New energy battery charging and discharging energy optimization system and method based on machine learning

Abstract

The present invention discloses a new energy battery charging and discharging energy optimization system and method based on machine learning, which belongs to the field of battery charging and discharging technology, and includes a data acquisition module, a demand analysis module and an optimization module; the data acquisition module is used to collect various data related to the charging and discharging of the new energy battery in real time, and transmit the collected data to subsequent modules; the demand analysis module is used to analyze the collected data, calculate the user's daily minimum cruising range requirement, provide an accurate basis for battery charging and discharging optimization, and generate a charging and discharging rate range; the optimization module optimizes charging and discharging according to the needs of the electric vehicle and the battery status, so as to improve the battery life and the vehicle's cruising range, and achieve fast charging while avoiding damage to the battery.

Description

Technical Field

[0001] The present invention belongs to the technical field of battery charging and discharging, and relates to a new energy battery charging and discharging energy optimization system and method based on machine learning. Background Art

[0002] Batteries are the core components of electric vehicle energy storage systems. Their performance and stability are directly related to the range and safety of electric vehicles. The market also places higher demands on the performance and endurance of new energy batteries.

[0003] The existing patent application with publication number CN116908728A discloses a method for optimizing a battery charge and discharge regime, comprising the following steps: establishing an electrochemical model based on a charge and discharge cycle process according to battery materials, electrode structures and battery structures; adding a battery life attenuation factor to the electrochemical model to establish a life attenuation model; simulating battery charge and discharge tests to optimize the life attenuation model; and comparing the life attenuation trends of the battery under different charge and discharge regimes according to charging requirements to screen out the optimal battery charge and discharge regime.

[0004] Although the prior art obtains a better battery charging and discharging system and provides a reliability reference for battery life testing and battery design, it does not take into account the dynamic changes in battery parameters and the specific needs of users. For example, in actual use, the battery parameter values ​​will continue to change with driving time and distance, and different users have different usage scenarios and needs for electric vehicles. Therefore, this application provides a new energy battery charging and discharging energy optimization system and method based on machine learning. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a new energy battery charging and discharging energy optimization system and method based on machine learning, to obtain multi-source data in real time, to deeply analyze user driving habits, to scientifically fit battery performance parameters, to ensure the optimal balance between battery life and charging speed, and to optimize charging and discharging according to the actual needs of electric vehicles and battery status.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] New energy battery charging and discharging energy optimization system based on machine learning, including: data acquisition module, demand analysis module and optimization module;

[0008] The data acquisition module is used to collect various data related to the charging and discharging of new energy batteries in real time, and transmit the collected data to the demand analysis module and the optimization module;

[0009] The demand analysis module is used to process the collected data, analyze user needs, calculate the user's daily minimum mileage requirement, and generate a charge and discharge rate range;

[0010] The optimization module optimizes charging and discharging according to the needs of the electric vehicle and the battery status;

[0011] The specific steps of analyzing user needs include:

[0012] Real-time recording of electric vehicle acceleration and speed Over time The changes in the speed of the electric vehicle are calculated, and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed ;

[0013] Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient ; The expression is as follows:

[0014] ;

[0015] In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and ;

[0016] Set the safety margin factor to , calculate the minimum mileage required by the user per day ; The expression is as follows:

[0017]

[0018] In the formula, is the average mileage;

[0019] The specific steps to generate the charge and discharge rate range include:

[0020] Build fitting formulas including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression ; The expression is as follows:

[0021] ;

[0022] ;

[0023] In the formula, , , , is the fitting parameter;

[0024] The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , and construct constraints;

[0025] Using binary coding, the charge and discharge rate The value range of to divide;

[0026] Randomly generate a certain number of initial individuals;

[0027] Set the fitness function according to the constraints and optimization goals ;

[0028] Calculate the fitness value of each individual and use genetic algorithm for iterative optimization until the fitness value of the best individual in the group is continuously Generations no longer improve;

[0029] When the termination conditions are met, the charge and discharge rate range is generated .

