comfort, and driving safety. (4) To address the issues of single constraints and poor implementation, a four-fold boundary constraint correction mechanism is established to conduct safety screening of iterative individuals, ensuring that the regenerative braking torque is always within the physically feasible domain, thereby improving the practicality of the algorithm in engineering. (5) Ultimately, without changing the original vehicle hardware structure, the system achieves comprehensive optimization of braking energy recovery efficiency, braking smoothness, battery life and driving stability, thus solving the industry pain points of limited range, poor braking experience and insufficient safety under harsh conditions for new energy vehicles. Attached Figure Description

[0023] Figure 1 This is a flowchart of the braking energy recovery method based on chaotic perturbation adaptive genetic algorithm in the first embodiment of the present invention; Figure 2 This is a structural block diagram of the braking energy recovery device based on the chaotic perturbation adaptive genetic algorithm in the third embodiment of the present invention.

[0024] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0025] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0026] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] Example 1 Please see Figure 1 The figure shows a braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm in the first embodiment of the present invention, the method including steps S10 to S14.

[0029] Step S10: Obtain multi-source state parameters of the vehicle from the multi-source sensors deployed in the vehicle.

[0030] The vehicle is equipped with a multi-source sensing system consisting of various types of sensors, all of which communicate with the vehicle controller via the vehicle's CAN bus. After the vehicle is powered on and in operation, all sensors continue to work, collecting various status data during vehicle operation according to a set sampling frequency.

[0031] The sensors selected in this embodiment include a brake pedal travel sensor, a vehicle speed sensor, a battery management system-related data acquisition sensor, a road surface adhesion detection sensor, a temperature acquisition sensor, and a motor speed sensor. Each sensor performs a specific task for data acquisition. The brake pedal travel sensor collects the brake pedal opening signal to determine the driver's braking intention; the vehicle speed sensor collects the vehicle's current speed in real time; the battery management system simultaneously collects the remaining battery charge, real-time charging current, and real-time battery temperature; the road surface adhesion detection sensor identifies the current road surface adhesion status and outputs the road surface adhesion coefficient; the temperature acquisition sensor assists in verifying battery temperature data; and the motor speed sensor collects the real-time speed of the integrated generator motor.

[0032] All multi-source status parameters collected by the sensors are transmitted to the vehicle controller in real time via the CAN bus. The vehicle controller filters, denoises, and performs numerical calibration on the received data, eliminating abnormal interference data generated during transmission and retaining valid and usable vehicle status parameters, providing an accurate data foundation for subsequent model building and algorithm calculation.

[0033] Step S11: Construct a comprehensive fitness objective function for vehicle braking energy recovery based on the collected multi-source state parameters.

[0034] In this process, after preprocessing the multi-source state parameters, the vehicle controller retrieves a preset comprehensive fitness objective function model. This objective function is the core evaluation model for multi-objective optimization in this embodiment of the invention, comprehensively considering four dimensions: energy recovery capability during braking, vehicle braking smoothness, power battery loss, and vehicle braking and driving safety. The vehicle controller substitutes all parameters collected and calibrated in the first step, such as vehicle speed, motor speed, remaining battery charge, battery temperature, road surface adhesion coefficient, and brake pedal opening, into the corresponding sub-item calculation formulas of the objective function to calculate the four evaluation indicators: instantaneous energy recovery, braking impact, battery temperature rise, and tire adhesion margin. Simultaneously, it dynamically configures the weight coefficients within the objective function based on the vehicle's real-time operating conditions, ultimately completing the construction of the comprehensive fitness objective function. This objective function serves as the sole criterion for subsequent genetic algorithms to evaluate the quality of individuals in the population; a higher function calculation result indicates better overall performance of the corresponding braking torque distribution scheme.

[0035] For example, this embodiment of the invention breaks away from traditional single-step optimization and establishes a four-objective coupled global optimization mathematical model in the continuous time domain. The optimization objectives include: energy recovery, braking smoothness, battery wear, and tire safety margin; specifically including: ; In the formula: , This refers to the start and end times of a continuous braking process. For instantaneous energy recovery; Braking impact; For battery temperature rise; This refers to the tire adhesion margin; , , , For dynamic weighting coefficients, For time.

