Intelligent charging control method and system for pure electric vehicle

Abstract

The embodiment of the invention provides an intelligent charging control method and system for a pure electric vehicle, and the method comprises the steps: obtaining battery charging parameters which comprise a battery parameter, a historical charging parameter and a charging environment parameter; the battery charging parameters are input into a neural network model for model training, an algorithm module of the model training comprises a temperature rise risk index, a voltage sudden change trend and constant-current and constant-voltage switching, and a basic risk decision is output and comprises a temperature rise intervention index, a voltage trend index and a constant-current and constant-voltage inflection point; and after the charging period of the battery is detected, an empirical data set in the charging period is obtained, the decision value of the basic risk decision is determined in combination with the empirical data set and an evaluation algorithm, and the neural network model is updated through the decision value.

Description

Technical Field

[0001] This invention relates to the field of vehicle battery technology, and in particular to an intelligent charging control method and system for pure electric vehicles. Background Technology

[0002] As the market share of pure electric vehicles increases, electric vehicle batteries are also developing. Electric vehicle batteries are chemical energy storage devices that provide driving power for electric vehicles, and they are divided into two categories: rechargeable batteries and fuel cells. Rechargeable batteries store electrical energy through reversible electrochemical reactions. The entire system consists of multiple battery packs and corresponding management systems, high-voltage circuits, low-voltage circuits, and mechanical assemblies. Types include lead-acid batteries, nickel-metal hydride batteries, and lithium-ion batteries, among which lithium-ion batteries dominate the market due to their high energy density and long cycle life.

[0003] Compared to traditional gasoline vehicles, electric vehicles powered by batteries can provide power through a charging cycle, which can save resources and reduce environmental pollution.

[0004] However, in the existing technology, the following fixed defects in the charging protection of pure electric vehicle batteries have not yet been resolved, including battery protection delay caused by static protection lag, thermal runaway caused by poor high temperature adaptability, and battery life degradation problem equivalent to lack of current balance, which reduces the user experience. Summary of the Invention

[0005] To address the problems existing in the prior art, embodiments of the present invention provide a method and system for intelligent charging control of pure electric vehicles.

[0006] This invention provides an intelligent charging control method for pure electric vehicles, the method comprising:

[0007] Obtain battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters.

[0008] The battery charging parameters are input into a neural network model for model training. The algorithm module for model training includes a temperature rise risk index, a voltage change trend, and a constant current / constant voltage switching mechanism. The model outputs basic risk decisions, which include a temperature rise intervention index, a voltage trend index, and a constant current / constant voltage inflection point.

[0009] After the battery has undergone a charging cycle, an empirical data set within the charging cycle is acquired. Combining the empirical data set with the evaluation algorithm, the decision value of the basic risk decision is determined, and the neural network model is updated based on the decision value.

[0010] In one embodiment, the empirical data set includes:

[0011] Initial state, decision action, action reward, and new state.

[0012] In one embodiment, the calculation formula for determining the decision value of the basic risk decision by combining the empirical data set and the evaluation algorithm includes:

[0013] A t =R+γ*V(S')-V(S)

[0014] Among them, A t Let R be the evaluation value of the decision action, γ be the action reward, γ be the discount factor, V be the value prediction of the evaluation algorithm, S' be the new state, and S be the initial state.

[0015] In one embodiment, the calculation formula for updating the neural network model using the decision value includes:

[0016] L(θ)=E[min(r(θ)*A t ,clip(r(θ),1-ε,1+ε)*A t )]

[0017] Where r(θ) is the probability of the new decision / the probability of the basic risk decision, the Clip function restricts r(θ) to the range of [1-ε, 1+ε], and E is the batch sampling average.

[0018] In one embodiment, the action reward includes:

[0019] Rewards include charging speed, battery health, safety risk penalties, and user satisfaction.

[0020] In one embodiment, the method further includes:

[0021] Based on the trigger condition values ​​in the basic risk decision-making, four risk protection levels are determined during the battery charging process, and corresponding response measures are configured.

[0022] This invention provides an intelligent charging control system for pure electric vehicles, the system comprising:

[0023] The parameter module is used to acquire battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters.

[0024] The training module is used to input the battery charging parameters into the neural network model for model training. The algorithm module for model training includes temperature rise risk index, voltage change trend and constant current and constant voltage switching, and outputs basic risk decisions. The basic risk decisions include: temperature rise intervention index, voltage trend index and constant current and constant voltage inflection point.