[0030] Furthermore, the specific steps of analyzing user needs include:

[0031] Set the number of times a user has performed The second trip, The mileage of this trip is , , record The start time of the vehicle in the trip and end time , and calculate the The travel time of the trip ;

[0032] Calculate total mileage and total travel time ;

[0033] Record No. The start time of the stop and end time , and calculate the total parking time ;in, , The number of stops in a day;

[0034] Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time .

[0035] Furthermore, the optimization module is configured with a charging optimization strategy and a discharging optimization strategy;

[0036] The charging optimization strategy determines the charging mode to be performed by the battery by constructing a charging score model;

[0037] The discharge optimization strategy determines the remaining power at startup and adjusts the discharge current in real time during driving, and adjusts the discharge cut-off voltage according to the health status of the battery.

[0038] Furthermore, the specific steps of the charging optimization strategy include:

[0039] Comprehensively consider the battery's remaining capacity SOC, health status SOH and minimum cruising range Frequency of use , build a charging score model; the expression of the charging score model is as follows:

[0040] ;

[0041] In the formula, The remaining power when the battery is fully charged. is the remaining power threshold, is the threshold of good health status, The remaining power in the battery can support the driving range. , , , is the weight coefficient, and ;

[0042] Using the actual data of the battery, calculate the charging score required for the battery in the current state , and set the charging score threshold to ;

[0043] Based on the calculated charging score and charging score threshold The size relationship can be used to determine the charging method to be adopted;

[0044] like , using charging current Fast charging;

[0045] like , using charging current Perform trickle charge.

[0046] Furthermore, the specific steps of the discharge optimization strategy include:

[0047] When the electric car starts, get the remaining battery power in the current state , and calculate the theoretical driving mileage that the remaining battery power can support ; The expression is as follows:

[0048] ;

[0049] In the formula, The mileage per unit of battery power, is the rated capacity of the battery;

[0050] like , the remaining battery power is sufficient to meet the minimum driving range, according to the original discharge current Performing a discharge operation;

[0051] like , the remaining battery power is insufficient to meet the minimum cruising range, and the discharge current is adjusted to ; The expression is as follows:

[0052] .

[0053] Furthermore, the specific steps of the discharge optimization strategy also include:

[0054] Continuously obtain the remaining battery power during the driving of the electric vehicle And update in real time;

[0055] when , maintain the original discharge current for discharge operation; when , reduce the discharge current to ; is the remaining power threshold;

[0056] Continuously obtain the remaining battery power And update in real time, and calculate the current battery discharge cut-off voltage ;

[0057] The actual voltage of the battery is Discharge cut-off voltage For comparison; if , continue driving; if , take protective measures.

[0058] New energy battery charging and discharging energy optimization method based on machine learning, including:

[0059] Step S1: real-time collection of various data related to the charging and discharging of new energy batteries;

[0060] Step S2: Analyze the collected data and calculate the user's minimum daily mileage requirement;

[0061] Step S3: fitting the relationship expression between battery life, charging speed and charge and discharge rate, and generating the charge and discharge rate range;

[0062] Step S4: Optimize charging and discharging according to the needs of the electric vehicle and the battery status.

[0063] Furthermore, the step S2 specifically includes:

[0064] Set the number of times a user has performed The second trip, The mileage of this trip is , , record The start time of the vehicle in the trip and end time , and calculate the The travel time of the trip ;

[0065] Calculate total mileage and total travel time ;

[0066] Record No. The start time of the stop and end time , and calculate the total parking time ;in, , The number of stops in a day;

[0067] Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time ;

[0068] Real-time recording of electric vehicle acceleration and speed Over time The changes in the speed of the electric vehicle are calculated, and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed ;

[0069] Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient ; The expression is as follows:

[0070] ;

[0071] In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and ;

[0072] Set the safety margin factor to , calculate the minimum mileage required by the user per day ; The expression is as follows:

[0073]

[0074] In the formula, is the average mileage.

[0075] Furthermore, the step S3 specifically includes:

[0076] Build fitting formulas including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression ; The expression is as follows:

[0077] ;

[0078] ;

[0079] In the formula, , , , is the fitting parameter;

[0080] The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , and construct constraints;

[0081] Using binary coding, the charge and discharge rate The value range of to divide;

[0082] Randomly generate a certain number of initial individuals;

[0083] Set the fitness function according to the constraints and optimization goals ;

[0084] Calculate the fitness value of each individual and use genetic algorithm for iterative optimization until the fitness value of the best individual in the group is continuously Generations no longer improve;

[0085] When the termination conditions are met, the charge and discharge rate range is generated .