[0036] This function is an integral-form multi-objective coupled function capable of global optimization under continuous braking conditions. It uses the integral over the continuous driving time period as the evaluation criterion to achieve optimal global cumulative benefit.

[0037] The formula for calculating braking impact is as follows: ; In the formula: To slow down the vehicle, For vehicle speed, The formula for calculating instantaneous energy recovery is: ; In the formula: For motor power generation efficiency; This provides regenerated torque for the motor. The angular velocity of the motor. The step size.

[0038] During actual operation, the vehicle controller continuously collects vehicle speed data and uses differential operations to solve for vehicle deceleration and braking impact. The calculation results are used for both the calculation of the comprehensive fitness objective function and the verification of braking ride comfort constraints. In this embodiment, the calculation step size is consistent with the vehicle controller's ten-millisecond control cycle. The calculation logic of the instantaneous energy recovery formula is as follows: first, the instantaneous power generation of the motor is calculated using the motor torque and angular velocity; then, the recovered energy within the corresponding time period is calculated by combining the time step; and the calculation result is directly substituted into the integral term of the comprehensive fitness objective function.

[0039] In addition, in some optional embodiments of the present invention, the step of constructing the comprehensive fitness objective function for vehicle braking energy recovery further includes: Based on the vehicle's current operating conditions, the dynamic weighting coefficients are adaptively adjusted: ; In the formula: The road surface adhesion coefficient; , , for , , The corresponding initial baseline weights, , , This is the proportional adjustment factor for the operating condition. The road surface adhesion coefficient, This is the absolute value of the braking impact. Real-time temperature of the power battery. This indicates the remaining battery power.

[0040] Specifically, the system automatically reduces energy weight and increases safety weight under conditions of high SOC, low temperature, and low adhesion, achieving adaptive offset based on operating conditions. In particular, the formula automatically reduces [the weight] when the remaining battery charge is high or the real-time battery temperature is high. The value of the energy recovery index is reduced, thus weakening the optimization priority of the energy recovery index and avoiding battery overcharging and high temperature damage. This refers to the absolute value of the braking impact, focusing only on the severity of the braking impact without distinguishing the direction of the impact. When the braking impact increases and the vehicle's jerking sensation intensifies, The value increases accordingly, prioritizing the optimization of braking smoothness indicators. Under low-adhesion road surfaces... The algorithm automatically increases speed and prioritizes vehicle braking stability.

[0041] Step S12: When the vehicle is in braking state, the regenerative torque sequence of the vehicle is encoded to form an initial population, and the fitness of individuals in the initial population is calculated using the comprehensive fitness objective function. The vehicle controller first determines the current operating condition of the vehicle based on parameters such as brake pedal opening and vehicle deceleration. If it detects that the driver has pressed the brake pedal and the vehicle has begun to decelerate, it determines that the vehicle has entered the braking state.

[0042] This invention employs real-number encoding to complete the population encoding process. Unlike traditional binary encoding, real-number encoding directly corresponds to torque values ​​in actual engineering, reducing data conversion steps and improving computational efficiency. Combining the operational characteristics of continuous vehicle braking, the regenerative braking torques of the motor at different sampling times within a continuous braking duration are arranged chronologically to form a regenerative torque sequence. Each regenerative torque sequence represents an individual in the genetic algorithm. The vehicle controller randomly generates multiple sets of regenerative torque sequences conforming to the basic numerical range according to a preset population size. All regenerative torque sequences together form the initial population of the genetic algorithm. After the initial population is constructed, each individual within the population is traversed. The regenerative braking torque values ​​in each set of regenerative torque sequences, along with the collected multi-source vehicle state parameters, are substituted into the constructed comprehensive fitness objective function. The comprehensive fitness value corresponding to each individual in the population is calculated one by one, and the fitness results of all individuals in the population are completely recorded for subsequent population iteration and selection.

[0043] Step S13: Based on adaptive crossover mutation and chaotic perturbation, iteratively optimize the initial population to obtain the optimal individual of the final excellent population.

