Intelligent management method and system for charging and discharging of charging boxes

Abstract

This application provides a charging box intelligent charging and discharging management method and system, belonging to the field of charging management technology. The method includes: constructing a battery state vector and a user behavior-grid state vector based on the current battery state, user historical behavior characteristics, and grid information. The battery state vector is input into a preset battery prediction model to obtain the predicted remaining battery lifespan and predicted battery health level. The user behavior-grid state vector, the predicted remaining battery lifespan, and the predicted battery health level are input into a preset charging prediction model to obtain the predicted charging demand and predicted charging period. The predicted remaining battery lifespan, predicted battery health level, predicted charging demand, and predicted charging period are input into a preset power prediction model to obtain the predicted charging power and predicted charging duration. The vehicle is charged according to the predicted charging power and predicted charging duration. This can extend battery lifespan and improve the grid's load capacity.

Description

Technical Field

[0001] This application relates to the field of charging management technology, and in particular to a method and system for intelligent management of charging and discharging of a charging box. Background Technology

[0002] With the widespread adoption of electric vehicles (EVs) and renewable energy, charging demand is increasing daily, and the management of charging infrastructure faces unprecedented challenges. Batteries are affected by different charging methods (such as fast charging and slow charging) and charging parameters (such as current and voltage) during charging, which directly impact the long-term health and lifespan of the battery. Currently, charging infrastructure typically uses fixed charging modes, unable to adjust in real-time according to the specific state of the battery. This can lead to accelerated aging and even battery damage during charging, especially during fast charging, due to overcharging or improper charging. Secondly, with the increasing number of charging stations, especially during peak demand periods, unreasonable allocation of grid resources can lead to grid overload. Therefore, reducing battery damage during charging and alleviating grid pressure have become urgent technical problems to be solved. Summary of the Invention

[0003] The main objective of this application is to propose a smart charging and discharging management method and system for charging boxes, which aims to reduce battery damage during charging and alleviate grid pressure.

[0004] To achieve the above objectives, a first aspect of this application proposes a smart charging and discharging management method for a charging box, the method comprising:

[0005] Obtain the vehicle's current battery status, user historical behavior characteristics, and power grid information;

[0006] A state harmonic term is constructed based on the current battery state, the user's historical behavior characteristics, and the power grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-power grid state vector;

[0007] The battery state vector is input into a preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level.

[0008] The user behavior-grid state vector, the predicted remaining battery life, and the predicted battery health level are input into a preset charging prediction model to obtain the predicted charging demand and the predicted charging period.

[0009] The predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period are input into a preset power prediction model to obtain the predicted charging power and predicted charging time.

[0010] The vehicle is charged according to the predicted charging power and the predicted charging duration.

[0011] In some embodiments, after inputting the predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period into a preset power prediction model to obtain the predicted charging power and predicted charging duration, the method further includes:

[0012] The predicted charging power is corrected based on the preset maximum sustainable safe charging power, the preset maximum battery life, the predicted remaining battery life, and the predicted charging power to obtain the corrected predicted charging power.

[0013] The predicted charging duration is corrected based on the predicted charging period and the predicted charging duration to obtain the corrected predicted charging duration.

[0014] In some embodiments, charging the vehicle based on the predicted charging power and predicted charging duration includes:

[0015] The predicted charging power is processed by current conversion to obtain the target charging current;

[0016] The vehicle is charged according to the target charging current and the predicted charging time.

[0017] In some embodiments, the current state of the battery includes the battery temperature, and the method further includes, during the charging of the vehicle based on the predicted charging power and predicted charging duration:

[0018] If the battery temperature exceeds the preset maximum safe temperature, and / or the target charging current exceeds the preset maximum allowable charging current, charging will stop.

[0019] In some embodiments, the battery prediction model includes a thermal-electrical joint anomaly sensitivity regularization term and a target weight dynamic adjustment mechanism;

[0020] The thermo-electric combined anomaly sensitivity regularization term is:

[0021]

[0022] in, This represents the regularization term for the combined thermal and electrical anomaly sensitivity. RUL indicates the actual remaining battery life. t This indicates the predicted remaining battery life. Indicates the actual battery health level, SoH t This indicates the predicted battery health level. This represents the rate of change of battery temperature in the i-th training sample. α1 represents the rate of change of the battery's internal resistance, β1 represents the weight of the error term in predicting the battery's remaining lifespan, and γ1 represents the weight of the error term in predicting the battery's health level.