[0086] Beneficial effects of the present invention:

[0087] Taking into account the dynamic changes of battery parameters, by collecting multi-dimensional data such as electric vehicles, batteries, environment and charging stations in real time, the battery status and user driving habits are accurately grasped, providing a solid foundation for the optimization strategy; by deeply analyzing key information such as user mileage, time, acceleration, etc., it flexibly responds to the personalized needs of different users, not only evaluates the impact of driving habits on energy consumption, but also sets the minimum daily driving range required for users, ensuring that the driving needs of electric vehicles in different scenarios are met; at the same time, pay attention to battery life and charging speed, and use genetic algorithms to iteratively optimize the range of charging and discharging rates, which not only protects the battery but also improves the user experience; and based on the real-time battery status and user needs, the charging and discharging strategy is intelligently adjusted, which not only extends the battery life, but also improves the vehicle's driving range, achieving the dual goals of fast charging and battery protection; by comprehensively considering dynamic parameters and user needs, efficient optimization of the charging and discharging energy of new energy batteries is achieved, improving the overall performance and user satisfaction of electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 A structural diagram of the energy optimization system for charging and discharging new energy batteries based on machine learning;

[0089] Figure 2 Flowchart for analyzing user needs;

[0090] Figure 3 A flow chart for generating a range of charge and discharge rates;

[0091] Figure 4 Flowchart for charging optimization strategy;

[0092] Figure 5 Flowchart for the discharge optimization strategy;

[0093] Figure 6 This is a flow chart of the new energy battery charging and discharging energy optimization method based on machine learning. DETAILED DESCRIPTION

[0094] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0095] Example 1

[0096] refer to Figures 1 to 5 As shown, this embodiment introduces a new energy battery charging and discharging energy optimization system based on machine learning, including: a data acquisition module, a demand analysis module and an optimization module;

[0097] The data acquisition module is used to collect various data related to the charging and discharging of new energy batteries from multiple channels in real time, including electric vehicle data, battery data, environmental data and charging station charging data, and transmit the collected data to subsequent modules; among them, electric vehicles include electric two-wheelers and three-wheelers, and the batteries are sodium-ion batteries. The electric vehicle data is the real-time operation data collected by the on-board sensors configured in the electric vehicle, covering the driving speed, acceleration, driving distance and motor load of the electric vehicle, reflecting the dynamic performance of the electric vehicle under different road conditions and driving conditions. The battery data is the key battery parameters collected in real time by the dedicated sensors installed in the battery pack, including the remaining power, real-time voltage, current, temperature, state of charge and internal resistance. The environmental data is the data of the environment in which the electric vehicle is located, including ambient temperature, humidity and air pressure. Different ambient temperatures are likely to cause changes in the charging and discharging performance of the battery, thereby affecting the battery's endurance and life. The charging station charging data includes the start and end time of each charge, charging time, charging power, and charging amount, which is used to understand the charging behavior of electric vehicles at different times and under different charging facilities, and helps to analyze the impact of different charging methods on battery performance.

[0098] The demand analysis module is used to process the driving data of the user's electric vehicle, analyze user needs, quantify the user's daily driving habits, evaluate the impact of battery consumption and driving habits, calculate the driving habit coefficient by setting the driving acceleration and speed thresholds, and set the safety margin to determine the user's daily minimum mileage requirement, providing an accurate basis for battery charging and discharging optimization; and fit the relationship expression between battery life, charging speed and charging and discharging rate, determine the feasible range based on the minimum acceptable life and charging speed, and use genetic algorithms to iteratively optimize to generate a charging and discharging rate range that can protect the battery and meet the usage needs. According to the chemical properties and physical properties of the battery, a too high charging and discharging rate will put greater pressure on the internal structure of the battery, accelerate battery aging, and damage the battery life. Although a too low charging and discharging rate is relatively mild, it will reduce the charging speed and affect the user's experience.