[0044] The vehicle controller initiates an iterative optimization process based on the initial population and the corresponding individual fitness values. This embodiment of the invention abandons the traditional genetic algorithm's fixed crossover and mutation probabilities, and instead adopts an adaptive crossover and mutation mechanism, dynamically adjusting the crossover and mutation probabilities according to the fitness of each individual in the population.

[0045] For high-fitness, superior individuals, the probability of crossover and mutation is reduced to prevent random alteration of the optimal regenerative torque sequence. For low-fitness, inferior individuals, the probability of crossover and mutation is increased to accelerate their evolution. During population iteration, the vehicle controller monitors the overall evolutionary state of the population in real time. When it detects that the optimal fitness value of the population has not been updated for several consecutive generations, it is determined that the algorithm is trapped in a local optimum. At this point, the Logistic chaotic perturbation mechanism is activated, using a chaotic sequence to disrupt the distribution of existing individuals in the population, expanding the search range of the algorithm and helping it escape the local optimum.

[0046] The algorithm continues to iterate according to the above-mentioned rules of adaptive crossover mutation combined with chaotic perturbation until the algorithm reaches the preset maximum number of iterations, or the optimal fitness value of the population tends to stabilize and no longer increases. At this time, the iteration operation is stopped, and the population after the current iteration is the excellent population. The individual with the highest comprehensive fitness value is selected from the excellent population as the optimal individual obtained by optimization.

[0047] Step S14: Based on the optimal individual, the regenerative braking torque and mechanical braking torque are obtained, a braking energy recovery strategy is formed, and energy is recovered from the current vehicle based on the braking energy recovery strategy.

[0048] After the algorithm iterates to obtain the optimal individual, the vehicle controller analyzes the regenerative torque sequence within the optimal individual. Each value in the sequence corresponds to the regenerative braking torque that the motor needs to output at each sampling moment during continuous braking. The vehicle controller, combined with the total braking demand torque of the vehicle collected earlier, calculates the mechanical braking torque that the hydraulic-mechanical braking mechanism needs to output at each sampling moment through torque allocation logic. The analyzed regenerative braking torque and the calculated mechanical braking torque are combined according to time sequence to form a complete braking energy recovery strategy.

[0049] The vehicle controller sends regenerative braking torque commands to the integrated drive-generator motor and mechanical braking torque commands to the hydraulic-mechanical brake actuator via the CAN bus. The two actuators work together according to the commands. The integrated drive-generator motor operates in regenerative braking mode, converting the inertial kinetic energy generated during vehicle braking into electrical energy stored in the power battery, thus completing the braking energy recovery process. At the same time, the hydraulic-mechanical brake actuator cooperates to output braking force, ensuring that the vehicle decelerates smoothly according to the driver's braking intention.

[0050] In summary, the braking energy recovery method based on chaotic perturbation adaptive genetic algorithm in the above embodiments of the present invention acquires multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle; constructs a comprehensive fitness objective function for vehicle braking energy recovery based on the acquired multi-source state parameters, including four optimization indicators: instantaneous recovered electrical energy, braking impact, battery temperature rise, and tire adhesion margin; when the vehicle is in braking state, an initial population is formed using the complete regenerative torque sequence under continuous braking time as encoded individuals, and the fitness of each individual in the initial population is calculated based on the multi-source state parameters using the comprehensive fitness objective function; and based on the individual fitness, adaptive crossover mutation that automatically changes with fitness is adopted to distinguish between good and bad individuals. The algorithm probabilistically completes population evolution and introduces chaotic perturbations to enrich population diversity when iteration stagnates. It then performs global iteration to optimize the initial population, ultimately obtaining the optimal individuals from the best population. Finally, based on these optimal individuals, the regenerative braking torque at each moment is analyzed, and combined with the total braking torque demand of the vehicle, a matching mechanical braking torque is calculated. This generates a global braking energy recovery strategy adapted to the current continuous operating conditions and executes energy recovery control. This approach covers the entire continuous braking time domain, considers the coupling effects of multiple physical quantities, dynamically balances recovery efficiency, ensures ride smoothness, and protects battery and driving safety—multiple objectives. The algorithm is less prone to premature convergence during iteration, and its real-time computation is adapted to the vehicle controller. The output torque command simultaneously satisfies multiple engineering boundary conditions related to the motor, battery, road surface, and driving experience. This solves the problem that existing braking energy recovery control methods cannot achieve globally optimal allocation of braking energy under continuous driving conditions.