[0025] The update module is used to detect the battery after it has gone through a charging cycle, obtain the experience data set within the charging cycle, combine the experience data set with the evaluation algorithm, determine the decision value of the basic risk decision, and update the neural network model based on the decision value.

[0026] In one embodiment, the system further includes:

[0027] The protection module is used to determine the four-level risk protection level during the battery charging process by comprehensively considering the trigger condition values ​​in the basic risk decision, and to configure the corresponding response measures.

[0028] This invention provides an electronic device, including a processor and a memory;

[0029] The processor is connected to the memory;

[0030] The memory is used to store executable program code;

[0031] The processor runs a program corresponding to the executable program code stored in the memory to perform the methods described in one or more embodiments.

[0032] This invention provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described intelligent charging control method for pure electric vehicles.

[0033] In view of the above, in one or more embodiments of this specification, battery charging parameters are obtained, including battery parameters, historical charging parameters, and charging environment parameters. These parameters are then input into a neural network model for training. The model training algorithm includes a temperature rise risk index, voltage surge trend, and constant current / constant voltage switching, and outputs basic risk decisions. These basic risk decisions include a temperature rise intervention index, a voltage trend index, and a constant current / constant voltage inflection point. After the battery has undergone a charging cycle, an empirical data set within that cycle is obtained. This empirical data set, combined with the evaluation algorithm, determines the decision value of the basic risk decisions, and updates the neural network model based on this decision value. This not only addresses the fundamental deficiencies in battery charging protection, high-temperature adaptation, and current balancing, but also dynamically and adaptively adjusts charging decisions during the charging process, thereby better meeting user charging requirements and improving the user experience. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a flowchart of an embodiment of an intelligent charging control method for pure electric vehicles provided in this specification.

[0036] Figure 2 This is a schematic diagram of the structure of an intelligent charging control system for a pure electric vehicle provided in one embodiment of this specification.

[0037] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this specification. Detailed Implementation

[0038] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the various examples. For example, the described methods may be performed in a different order than described, and steps may be added, omitted, or combined. Furthermore, features described in some examples may be combined in other examples.

[0039] As used herein, the term "comprising" and its variations are open terms meaning "including but not limited to". The term "based on" means "at least partially based on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other definitions, whether explicit or implicit, may be included below. Unless explicitly indicated by the context, the definition of a term shall remain consistent throughout the specification.

[0040] like Figure 1 As shown, this embodiment of the invention provides an intelligent charging control method for pure electric vehicles, including:

[0041] Step S102: Obtain battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters.

[0042] Specifically, during the charging process of pure electric vehicles, dynamic battery charging parameters are collected, including battery intrinsic parameters, historical charging parameters, and charging environment parameters. Battery intrinsic parameters can include basic battery parameters such as individual cell voltage and temperature; and further derived parameters, including voltage dispersion (standard deviation of all individual cell voltages), temperature change rate (the amount of temperature change per unit time), and internal resistance change rate (estimated through electrochemical impedance spectroscopy or voltage / current transient response). On the other hand, charging parameters include historical charging parameters, namely, the average final SOC (state of charge) of the battery over historical charging processes (e.g., over 10 historical charges) (used to determine user charging habits) and performance degradation data (the trend of capacity and internal resistance changes during historical charging). Charging environment parameters include ambient temperature and humidity, which affect heat dissipation efficiency.

[0043] In addition, for the above data, the parameters are preprocessed before subsequent processing steps. Noise and fault data can be removed by using sliding window averaging and anomaly detection algorithms. The corresponding data types are also standardized. Real-time data and historical charging parameters are aligned to form a time series, ensuring data consistency and input stability.

[0044] Step S104: Input the battery charging parameters into the neural network model for model training. The algorithm module for model training includes temperature rise risk index, voltage change trend and constant current and constant voltage switching, and outputs basic risk decisions. The basic risk decisions include: temperature rise intervention index, voltage trend index and constant current and constant voltage inflection point.

[0045] Specifically, the battery-related parameters obtained in the above steps are used for model training. The model training step includes three parts, and the algorithm modules for each part are as follows:

[0046] Module 1: The calculation of the temperature rise risk index will shift from "remedial measures after the temperature reaches a threshold" to "predicting temperature trends and intervening in advance." The core is to establish a temperature prediction model.