[0099] The optimization module optimizes charging and discharging according to the needs of electric vehicles and the battery status to increase the battery life and vehicle range, and achieve fast charging while avoiding damage to the battery. The module is configured with charging optimization strategy and discharging optimization strategy. The charging optimization strategy determines the specific charging method of the battery by constructing a charging score model. The discharging optimization strategy determines the remaining power at startup and adjusts the discharge current in real time during driving. At the same time, the discharge cut-off voltage is adjusted according to the battery health status SOH to prevent excessive discharge of the battery.

[0100] Going further, the specific steps for analyzing user needs include:

[0101] Set the number of times a user has performed The second trip, The mileage of this trip is , , each time the electric vehicle is started, the on-board sensor starts to record the initial value of the mileage. When the electric vehicle is turned off at the end of the trip, the final value of the mileage is recorded. The difference between the initial value and the final value is the mileage of each trip. At this time, the number of trips in a day is also Second, record The start time of the vehicle start in this trip and the end time of closing the vehicle , and calculate the The travel time of the trip , the expression is as follows:

[0102] ;

[0103] Calculate the total mileage of the electric vehicle on that day and total travel time , the expression is as follows:

[0104] ;

[0105] ;

[0106] Set the number of stops per day to , Each time you start the car, it is counted as a stop. However, the stop after the last trip is not counted. The start time of the stop and end time , , and calculate the total parking time of the electric vehicle on that day , the expression is as follows:

[0107] ;

[0108] Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time , the expression is as follows:

[0109] ;

[0110] ;

[0111] ;

[0112] In the formula, In the statistical interval Total mileage per day, In the statistical interval Total driving time for the day, In the statistical interval The total parking time of the day; by averaging the data over a period of time, it can more accurately grasp the user's daily driving habits and vehicle usage, help analyze the frequency and consumption of battery use, and understand the use and idleness of the vehicle in different time periods. For example, the average mileage can be used to estimate the approximate battery consumption, and the average parking time can be used to understand the idle time of the vehicle in different time periods, thereby providing a basis for the reasonable arrangement of vehicle use;

[0113] Using on-board sensors to record the acceleration of electric vehicles in real time and speed Over time The changes in the speed of the electric vehicle are calculated and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed , the expression is as follows:

[0114] ;

[0115] ;

[0116] in, For the Day's trip, For the Day's trip, For the Tianzhongdi The end time of starting the vehicle in the trip, For the Tianzhongdi The starting time of the vehicle start in the trip;

[0117] Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient , to evaluate the impact of driving habits on energy consumption, the expression is as follows:

[0118] ;

[0119] In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and , can be freely set by those skilled in the art. In this embodiment, Different driving habits will lead to different energy consumption of vehicles. Aggressive driving habits will increase energy consumption and require a higher range.

[0120] Set the safety margin factor to ,and , calculate the minimum mileage required by the user per day , the expression is as follows:

[0121] ;

[0122] In the formula, a certain safety margin is added to cope with emergencies, such as temporary changes in routes and increased mileage due to road congestion, to ensure that electric vehicles have enough power to meet normal driving needs within a day.

[0123] Furthermore, the specific steps of generating the charge and discharge rate range include:

[0124] Select multiple groups of sodium-ion batteries of the same specifications, record their lifespan and charging speed at different charge and discharge rates, and use programming tools such as Python's NumPy and SciPy libraries to fit the data through nonlinear least squares method to generate fitting formulas, including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression , the expression is as follows:

[0125] ;

[0126] ;

[0127] In the formula, , , , is the fitting parameter;

[0128] The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , according to the fitting formula and constraints, determine the feasible range of charge and discharge rate to ensure that the optimization results meet the performance requirements in practical applications; the constraints are as follows:

[0129] ;

[0130] ;

[0131] Using binary coding, the charge and discharge rate The value range of Divide into discrete binary strings, where , are the lower and upper limits of the charge and discharge rate, respectively. is the length of the binary code;

[0132] A certain number of initial individuals (chromosomes) are randomly generated to represent different candidate values ​​of charge and discharge rates;

[0133] Set the fitness function according to the constraints and optimization goals , to reflect the individual's ability to meet constraints and optimization goals. The optimization goal is to find the optimal charge and discharge rate range that takes into account both battery life and charging speed. The expression is as follows:

[0134] ;