[0051] Example 2 This embodiment also proposes a braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm. The difference between the braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm in this embodiment and the braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm in Embodiment 1 is as follows: In the step of iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population: Based on the four boundary constraints of motor torque, battery charging, road surface adhesion, and braking smoothness, all individuals in the population are verified, and individuals that do not meet the constraints are removed: The expression for the motor torque constraint is as follows: ; The expression for the battery charging constraint is: ; The expression for road surface adhesion constraints is: ; The expression for the braking ride comfort constraint is: ; In the formula, For the first The maximum regenerative torque that the motor is allowed to output at each braking sampling time. This is the total braking torque; For the overall vehicle weight; The tire's rolling radius, Provides real-time charging current for the power battery. This is the maximum allowable charging current for the power battery. Remaining battery power The road surface adhesion coefficient, It is the acceleration due to gravity. This refers to the braking impact.

[0052] Among them, the motor torque constraint is used to limit the range of regenerative braking torque output by the motor at a single moment. The constraint rule requires that all regenerative braking torque values ​​must be greater than or equal to zero and less than or equal to the maximum regenerative torque of the motor at the current moment. A negative torque value means that the motor has entered the driving state and cannot achieve energy recovery. If the torque value exceeds the upper limit, it will cause the motor to be overloaded and damaged. Therefore, individuals that do not meet the constraint are directly eliminated. Battery charging constraints are used to protect the power battery and prevent high-current charging and overcharging. The constraint rules are divided into two parts: first, the real-time charging current of the power battery cannot exceed the maximum allowable charging current of the battery to prevent high-current charging from damaging the cells; second, the remaining battery charge cannot exceed 0.95%. When the battery charge is close to full charge, the energy recovery intensity is forcibly limited to prevent overcharging damage. If any individual's parameters do not meet either of these conditions, that individual will be eliminated. Road surface adhesion constraints are used to ensure the driving stability of a vehicle during braking and to prevent wheel lock-up and slippage. Braking smoothness constraints are used to evaluate the ride comfort during vehicle braking. The larger the absolute value of the impact value, the more drastic the change in vehicle deceleration, and the more obvious the jerking and pitching phenomena inside the vehicle. This constraint limits the braking impact to a range of -10 to +10 meters per cubic second. Once the braking impact of an individual exceeds this range, it indicates that the torque distribution scheme will seriously affect the ride experience, and such individuals will be directly eliminated.

[0053] In addition, in some optional embodiments of the present invention, the step of iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population includes: When the population continuously fails to update its optimal value after a preset number of generations, a Logistic chaotic sequence is introduced to disrupt the distribution of individuals.

[0054] In this system, after each population iteration, the vehicle controller records the maximum fitness value within the current population and compares it with the optimal fitness values ​​of previous generations. In this embodiment, three consecutive generations are set as the preset number of generations. When it is detected that the optimal fitness value of the population has not been updated for three consecutive generations, it can be determined that the current genetic algorithm has entered a local optimum, and adaptive crossover mutation alone cannot further improve the population quality. At this time, the vehicle controller automatically triggers Logistic chaotic sequence related operations, using the traversability and randomness of the chaotic sequence to scramble the distribution of all individuals within the current population, regenerating new population individuals, breaking the original local optimum state, and continuing iterative optimization. This mechanism, as a supplementary optimization method to adaptive crossover mutation, effectively improves the global search capability of the genetic algorithm, ensuring that the algorithm can find the globally optimal braking torque distribution scheme. For example, the adaptive crossover probability formula is as follows:

[0055] The formula for adaptive mutation probability is:

[0056] The chaotic perturbation update formula is

[0057] In the formula: The adaptive crossover probability for the current individual. To preset the maximum crossover probability, To preset the minimum crossover probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. This represents the adaptive mutation probability of the current individual. To preset the maximum mutation probability, To preset the minimum mutation probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. for The normalized population individual variables of the th generation For the next generation of individual variables updated by chaotic mapping, These are the control parameters for Logistic chaotic bifurcation.