[0023] The target weight dynamic adjustment mechanism is as follows:

[0024]

[0025] Among them, SoH t This represents the predicted battery health level, where σ represents the Sigmoid function. Indicates the actual battery health level, κ represents the voltage regulation coefficient, V nom Indicates the nominal voltage, V t ∈ represents the battery voltage, and ∈ represents a small constant.

[0026] In some embodiments, the loss function of the charging prediction model is:

[0027]

[0028] in, Let Q represent the loss function of the charging prediction model. predict Q represents the predicted charging demand. real Indicates the actual charging demand, t window The predicted charging period is represented by t. real The actual charging period is represented by λ, which represents the adjustment coefficient, and SoH. t This indicates the predicted battery health level.

[0029] In some embodiments, the power prediction model includes an optimizer, which is:

[0030]

[0031] Among them, Q predict P represents the predicted charging demand. t The predicted charging power is represented by T, and the predicted charging time is represented by SoH. t P represents the predicted battery health level. grid This indicates the current grid load threshold.

[0032] To achieve the above objectives, a second aspect of this application proposes a charging and discharging intelligent management system for a charging box, the system comprising:

[0033] The acquisition module is used to acquire the vehicle's current battery status, user historical behavior characteristics, and power grid information;

[0034] A construction module is used to construct a state harmonic term based on the current state of the battery, the user's historical behavior characteristics, and the power grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-power grid state vector;

[0035] The first prediction module is used to input the battery state vector into a preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level.

[0036] The second prediction module is used to input the user behavior-grid state vector, the predicted remaining battery life and the predicted battery health level into a preset charging prediction model to obtain the predicted charging demand and the predicted charging period.

[0037] The third prediction module is used to input the predicted remaining battery life, predicted battery health level, predicted charging demand and predicted charging period into a preset power prediction model to obtain the predicted charging power and predicted charging time.

[0038] A charging module is used to charge the vehicle according to the predicted charging power and the predicted charging time.

[0039] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0040] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0041] The intelligent charging and discharging management method and system proposed in this application acquires the current battery state, user historical behavior characteristics, and power grid information. A state harmonic term is constructed based on the current battery state, user historical behavior characteristics, and power grid information; this term includes a battery state vector and a user behavior-power grid state vector. The battery state vector is input into a preset battery prediction model to obtain the predicted remaining battery life and predicted battery health level. The user behavior-power grid state vector, the predicted remaining battery life, and the predicted battery health level are input into a preset charging prediction model to obtain the predicted charging demand and predicted charging period. The predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period are input into a preset power prediction model to obtain the predicted charging power and predicted charging duration. The vehicle is charged according to the predicted charging power and predicted charging duration. By incorporating factors such as the current battery state, user historical behavior characteristics, and power grid information into the charging decision-making process in real time, the charging strategy is optimized, thereby extending battery life and improving the power grid's load capacity. Attached Figure Description

[0042] Figure 1 This is a flowchart of the intelligent charging and discharging management method for a charging box provided in an embodiment of this application;

[0043] Figure 2 yes Figure 1 The flowchart of step S106 in the process;

[0044] Figure 3 This is a schematic diagram of the charging and discharging intelligent management system for the charging box provided in an embodiment of this application;

[0045] Figure 4 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] It should be noted that although functional modules are divided in the system diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0048] 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 application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0049] With the widespread adoption of electric vehicles (EVs) and renewable energy, charging demand is increasing daily, and the management of charging infrastructure faces unprecedented challenges. Batteries are affected by different charging methods (such as fast charging and slow charging) and charging parameters (such as current and voltage) during charging, and these factors directly affect the long-term health and lifespan of the battery. In related technologies, charging facilities typically employ fixed charging modes, unable to adjust in real time according to the specific state of the battery. This means that during charging, especially fast charging, batteries may age faster due to overcharging or improper charging, or even be damaged. Secondly, with the increase in the number of charging stations, especially during peak demand periods, unreasonable allocation of grid resources can lead to grid overload.