[0135] Calculate the fitness value of each individual and use genetic algorithm for iterative optimization; including: use roulette wheel selection method to select excellent individuals from the current population to enter the next generation; pair the selected individuals in a way that the adjacent individuals are paired to produce offspring individuals; perform mutation operation on each offspring individual to increase the diversity of the population;

[0136] Repeat the above steps until the fitness value of the best individual in the group is continuously The generation is no longer improved, and the expression is as follows:

[0137] ;

[0138] In the formula, is the charge and discharge rate value of the current iteration; For the The fitness value of the best individual at the iteration;

[0139] When the termination condition is met, a set of optimal individuals is obtained to generate a range of charge and discharge rates. , is the optimal range under the constraints of battery life and charging speed, where and These are the minimum and maximum values ​​of the charging rate that are ultimately determined.

[0140] Furthermore, the specific steps of the charging optimization strategy include:

[0141] In order to accurately adjust the charging parameters according to the battery status and user needs, a charging score model is constructed to calculate the actual score required for the battery in the current state. , used to determine whether the battery is fast charged or trickle charged, taking into account the battery's remaining capacity SOC, health status SOH and minimum cruising range Frequency of use , the expression is as follows:

[0142] ;

[0143] In the formula, The remaining power when the battery is fully charged. is the remaining power threshold, which is used to determine whether the remaining battery power is at a low level. is the health status threshold, which is used to determine whether the battery is in good health. The remaining power in the battery can support the driving range. , , , , is the weight coefficient, and , defined by those skilled in the art, in this embodiment, , ;

[0144] Using the actual data of the battery, calculate the charging score required for the battery in the current state , and set the charging score threshold to ;

[0145] Based on the calculated charging score and charging score threshold The size relationship can be used to determine which charging method to use; if The remaining battery power is relatively low, the battery is in poor health, the battery life requirement is high, or the usage frequency is high, indicating that the battery needs to be replenished quickly and a larger charging current is used. Fast charging; if , the battery is in a relatively good condition and does not require fast charging. In order to protect the battery, a smaller charging current is used Perform trickle charge. and It is determined by those skilled in the art in combination with actual conditions.

[0146] Furthermore, the charging current and All by the charge and discharge rate range OK, the expression is as follows:

[0147] ;

[0148] ;

[0149] In the formula, is the rated capacity of the battery.

[0150] Furthermore, the specific steps of the discharge optimization strategy include:

[0151] When the electric car starts, immediately obtain the remaining battery power in the current state , and calculate the theoretical driving mileage that the remaining battery power can support , to determine whether the remaining power meets the minimum cruising range; the expression is as follows:

[0152] ;

[0153] In the formula, The driving distance per unit of battery power;

[0154] like , the remaining battery power is sufficient to meet the minimum driving range, according to the original discharge current Perform discharge operation; if , the remaining battery power is insufficient to meet the minimum cruising range, and the discharge current is adjusted to , further reducing the discharge current to extend the vehicle's mileage; the expression is as follows:

[0155] ;

[0156] Continuously obtain the remaining battery power during the driving of the electric vehicle And update in real time to further adjust the discharge current according to the latest remaining power of the battery. is constantly updated, and adjustments are also constantly updated; when , keep the original discharge current for discharge operation, the discharge current at this time is (If the battery is initially determined to be sufficient) or (If the initial judgment is that the battery is insufficient); when , reduce the discharge current to , to protect the battery, the expression is as follows:

[0157] ;

[0158] Continuously obtain the remaining battery power And update in real time, and calculate the current battery discharge cut-off voltage ,along with The discharge cut-off voltage should be appropriately increased to avoid damage to the battery caused by excessive discharge; the expression is as follows:

[0159] ;

[0160] In the formula, is the initial discharge cut-off voltage of the battery, is the adjustment coefficient, which can be determined by those skilled in the art;

[0161] During the entire driving process, the actual voltage of the battery is Discharge cut-off voltage For comparison; if , continue driving; if , immediately take protective measures, such as further reducing the discharge current or cutting off the discharge circuit, to prevent the battery from over-discharging.