[0058] In summary, the braking energy recovery method based on chaotic perturbation adaptive genetic algorithm in the above embodiments of the present invention acquires multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle; constructs a comprehensive fitness objective function for vehicle braking energy recovery based on the acquired multi-source state parameters, including four optimization indicators: instantaneous recovered electrical energy, braking impact, battery temperature rise, and tire adhesion margin; when the vehicle is in braking state, the complete regenerative torque sequence under continuous braking time sequence is used as the encoded individuals to form an initial population, and the fitness of each individual in the initial population is calculated based on the multi-source state parameters using the comprehensive fitness objective function; based on the individual fitness, the method distinguishes between good and bad individuals, and adopts an adaptive crossover and mutation probability that automatically changes with fitness. The algorithm completes population evolution efficiently and introduces chaotic perturbations to enrich population diversity when the population iteration stagnates. It then performs global iteration to optimize the initial population, obtaining the optimal individuals from the final superior population. Finally, based on the optimal individuals, the regenerative braking torque at each moment is analyzed, and combined with the total braking torque demand of the entire vehicle, a matching mechanical braking torque is calculated. This generates a global braking energy recovery strategy adapted to the current continuous operating conditions and executes energy recovery control. This approach covers the entire continuous braking time domain, considers the coupling effects of multiple physical quantities, dynamically balances recovery efficiency, ensures ride smoothness, and addresses multiple objectives related to battery protection and driving safety. The algorithm is less prone to premature convergence during iteration, and its real-time computation is adapted to the vehicle controller. The output torque command simultaneously satisfies multiple engineering boundary conditions related to the motor, battery, road surface, and driving experience. This solves the problem that existing braking energy recovery control methods cannot achieve globally optimal allocation of braking energy under continuous driving conditions.

[0059] Example 3 Please see Figure 2 The image shows a braking energy recovery device based on a chaotic perturbation adaptive genetic algorithm proposed in the third embodiment of the present invention. The device includes: The acquisition module 100 is used to acquire multi-source state parameters of the vehicle collected by multi-source sensors deployed in the vehicle. Module 200 is used to construct a comprehensive fitness objective function for vehicle braking energy recovery based on the collected multi-source state parameters; The component module 300 is used to encode the regenerative torque sequence of the vehicle to form an initial population when the vehicle is in a braking state, and to calculate the fitness of individuals in the initial population using a comprehensive fitness objective function. The iteration module 400 is used to iteratively optimize the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population. The recovery module 500 is used to obtain the regenerative braking torque and mechanical braking torque based on the optimal individual, form a braking energy recovery strategy, and perform energy recovery on the current vehicle based on the braking energy recovery strategy.

[0060] The functions or operation steps implemented by the above modules are largely the same as those in the above method embodiments, and will not be repeated here.

[0061] Example 4 In another aspect, the present invention provides a readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the steps of the method described in any one of the above embodiments one to two.

[0062] Example 5 In another aspect, the present invention provides a vehicle, wherein the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any one of the methods described in embodiments one to two above.

[0063] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0064] Those skilled in the art will understand that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequential list of executable instructions for implementing logical functions, and can be embodied in any computer-readable storage medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable storage medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0065] More specific examples (a non-exhaustive list) of computer-readable storage media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable storage media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0066] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0067] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0068] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A braking energy recovery method based on a chaotic perturbation adaptive genetic algorithm, characterized in that, The method, applied in a vehicle controller, includes: Acquire multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle; A comprehensive fitness objective function for vehicle braking energy recovery is constructed based on the collected multi-source state parameters. When the vehicle is braking, the regenerative torque sequence of the vehicle is encoded to form an initial population, and the fitness of individuals in the initial population is calculated using the comprehensive fitness objective function. The optimal individuals of the final excellent population are obtained by iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation. Based on the optimal individual, the regenerative braking torque and mechanical braking torque are obtained, a braking energy recovery strategy is formed, and energy is recovered from the current vehicle based on the braking energy recovery strategy.

2. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 1, characterized in that, In the step of iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individual of the final excellent population: Based on the four boundary constraints of motor torque, battery charging, road surface adhesion, and braking smoothness, all individuals in the population are verified, and individuals that do not meet the constraints are removed: The expression for the motor torque constraint is as follows: ; The expression for the battery charging constraint is: ; The expression for road surface adhesion constraints is: ; The expression for the braking ride comfort constraint is: ; In the formula, For the first The maximum regenerative torque that the motor is allowed to output at each braking sampling time. This is the total braking torque; For the overall vehicle weight; The tire's rolling radius, Provides real-time charging current for the power battery. This is the maximum allowable charging current for the power battery. Remaining battery power The road surface adhesion coefficient, It is the acceleration due to gravity. This refers to the braking impact.

3. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 1, characterized in that, The expression for the comprehensive fitness objective function is: ； In the formula: , This refers to the start and end times of a continuous braking process. For instantaneous energy recovery; Braking impact; For battery temperature rise; This refers to the tire adhesion margin; , , , For dynamic weighting coefficients, For time.

4. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 3, characterized in that, The formula for calculating braking impact is: ； In the formula: To slow down the vehicle, For vehicle speed, The formula for calculating instantaneous energy recovery is: ； In the formula: For motor power generation efficiency; This provides regenerated torque for the motor. The angular velocity of the motor. The step size.

5. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 3, characterized in that, The step of constructing the comprehensive fitness objective function for vehicle braking energy recovery also includes: Based on the vehicle's current operating conditions, the dynamic weighting coefficients are adaptively adjusted: ； In the formula: The road surface adhesion coefficient; , , for , , The corresponding initial baseline weights, , , This is the proportional adjustment factor for the operating condition. The road surface adhesion coefficient, This is the absolute value of the braking impact. Real-time temperature of the power battery. This indicates the remaining battery power.

6. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 1, characterized in that, The steps for iteratively optimizing the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individuals of the final excellent population include: When the population continuously fails to update its optimal value after a preset number of generations, a Logistic chaotic sequence is introduced to disrupt the distribution of individuals.

7. The braking energy recovery method based on chaotic perturbation adaptive genetic algorithm according to claim 6, characterized in that, The adaptive crossover probability formula is: The formula for adaptive mutation probability is: The chaotic perturbation update formula is In the formula: The adaptive crossover probability for the current individual. To preset the maximum crossover probability, To preset the minimum crossover probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. This represents the adaptive mutation probability of the current individual. To preset the maximum mutation probability, To preset the minimum mutation probability, This represents the overall fitness value of a single individual in the current population. The maximum fitness of all individuals in the current population. This represents the average fitness of all individuals in the current population. for The normalized population individual variables of the th generation For the next generation of individual variables updated by chaotic mapping, These are the control parameters for Logistic chaotic bifurcation.

8. A braking energy recovery device based on a chaotic perturbation adaptive genetic algorithm, characterized in that, The device, used in vehicle controllers, includes: The acquisition module is used to acquire multi-source state parameters of the vehicle from multi-source sensors deployed in the vehicle. The module is used to construct a comprehensive fitness objective function for vehicle braking energy recovery based on the collected multi-source state parameters; The component module is used to encode the regenerative torque sequence of the vehicle to form an initial population when the vehicle is in a braking state, and to calculate the fitness of individuals in the initial population using a comprehensive fitness objective function. The iterative module is used to iteratively optimize the initial population based on adaptive crossover mutation and chaotic perturbation to obtain the optimal individuals of the final excellent population. The recovery module is used to obtain the regenerative braking torque and mechanical braking torque based on the optimal individual, form a braking energy recovery strategy, and perform energy recovery on the current vehicle based on the braking energy recovery strategy.

9. A storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 7.

10. A vehicle, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the method as described in any one of claims 1 to 7.

Images