[0047] Module one is the temperature prediction model. Input data includes the battery cell temperature, temperature change rate, charging current, elevator voltage, internal resistance, and ambient temperature from the aforementioned data. Combining this input data, a neural network (optionally an LSTM for time series processing) processes the input data as a time series window (e.g., data every 30 seconds). After model training, the trend of cell temperature change is determined, thus outputting a predicted cell temperature for the next 5-10 seconds.

[0048] Additionally, a temperature risk index, namely the first risk index (temperature rise intervention index), can be calculated based on the predicted cell temperature. Specifically, this involves calculating the distance between the predicted temperature and a safe threshold (e.g., 55°C). The difference between these two values ​​is not a simple difference, but rather a normalized exponential function. This outputs the temperature-related first risk index.

[0049] Module 2: Analysis of voltage mutation trends to identify subtle faults and inconsistencies within the battery, enabling early fault diagnosis and proactive balancing.

[0050] The input data for Module 2 includes the voltage array of all individual cells, voltage dispersion (standard deviation), voltage change rate of each individual cell, and internal resistance change rate. A neural network model for voltage prediction is trained using this input data. A second risk index is then calculated based on the trained model, and this index is used to predict voltage trends, including voltage dispersion trends and voltage sags.

[0051] Specifically, when the voltage dispersion exceeds a preset threshold (50mV) and continues to rise, it is determined that the current battery is experiencing a deterioration in consistency. Furthermore, the short-term average voltage of each individual cell is calculated, and the deviation between the current value and the average value is compared in real time. If the voltage of any individual cell drops by more than 5% within 1 second, and external factors such as sudden current changes can be ruled out (through cross-validation of current data), then an internal micro-short circuit is suspected to have occurred in that individual cell.

[0052] Module 3: Constant Current to Constant Voltage Switching Point Optimization. The switching point during the charging stage is changed from a fixed SOC value to a dynamic value based on the real-time health status and operating conditions of the battery, achieving the optimal balance between charging efficiency and lifespan.

[0053] Module 3's input data includes battery state, cell temperature, voltage change with charge (voltage inflection point), and charging current. Towards the end of constant current (CC) charging, the voltage rise rate accelerates. Traditional BMSs wait until the voltage reaches its absolute upper limit (e.g., 4.2V) before switching to constant voltage (CV) charging, at which point the battery is already under high stress. The above input data can be fed into a neural network model for learning. By learning from a large amount of battery aging data, the arrival of the voltage inflection point can be determined earlier. Analyzing the change in voltage with charge, when this value begins to increase non-linearly, it means that continuing constant current charging will significantly accelerate aging.

[0054] The model is corrected based on both battery state and individual cell temperature. When the battery state is low (aged battery): its withstandable polarization voltage is lower, so the model will switch the switching point earlier (e.g., from 80% SOC to 75%). Or when the temperature is low: the battery internal resistance is high, and the voltage is artificially high, so the model will also switch the switching point earlier to avoid overcharging risk. Or when the temperature is suitable and the battery state is high: the model can appropriately delay the switching point to maximize constant current fast charging within a safe range and save time. This determines the inflection point from CC to CV under different temperatures and battery states.

[0055] By integrating the trigger condition values ​​in the basic risk decision-making process, four levels of risk protection can be determined during battery charging, and corresponding response measures can be configured. Specific protection level classifications may include:

[0056] For Level 1 warnings, when the differential pressure of a single unit is greater than 50mV or the temperature is greater than 45℃, the corresponding response measures can be to activate the liquid cooling system and push the warning through the central control screen.

[0057] Level 2 current limiting: if SOH is less than 80% or current imbalance (standard deviation) is greater than 15%, the response measure is to switch trickle charging and activate the active balancing circuit.

[0058] Level 3 interruption: When a short circuit / reverse connection / temperature > 60℃ occurs, the corresponding measure is to disconnect the main circuit.

[0059] In a Level 4 emergency, if a BMS dead loop or communication failure occurs, the BMS can be restarted via a relay controlled by the VCU.

[0060] Step S106: After the battery has gone through a charging cycle, obtain the experience data set within the charging cycle, combine the experience data set with the evaluation algorithm, determine the decision value of the basic risk decision, and update the neural network model based on the decision value.

[0061] Specifically, after each charging cycle, an experience data set for that charging cycle is acquired. This experience data set includes the initial state, decision action, action reward, and new state.

[0062] The initial state is the state before the start of the charging cycle. The data in the state includes the battery (average health) status, initial temperature, ambient temperature, (user-defined) charging mode, and the target SOC for this charge.