[0162] Example 2

[0163] See also Figure 6 Another embodiment provided by the present invention is a new energy battery charging and discharging energy optimization method based on machine learning, comprising the following steps:

[0164] Step S1: collecting various data related to the charging and discharging of new energy batteries from multiple channels in real time, including electric vehicle data, battery data, environmental data, and charging station charging data;

[0165] Step S2: Analyze the driving data of the user's electric vehicle, quantify the user's daily driving habits, and determine the user's minimum daily mileage requirement;

[0166] Step S3: Fitting the relationship expression between battery life, charging speed and charge and discharge rate, determining the feasible range according to the minimum acceptable life and charging speed, and using genetic algorithm to iteratively optimize and generate the charge and discharge rate range;

[0167] Step S4: Optimize charging and discharging according to the needs of the electric vehicle and the battery status to increase the battery life and the vehicle's cruising range, and achieve fast charging while avoiding damage to the battery.

[0168] Furthermore, step S2 specifically includes:

[0169] Set the number of times a user has performed The second trip, The mileage of this trip is , , each time the electric vehicle is started, the on-board sensor starts to record the initial value of the mileage. When the electric vehicle is turned off at the end of the trip, the final value of the mileage is recorded. The difference between the initial value and the final value is the mileage of each trip. At this time, the number of trips in a day is also Second, record The start time of the vehicle start in this trip and the end time of closing the vehicle , and calculate the The travel time of the trip , the expression is as follows:

[0170] ;

[0171] Calculate the total mileage of the electric vehicle on that day and total travel time , the expression is as follows:

[0172] ;

[0173] ;

[0174] Set the number of stops per day to , Each time you start the car, it is counted as a stop. However, the stop after the last trip is not counted. The start time of the stop and end time , , and calculate the total parking time of the electric vehicle on that day , the expression is as follows:

[0175] ;

[0176] Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time , the expression is as follows:

[0177] ;

[0178] ;

[0179] ;

[0180] In the formula, In the statistical interval Total mileage per day, In the statistical interval Total driving time for the day, In the statistical interval The total parking time of the day; by averaging the data over a period of time, it can more accurately grasp the user's daily driving habits and vehicle usage, help analyze the frequency and consumption of battery use, and understand the use and idleness of the vehicle in different time periods. For example, the average mileage can be used to estimate the approximate battery consumption, and the average parking time can be used to understand the idle time of the vehicle in different time periods, thereby providing a basis for the reasonable arrangement of vehicle use;

[0181] Using on-board sensors to record the acceleration of electric vehicles in real time and speed Over time The changes in the speed of the electric vehicle are calculated and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed , the expression is as follows:

[0182] ;

[0183] ;

[0184] in, For the Day's Second trip, For the Day's Second trip, For the Tianzhongdi The end time of starting the vehicle in the trip, For the Tianzhongdi The starting time of the vehicle start in the trip;

[0185] Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient , to evaluate the impact of driving habits on energy consumption, the expression is as follows:

[0186] ;

[0187] In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and , can be freely set by those skilled in the art. In this embodiment, Different driving habits will lead to different energy consumption of vehicles. Aggressive driving habits will increase energy consumption and require a higher range.

[0188] Set the safety margin factor to ,and , calculate the minimum mileage required by the user per day , the expression is as follows:

[0189] ;

[0190] In the formula, a certain safety margin is added to cope with emergencies, such as temporary changes in routes and increased mileage due to road congestion, to ensure that electric vehicles have enough power to meet normal driving needs within a day.

[0191] Furthermore, step S3 specifically includes:

[0192] Select multiple groups of sodium-ion batteries of the same specifications, record their lifespan and charging speed at different charge and discharge rates, and use programming tools such as Python's NumPy and SciPy libraries to fit the data through nonlinear least squares method to generate fitting formulas, including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression , the expression is as follows:

[0193] ;

[0194] ;

[0195] In the formula, , , , is the fitting parameter;

[0196] The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , according to the fitting formula and constraints, determine the feasible range of charge and discharge rate to ensure that the optimization results meet the performance requirements in practical applications; the constraints are as follows:

[0197] ;

[0198] ;

[0199] Using binary coding, the charge and discharge rate The value range of Divide into discrete binary strings, where is the length of the binary code;