[0050] Based on this, embodiments of this application provide a charging box intelligent charging and discharging management method and system. By incorporating factors such as the current battery state, historical user behavior characteristics (user charging needs), and grid information into the charging decision-making process in real time, the charging strategy is optimized. Specifically, the system can dynamically adjust based on real-time battery monitoring data (such as battery voltage, temperature, internal resistance, etc.) to select the most suitable charging method for different batteries, avoiding damage to battery health caused by overcharging and fast charging. This extends battery life and reduces maintenance costs. Secondly, addressing the problems of excessive congestion of charging resources and insufficient grid load management, the system enables charging piles to communicate with the grid system in real time, dynamically adjusting charging power and charging time periods based on grid information and user charging needs. This allows charging piles to avoid peak grid load periods, reducing grid pressure, and accelerates charging when grid load is light, improving charging efficiency.

[0051] The intelligent charging and discharging management method and system for charging boxes provided in this application are specifically described through the following embodiments. First, the intelligent charging and discharging management method for charging boxes in this application embodiment is described.

[0052] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0053] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0054] The intelligent charging and discharging management method for charging boxes provided in this application can be applied to terminals, servers, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the intelligent charging and discharging management method for charging boxes, but is not limited to the above forms.

[0055] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0056] Please refer to Figure 1 , Figure 1 This is a flowchart of the intelligent charging and discharging management method for a charging box provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0057] Step S101: Obtain the current battery status of the vehicle, user historical behavior characteristics, and power grid information;

[0058] Step S102: Construct a state harmonic term based on the current battery state, user historical behavior characteristics, and grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-grid state vector;

[0059] Step S103: Input the battery state vector into the preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level.

[0060] Step S104: Input the user behavior-grid state vector, the predicted remaining battery life and the predicted battery health level into the preset charging prediction model to obtain the predicted charging demand and the predicted charging period.

[0061] Step S105: Input the predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period into the preset power prediction model to obtain the predicted charging power and predicted charging time.

[0062] Step S106: Charge the vehicle according to the predicted charging power and the predicted charging time.

[0063] In step S101 of some embodiments, the current state of the battery includes the current battery voltage V. t Battery temperature T t and battery internal resistance R t The data is read by the BMS module via the CAN bus interface. User historical behavior characteristics include the user's historical average charging amount Q. avg User historical charging volume variance Q var Average charging time t per user avg This value is calculated by the charging box's local control unit based on the user's last N charging actions. Grid information includes the current grid load L. t And the current electricity price W t The data is obtained in real time by the local charging box control unit via API. There are a total of 8 original input variables, all of which are statistics within the current or short-term window.

[0064] In step S102 of some embodiments, all original input variables are constructed into a current state vector X0 = [V t ,T t ,R t |Q avg Q var ,t avg |L t W t Subsequently, in order to enhance the model's ability to perceive system state imbalances (such as "strong user demand + poor battery condition"), a state reconciliation term is introduced to integrate the relationship between behavioral demand and battery supply in advance, as shown in the following formula (1):

[0065]

[0066] Where X′0 represents the state harmonic term, and X0 represents the current state vector. λ represents the harmonic coefficient, with an empirical value of 0.1 to 0.3, controlling the weight of this term in the input vector. Q avg This represents the user's historical average charging amount, in kWh. Q max This indicates the maximum single charging capacity set in the system, in kWh, such as 20kWh. V t This indicates the current battery voltage, in volts (V), and is provided by the BMS module. (V) nom This indicates the nominal voltage, in volts (V), such as 48V. ∈ indicates a minimum value to prevent the denominator from being zero; a value of 10 is recommended. -3 Extract the user behavior-grid state vector X from the state harmonic term X′0. u ={Q avg Q var ,t avg ,L t W t} and battery state vector X b ={V t ,T t ,R t}

[0067] It's important to note that the state harmonization term reflects the "tension" between user charging demand and battery state, enabling the system to identify unfavorable combinations of high load, high demand, and low battery state at the input stage, thus guiding the scheduling model towards a conservative or delayed strategy in advance. This step is used to construct a structurally unified and temporally consistent input vector X0, serving as the core input for subsequent battery health assessment and user behavior modeling. It integrates three types of information: current battery state, historical user behavior characteristics, and grid information, and introduces a normalized state harmonization term to reflect the dynamic tension between states across different dimensions. All data comes from directly accessible sensors or interfaces in the charging box deployment environment, and the vector construction process can be completed in real time in the local controller or edge computing module.