[0063] The decision-making process involves inputting the initial state data into the neural network models in modules one, two, and three of the above steps, and summarizing the decision-making actions based on the output of the neural network models throughout the entire charging cycle. These decision-making actions may include: the ratio of the average charging current to the maximum current threshold (module two), the average trigger level of the temperature rise risk index (module one), and the offset of the CC-CV switching point relative to the standard value (module three).

[0064] Action rewards are scalar scores for the evaluation decisions made during the current charging cycle in the initial state, serving as the driving force for model training. These include, but are not limited to: charging speed rewards (higher rewards for shorter durations), battery health rewards (higher rewards for less capacity decay), safety risk penalties (the degree to which the maximum temperature exceeds the safety threshold), and user satisfaction rewards (the ratio of actual charging amount to the user's target charging amount).

[0065] The new state is the state after the charging cycle ends.

[0066] By combining empirical data sets, training algorithms, such as the proximal policy optimization algorithm, can be used for training. The training process includes collecting several empirical data sets. Then, the neural networks in modules one, two, and three, along with the corresponding value network, are used to evaluate the decision value. The evaluated value of the decision action in each empirical data set is calculated, and the formula for the evaluated value can be:

[0067] A t =R+γ*V(S')-V(S)

[0068] Among them, A t Let R be the evaluation value of the decision action, γ be the action reward, γ be the discount factor, V be the value network prediction, S' be the new state, and S be the initial state.

[0069] Then, the network parameters θ are updated through optimization steps to maximize the objective function, where the objective function includes:

[0070] L(θ)=E[min(r(θ)*A t ,clip(r(θ),1-ε,1+ε)*A t )]

[0071] Where r(θ) = (new decision probability / old decision probability), it represents the change in preference for the same action relative to the old decision. The Clip function restricts r(θ) to the range [1-ε, 1+ε], ensuring that each decision update does not deviate too far from the old decision, thus guaranteeing the stability of the training process. The Min operation selects a more conservative option between the original update and the truncated update, further ensuring stability. E is the batch sampling average.

[0072] Finally, the weights θ of the neural network model are adjusted using a gradient ascent algorithm (such as Adam) to maximize L(θ). At the same time, the value network is also updated to better predict long-term returns.

[0073] This invention provides an intelligent charging control method for pure electric vehicles. The method acquires battery charging parameters, including battery intrinsic parameters, historical charging parameters, and charging environment parameters. These parameters are then input into a neural network model for training. The model training algorithm includes a temperature rise risk index, voltage surge trend, and constant current / constant voltage switching. The method outputs basic risk decisions, including a temperature rise intervention index, a voltage trend index, and a constant current / constant voltage inflection point. After the battery has undergone a charging cycle, the method acquires a set of empirical data within that cycle. Combining this empirical data with the evaluation algorithm, the method determines the decision value of the basic risk decisions and updates the neural network model based on these decision values. This approach not only fundamentally addresses deficiencies in battery charging protection, high-temperature adaptation, and current balancing, but also dynamically and adaptively adjusts charging decisions during the charging process, thereby better meeting user charging requirements and improving the user experience.

[0074] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of an intelligent charging control system for pure electric vehicles provided in an embodiment of this application. Figure 2 As shown, the system includes:

[0075] The parameter module S202 is used to acquire battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters.

[0076] The training module S204 is used to input the battery charging parameters into the neural network model for model training. The algorithm module for model training includes temperature rise risk index, voltage change trend and constant current and constant voltage switching, and outputs basic risk decisions. The basic risk decisions include: temperature rise intervention index, voltage trend index and constant current and constant voltage inflection point.

[0077] The update module S206 is used to detect the battery after it has gone through a charging cycle, obtain the experience data set within the charging cycle, combine the experience data set with the evaluation algorithm, determine the decision value of the basic risk decision, and update the neural network model based on the decision value.

[0078] In another embodiment, a smart charging control system for pure electric vehicles further includes:

[0079] The protection module is used to determine the four-level risk protection level during the battery charging process by comprehensively considering the trigger condition values ​​in the basic risk decision, and to configure the corresponding response measures.

[0080] Those skilled in the art will clearly understand that the technical solutions of the embodiments of this application can be implemented by means of software and / or hardware. In this specification, "unit" and "module" refer to software and / or hardware that can independently complete or cooperate with other components to complete a specific function, wherein the hardware may be, for example, a field-programmable gate array (FPGA), an integrated circuit (IC), etc.

[0081] Each processing unit and / or module in the embodiments of this application can be implemented by an analog circuit that implements the functions described in the embodiments of this application, or by software that executes the functions described in the embodiments of this application.