[0200] A certain number of initial individuals (chromosomes) are randomly generated to represent different candidate values ​​of charge and discharge rates;

[0201] Set the fitness function according to the constraints and optimization goals , to reflect the individual's ability to meet constraints and optimization goals. The optimization goal is to find the optimal charge and discharge rate range that takes into account both battery life and charging speed. The expression is as follows:

[0202] ;

[0203] Calculate the fitness value of each individual and use genetic algorithm for iterative optimization; including: use roulette wheel selection method to select excellent individuals from the current population to enter the next generation; pair the selected individuals in a way that the adjacent individuals are paired to produce offspring individuals; perform mutation operation on each offspring individual to increase the diversity of the population;

[0204] Repeat the above steps until the fitness value of the best individual in the group is continuously The generation is no longer improved, and the expression is as follows:

[0205] ;

[0206] In the formula, is the charge and discharge rate value of the current iteration; For the The fitness value of the best individual at the iteration;

[0207] When the termination condition is met, a set of optimal individuals is obtained to generate a range of charge and discharge rates. , is the optimal range under the constraints of battery life and charging speed, where and These are the minimum and maximum values ​​of the charging rate that are ultimately determined.

[0208] In summary, the present invention collects multi-dimensional data of electric vehicles, batteries, environment and charging stations in real time, quantifies the user's range requirements based on the user's historical usage information, and uses a genetic algorithm to determine the appropriate range of charge and discharge rates in combination with the relationship between battery life and charging speed; and implements charging and discharging optimization strategies based on the actual needs of the electric vehicle and the battery status; when charging, the best charging method is selected based on the charging score model; when discharging, the discharge parameters are dynamically adjusted during starting and driving, and the discharge cut-off voltage is adjusted according to the battery health status to ensure that the battery is charged and discharged efficiently and safely, extend the battery life and improve the range of the electric vehicle.

[0209] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A new energy battery charging and discharging energy optimization system based on machine learning, characterized in that: include: Data collection module, demand analysis module and optimization module; The data acquisition module is used to collect various data related to the charging and discharging of new energy batteries in real time, and transmit the collected data to the demand analysis module and the optimization module; The demand analysis module is used to process the collected data, analyze user needs, calculate the user's daily minimum mileage requirement, and generate a charge and discharge rate range; The optimization module optimizes charging and discharging according to the needs of the electric vehicle and the battery status; The specific steps of analyzing user needs include: Real-time recording of electric vehicle acceleration and speed Over time The changes in the speed of the electric vehicle are calculated and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed ; Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient ; The expression is as follows: ； In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and ; Set the safety margin factor to , calculate the minimum mileage required by the user per day ; The expression is as follows: ； In the formula, is the average mileage; The specific steps to generate the charge and discharge rate range include: Build fitting formulas including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression ; The expression is as follows: ； ； In the formula, , , , is the fitting parameter; The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , and construct constraints; Using binary coding, the charge and discharge rate The value range of to divide; Randomly generate a certain number of initial individuals; Set the fitness function according to the constraints and optimization goals ; Calculate the fitness value of each individual and use genetic algorithm for iterative optimization until the fitness value of the best individual in the group is continuously Generations no longer improve; When the termination conditions are met, the charge and discharge rate range is generated .

2. The new energy battery charging and discharging energy optimization system based on machine learning according to claim 1 is characterized in that: The specific steps of analyzing user needs also include: Set the number of times a user has performed The second trip, The mileage of this trip is , , record The starting time of the vehicle in the trip and end time , and calculate the The travel time of the trip ; Calculate total mileage and total travel time ; Record No. The start time of the stop and end time , and calculate the total parking time ;in, , The number of stops in a day; Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time .

3. The new energy battery charging and discharging energy optimization system based on machine learning according to claim 2 is characterized in that: The optimization module is configured with a charging optimization strategy and a discharging optimization strategy; The charging optimization strategy determines the charging mode to be performed by the battery by constructing a charging score model; The discharge optimization strategy determines the remaining power at startup and adjusts the discharge current in real time during driving, and adjusts the discharge cut-off voltage according to the health status of the battery.