[0068] In step S103 of some embodiments, the input layer of the battery prediction model is a 3-dimensional vector X. b The hidden layers consist of two fully connected layers (64-dimensional and 32-dimensional), with ReLU activation function. The output layer has a dimension of 2, and outputs the predicted remaining battery life RUL. t And predicting battery health level SoH t .

[0069] Specifically, to improve the model's ability to identify abnormal states and to introduce physical priors, a thermo-electric interaction anomaly sensitive regularization term is introduced into the loss function of the battery prediction model, as shown in the following formula (2):

[0070]

[0071] in, This indicates a regularization term for the combined thermal and electrical anomaly sensitivity. Indicates the actual remaining battery life (from tag data), RUL t This indicates the predicted remaining battery life. This indicates the actual battery health level (from tag data), SoH t This indicates the predicted battery health level. This represents the rate of change of battery temperature in the i-th training sample. The two dimensions together represent the risk of rapid battery aging or abnormal charging. α1 represents the weight of the error term in predicting the remaining battery life, β1 represents the weight of the error term in predicting the battery health level, and γ1 represents the weight of the thermo-electric coupling regularization term. The first term in formula (2) is the RUL prediction error, the second term is the SoH prediction error, and the third term introduces a "coupling term between the rate of temperature rise and the rate of internal resistance rise" to penalize the model's insensitivity to drastic changes in battery state.

[0072] It should be noted that the physical meaning of this regularization term is that, in the charging box scenario, user behavior is uncontrollable and may frequently perform high-power charging; at this time, the battery temperature rises and the internal resistance fluctuates drastically, but if the model lacks such signal penalties, it is easy to misjudge it as healthy; the regularization term forces the model to pay attention to this high-risk range, making its judgment closer to the physical degradation process in the actual scenario; it can be understood as a temperature and resistance joint anomaly weighting mechanism, which is an important structural improvement that distinguishes it from conventional time series prediction methods.

[0073] Furthermore, to improve the prediction stability of the battery prediction model under different battery health states, a target weight dynamic adjustment mechanism is designed, as shown in the following formula (3):

[0074]

[0075] Among them, SoH t This indicates the predicted battery health level, and σ represents the Sigmoid function, ensuring that the output range is [0,1]. This indicates the actual battery health level (from label data), and κ represents the voltage regulation coefficient, with an empirical value of 0.5–1.0. V nom This indicates the nominal voltage, such as 48V. (V) t This represents the battery voltage, and ∈ represents a small constant to avoid division by zero.

[0076] The goal of the target weight dynamic adjustment mechanism is to dynamically adjust the health value predicted by the model if the current battery voltage is detected to be much lower than the nominal level during the charging process (such as after deep discharge or before charging is started), so as to avoid overestimating SoH due to data disturbance when the voltage is low. This design is particularly suitable for charging box scenarios with multiple access points and unstable state during the start-up phase, thus enhancing the robustness of the model.

[0077] The training phase of the battery prediction model is completed on the server side, and the model parameters can be deployed to the terminal via OTA. The inference phase is executed entirely locally in the charging box, with a single inference latency of no more than 50ms, meeting the system's real-time requirements.

[0078] After the model training is complete, the battery state vector X is... b Inputting the data into the battery prediction model yields the predicted remaining battery life (RUL). t And predicting battery health level SoH t Predicting Remaining Battery Life (RUL) t The unit is "charge-discharge cycles," which represents the battery's expected remaining lifespan and can be used as an assessment indicator of battery replacement plans and the acceptability of fast charging. Battery health rating (SoH) t , is a continuous number between [0,1], representing the current health level of the battery. The closer it is to 1, the healthier it is. It is used directly for user behavior prediction and charging scheduling strategies.