[0082] See Figure 3 It shows a schematic diagram of the structure of an electronic device according to an embodiment of this application, which can be used to implement... Figure 1 The method in the illustrated embodiment. (As shown) Figure 3 As shown, the electronic device 300 may include: at least one processor 301, at least one network interface 304, user interface 303, memory 305, and at least one communication bus 302.

[0083] The communication bus 302 is used to enable communication between these components.

[0084] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0085] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0086] The processor 301 may include one or more processing cores. The processor 301 connects to various parts within the electronic device 300 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling data stored in the memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display on the screen; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0087] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory 305 may include a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. Figure 3 As shown, the memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and program instructions.

[0088] exist Figure 3In the illustrated electronic device 300, the user interface 303 is mainly used to provide an input interface for the user and acquire user input data; while the processor 301 can be used to call the image-based interactive application stored in the memory 305 and specifically perform the following operations: acquire battery charging parameters, including battery parameters, historical charging parameters, and charging environment parameters; input the battery charging parameters into a neural network model for model training, the algorithm module for model training includes temperature rise risk index, voltage change trend, and constant current / constant voltage switching, and output basic risk decisions, including temperature rise intervention index, voltage trend index, and constant current / constant voltage inflection point; detect after the battery has undergone a charging cycle, acquire the experience data set within the charging cycle, combine the experience data set and the evaluation algorithm to determine the decision value of the basic risk decisions, and update the neural network model based on the decision value.

[0089] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0090] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0091] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0092] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0093] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0095] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0096] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0097] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

Claims

1. A method for intelligent charging control of pure electric vehicles, the method comprising: Obtain battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters. The battery charging parameters are input into a neural network model for model training. The algorithm module for model training includes a temperature rise risk index, a voltage change trend, and a constant current / constant voltage switching mechanism. The model outputs basic risk decisions, which include a temperature rise intervention index, a voltage trend index, and a constant current / constant voltage inflection point. After the battery has undergone a charging cycle, an empirical data set within the charging cycle is acquired. Combining the empirical data set with the evaluation algorithm, the decision value of the basic risk decision is determined, and the neural network model is updated based on the decision value.

2. The method according to claim 1, characterized in that, The empirical data set includes: Initial state, decision action, action reward, and new state.

3. The method according to claim 2, characterized in that, The formula for calculating the decision value of the basic risk decision, by combining the empirical data set and the evaluation algorithm, includes: A t =R+γ*V(S')-V(S) Among them, A t Let R be the evaluation value of the decision action, γ be the action reward, γ be the discount factor, V be the value prediction of the evaluation algorithm, S' be the new state, and S be the initial state.

4. The method according to claim 3, characterized in that, The calculation formula for updating the neural network model using the decision value includes: L(θ)=E[min(r(θ)*A t ,clip(r(θ),1-ε,1+ε)*A t )] Where r(θ) is the probability of the new decision / the probability of the basic risk decision, the Clip function restricts r(θ) to the range of [1-ε, 1+ε], and E is the batch sampling average.

5. The method according to claim 1, characterized in that, The action reward includes: Rewards include charging speed, battery health, safety risk penalties, and user satisfaction.

6. The method according to claim 1, characterized in that, The method further includes: Based on the trigger condition values ​​in the basic risk decision-making, four risk protection levels are determined during the battery charging process, and corresponding response measures are configured.

7. A smart charging control system for pure electric vehicles, characterized in that, The system includes; The parameter module is used to acquire battery charging parameters, which include battery parameters, historical charging parameters, and charging environment parameters. The training module is used to input the battery charging parameters into the neural network model for model training. The algorithm module for model training includes temperature rise risk index, voltage change trend and constant current and constant voltage switching, and outputs basic risk decisions. The basic risk decisions include: temperature rise intervention index, voltage trend index and constant current and constant voltage inflection point. The update module is used to detect the battery after it has gone through a charging cycle, obtain the experience data set within the charging cycle, combine the experience data set with the evaluation algorithm, determine the decision value of the basic risk decision, and update the neural network model based on the decision value.

8. The system according to claim 7, characterized in that, The system also includes: The protection module is used to determine the four-level risk protection level during the battery charging process by comprehensively considering the trigger condition values ​​in the basic risk decision, and to configure the corresponding response measures.

9. An electronic device, comprising a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the method as described in any one of claims 1-6.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of claims 1-6.

Images