4. The new energy battery charging and discharging energy optimization system based on machine learning according to claim 3 is characterized in that: The specific steps of the charging optimization strategy include: Comprehensively consider the battery's remaining capacity SOC, health status SOH and minimum cruising range Frequency of use , build a charging score model; the expression of the charging score model is as follows: ； In the formula, The remaining power when the battery is fully charged. is the remaining power threshold, is the threshold of good health status, The remaining power in the battery can support the driving range. , , , is the weight coefficient, and ; Using the actual data of the battery, calculate the charging score required for the battery in the current state , and set the charging score threshold to ; Based on the calculated charging score and charging score threshold The size relationship can be used to determine the charging method to be adopted; like , using charging current Fast charging; like , using charging current Perform trickle charge.

5. The new energy battery charging and discharging energy optimization system based on machine learning according to claim 4 is characterized in that: The specific steps of the discharge optimization strategy include: When the electric car starts, get the remaining power of the battery in the current state , and calculate the theoretical driving mileage that the remaining battery power can support ; The expression is as follows: ； In the formula, The mileage per unit of battery power, is the rated capacity of the battery; like , the remaining battery power is sufficient to meet the minimum driving range, according to the original discharge current Performing a discharge operation; like , the remaining battery power is insufficient to meet the minimum cruising range, and the discharge current is adjusted to ; The expression is as follows: 。 6. The new energy battery charging and discharging energy optimization system based on machine learning according to claim 5 is characterized in that: The specific steps of the discharge optimization strategy also include: Continuously obtain the remaining battery power during the driving of the electric vehicle And update in real time; when , maintain the original discharge current for discharge operation; when , reduce the discharge current to ; is the remaining power threshold; Continuously obtain the remaining battery power And update in real time, and calculate the current battery discharge cut-off voltage ; The actual voltage of the battery is Discharge cut-off voltage For comparison; if , continue driving; if , take protective measures.

7. A new energy battery charging and discharging energy optimization method based on machine learning, which is implemented based on a new energy battery charging and discharging energy optimization system based on machine learning as described in any one of claims 1 to 6, characterized in that: include: Step S1: real-time collection of various data related to the charging and discharging of new energy batteries; Step S2: Analyze the collected data and calculate the user's minimum daily mileage requirement; Step S3: fitting the relationship expression between battery life, charging speed and charge and discharge rate, and generating the charge and discharge rate range; Step S4: Optimize charging and discharging according to the needs of the electric vehicle and the battery status.

8. The method for optimizing charging and discharging energy of new energy batteries based on machine learning according to claim 7 is characterized in that: The step S2 specifically includes: Set the number of times a user has performed The second trip, The mileage of this trip is , , record The starting time of the vehicle in the trip and end time , and calculate the The travel time of the trip ; Calculate total mileage and total travel time ; Record No. The start time of the stop and end time , and calculate the total parking time ;in, , The number of stops in a day; Set the statistical interval to Day, The number of trips per day is , , calculate the average mileage , average driving time and average parking time ; Real-time recording of electric vehicle acceleration and speed Over time The changes in the speed of the electric vehicle are calculated, and the average acceleration of the electric vehicle is calculated within the statistical interval. and average speed ; Set the driving acceleration threshold to , driving speed threshold is , calculate the user's driving habit coefficient ; The expression is as follows: ； In the formula, is the weight coefficient of acceleration, is the weight coefficient of speed, and ; Set the safety margin factor to , calculate the minimum mileage required by the user per day ; The expression is as follows: ； In the formula, is the average mileage.

9. The method for optimizing charging and discharging energy of new energy batteries based on machine learning according to claim 8 is characterized in that: The step S3 specifically includes: Build fitting formulas including battery life and charge and discharge rate The relational expression , Charging speed and charge and discharge rate The relational expression ; The expression is as follows: ； ； In the formula, , , , is the fitting parameter; The minimum acceptable lifetime of sodium-ion batteries is set to , the minimum acceptable charging speed is , and construct constraints; Using binary coding, the charge and discharge rate The value range of to divide; Randomly generate a certain number of initial individuals; Set the fitness function according to the constraints and optimization goals ; Calculate the fitness value of each individual and use genetic algorithm for iterative optimization until the fitness value of the best individual in the group is continuously Generations no longer improve; When the termination conditions are met, the charge and discharge rate range is generated .

Images