[0079] In step S104 of some embodiments, the input layer dimension of the charging prediction model is 7, corresponding to 7 input variables, namely the user behavior-grid state vector X. u ={Q avg Q var ,t avg ,L t W t Predicting the remaining battery life (RUL) t Battery health rating SoH t The first hidden layer is a fully connected layer with 64 nodes and the ReLU activation function. The second hidden layer is also a fully connected layer with 32 nodes and the ReLU activation function. The output layer has two outputs: the predicted charging demand Q. predict and predicted charging period t window .

[0080] The charging prediction model is trained under supervision using a historical dataset. The sample format is: Input features {RUL} t SoH t Q avg Q var ,t avg ,L t W tThe tag represents the user's actual charging needs (Q). real and actual charging period t real The loss function is shown in the following formula (4):

[0081]

[0082] in, Let Q represent the loss function of the charging prediction model. predict Q represents the predicted charging demand. real Indicates the actual charging demand, t window t represents the predicted charging period. real This represents the actual charging period, λ represents the adjustment coefficient, with an empirical value of 0.1 to 0.3, and SoH. t This indicates a predicted battery health level. The model was trained using the Adam optimizer with a learning rate of 1e. -3 The number of training samples is no less than 1000. After deployment, the inference time is no more than 50ms, supporting real-time prediction.

[0083] It should be noted that the first two terms in formula (4) are standard regression losses used to fit the target output, and the third term is a regularization term, which reflects that "the worse the battery health, the more conservative the user's charging prediction needs to be." This regularization term effectively prevents the system from incorrectly predicting that the user will still engage in high-power charging behavior when the battery health is low, ensuring the rationality and feasibility of the prediction.

[0084] After the power prediction model is trained, the user behavior-grid state vector, the predicted remaining battery life, and the predicted battery health level are input into the preset charging prediction model to obtain the predicted charging demand Q. predict and predicted charging period t window Forecast charging demand Q predict Used to predict the amount of charging requests users may initiate in the next cycle, in kWh. Predicted charging period t. window The length of the user's expected charging period, in minutes, is used for prediction.

[0085] In step S105 of some embodiments, the aim is to generate a smart charging scheduling scheme that simultaneously satisfies the triple constraints of user demand, battery state, and grid conditions. This is based on the user charging behavior (Q) predicted in the previous step. predict ,t window ) and battery current health status (RUL) t SoH t ), calculate the predicted charging power P tThe predicted charging time T is considered. Due to uncontrollable user behavior, complex battery states, and scheduling constraints imposed by hardware capabilities and grid fluctuations, an integrated optimization structure of health awareness, user-driven, and grid-side coordination is proposed. This structure incorporates physical boundary constraints, a dynamic state weighting mechanism, and an asymmetric penalty term based on differentiated objectives at different health stages to ensure the executability of the policy output and system stability.

[0086] The entire scheduling process is implemented using a two-stage optimization strategy: the first stage is the construction and optimization of the objective function, and the second stage is the implementable correction mechanism under healthy conditions.

[0087] The scheduling optimization objective of the first-stage optimizer is shown in the following formula (5):

[0088]

[0089] Among them, Q predict P represents the predicted charging demand. t The predicted charging power is represented by T, and the predicted charging time is represented by SoH. t This indicates the predicted battery health level. (P) grid This represents the current grid load threshold, derived from the current grid load L. t Normalized values. α represents the weight of user preference, β represents the weight of health protection, and γ represents the weight of grid friendliness. The suggested initial values ​​are 3:5:2. In formula (5), the first term represents the user satisfaction loss, measuring the degree of deviation between the final strategy and user expectations; the second term introduces the battery health perception factor (1-SoH). t The power penalty is dynamically adjusted; the third term expresses the nonlinear growth pressure of the grid load in exponential form. The model calls the optimizer (such as a local linear programmable or gradient descent) once in each scheduling cycle to calculate P. t The optimal combination of T. This objective function is not a traditional linearly weighted objective, but rather based on SoH. t The state value dynamically adjusts the weights, and applies them to P. t A secondary penalty is introduced to enable the system to automatically limit power during the health deterioration phase. The third term of formula (5) is the grid stress index control term considered for the first time. Its exponential structure is used to simulate the nonlinear penalty of fast charging power during peak grid periods, reflecting the "system-level energy consumption collaborative response mechanism".

[0090] The second phase introduces a RUL-based... t The health stage correction mechanism is used to determine whether the optimal solution is practically feasible. That is, the predicted charging power is corrected based on the preset maximum sustainable safe charging power, the preset maximum battery life, the predicted remaining battery life, and the predicted charging power, to obtain the corrected predicted charging power, as shown in the following formula (6):

[0091]

[0092] in, This represents the corrected predicted charging power, in kW (P). t P represents the predicted charging power. safe RUL indicates the system's calibrated maximum sustainable safe charging power. t RUL indicates the predicted remaining battery life. max This indicates the battery's maximum lifespan.

[0093] The predicted charging time is corrected based on the predicted charging period and the predicted charging duration to obtain the corrected predicted charging time, as shown in the following formula (7):

[0094] T * =min(T,t) window (7)

[0095] Among them, T * This indicates the corrected predicted charging time, in minutes, where T represents the predicted charging time. window This indicates the predicted charging period.

[0096] This correction mechanism introduces the concept of dynamic threshold control over the battery's lifecycle, meaning that as the battery ages, the system automatically reduces the maximum charging power. This mechanism addresses the need for different scheduling strategies at different lifecycle stages in real-world scenarios, strengthening the system's long-term adaptability and device protection capabilities. Furthermore, to improve the stability of the model output, a short-time moving average method is used to analyze historical P... t The output results are averaged using a window to prevent frequent fluctuations in the control system.

[0097] Please see Figure 2 In some embodiments, step S106 may include, but is not limited to, steps S201 to S202:

[0098] Step S201: Perform current conversion processing on the predicted charging power to obtain the target charging current;

[0099] Step S202: Charge the vehicle according to the target charging current and the predicted charging time.

[0100] In step S201 of some embodiments, when the predicted charging power is corrected, the corrected predicted charging power is subjected to current conversion processing to obtain the target charging current. Otherwise, the predicted charging power is directly subjected to current conversion processing to obtain the target charging current.

[0101] Specifically, by It is converted into the target charging current and dynamically adjusted according to the real-time voltage. The control chip periodically reads V... real (t)(provided by BMS, sampled once per second), when the predicted charging power is overcorrected, the target charging current is calculated by the following formula (8):

[0102]

[0103] Among them, I target (t) represents the target charging current. V represents the corrected predicted charging power. real (t) represents the current voltage measured in real time, and ∈ indicates that the minimum value should be avoided by dividing by zero, set to 10. -3 .

[0104] In step S202 of some embodiments, the target charging current is controlled by the MCU to drive the MOSFET or DC-DC converter module to perform the charging behavior. The system's built-in clock module initializes the timer and sets the predicted charging duration to T. * The device will continue charging for the countdown period, and will automatically power off when the time is up. If an abnormal state is detected in advance, the charging will be terminated early.

[0105] In some embodiments, during the charging of the vehicle based on the predicted charging power and predicted charging duration, to prevent operational risks caused by prediction errors or sudden state changes, charging is stopped when the battery temperature exceeds a preset maximum safe temperature and / or the target charging current exceeds a preset maximum allowable charging current. Specifically, as shown in formula (9) below:

[0106] IF T t >T safe OR I target (t)>I max THEN Abort, (9)

[0107] Among them, T t T represents the battery temperature. safe Indicates the maximum safe temperature, I target (t) represents the target charging current, I max This indicates the maximum allowable charging current. If any of the above conditions are triggered, the system will end the current charging cycle early and record the status flag.

[0108] Steps S201 to S202 as shown in the embodiments of this application map the predicted charging power and predicted charging time into actual control execution operations, forming current control commands, time management mechanisms, and necessary safety protection boundary judgments in the charging box control chip, and finally completing the implementation of the scheduling strategy.

[0109] Steps S101 to S106 of this embodiment involve acquiring the vehicle's current battery state, user historical behavior characteristics, and grid information. A state harmonic term is constructed based on the current battery state, user historical behavior characteristics, and grid information; wherein the state harmonic term includes a battery state vector and a user behavior-grid state vector. The battery state vector is input into a preset battery prediction model to obtain the predicted remaining battery lifespan and predicted battery health level. The user behavior-grid state vector, the predicted remaining battery lifespan, and the predicted battery health level are input into a preset charging prediction model to obtain the predicted charging demand and predicted charging period. The predicted remaining battery lifespan, predicted battery health level, predicted charging demand, and predicted charging period are input into a preset power prediction model to obtain the predicted charging power and predicted charging duration. The vehicle is charged according to the predicted charging power and predicted charging duration. By incorporating factors such as the current battery state, user historical behavior characteristics, and grid information into the charging decision-making process in real time, the charging strategy is optimized, thereby extending battery lifespan and improving the grid's load capacity.

[0110] Please see Figure 3 This application also provides a charging box charging and discharging intelligent management system, which can realize the above-mentioned charging box charging and discharging intelligent management method. The system includes:

[0111] The acquisition module 301 is used to acquire the current battery status of the vehicle, user historical behavior characteristics, and power grid information;

[0112] Module 302 is used to construct a state harmonic term based on the current battery state, user historical behavior characteristics, and grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-grid state vector;

[0113] The first prediction module 303 is used to input the battery state vector into a preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level.

[0114] The second prediction module 304 is used to input the user behavior-grid state vector, the predicted remaining battery life and the predicted battery health level into the preset charging prediction model to obtain the predicted charging demand and the predicted charging period.

[0115] The third prediction module 305 is used to input the predicted remaining battery life, predicted battery health level, predicted charging demand and predicted charging period into a preset power prediction model to obtain the predicted charging power and predicted charging time.

[0116] The charging module 306 is used to charge the vehicle based on the predicted charging power and the predicted charging time.

[0117] The specific implementation method of the intelligent charging and discharging management system for the charging box is basically the same as the specific implementation method of the intelligent charging and discharging management method for the charging box described above, and will not be repeated here.

[0118] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described intelligent charging and discharging management method for the charging box. This electronic device can be any intelligent terminal, including tablet computers, in-vehicle computers, etc.

[0119] Please see Figure 4 , Figure 4 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0120] The processor 401 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0121] The memory 402 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 402 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 402 and is called and executed by the processor 401 to execute the intelligent charging and discharging management method for the charging box according to the embodiments of this application.

[0122] Input / output interface 403 is used to implement information input and output;

[0123] The communication interface 404 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0124] Bus 405 transmits information between various components of the device (e.g., processor 401, memory 402, input / output interface 403, and communication interface 404);

[0125] The processor 401, memory 402, input / output interface 403 and communication interface 404 are connected to each other within the device via bus 405.

[0126] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent charging and discharging management method for a charging box.

[0127] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0128] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0129] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0130] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0131] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0132] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0133] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0134] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above 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 mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of the system or units may be electrical, mechanical, or other forms.

[0135] The units described above 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.

[0136] 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.

[0137] 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 medium. 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 storage medium and includes multiple 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 of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0138] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for intelligent management of charging and discharging of a charging box, characterized in that, The method comprises: Obtain the vehicle's current battery status, user historical behavior characteristics, and power grid information; A state harmonic term is constructed based on the current battery state, the user's historical behavior characteristics, and the power grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-power grid state vector; The battery state vector is input into a preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level; wherein, the battery prediction model includes a thermal-electrical joint anomaly sensitivity regularization term and a target weight dynamic adjustment mechanism; The thermo-electric combined anomaly sensitivity regularization term is: ； in, This represents the regularization term for the combined thermal and electrical anomaly sensitivity. This indicates the actual remaining battery life. This indicates the predicted remaining battery life. Indicates the actual battery health level. This indicates the predicted battery health level. Indicates the first The rate of change of battery temperature in each training sample This represents the rate of change of the battery's internal resistance. This represents the weight of the error term in the prediction of the remaining battery life. This indicates the weight of the error term in the battery health level prediction. This indicates the weight of the thermo-electric coupling regularization term; The target weight dynamic adjustment mechanism is as follows: ； in, This represents the Sigmoid function. Indicates the voltage regulation coefficient. Indicates the nominal voltage. Indicates battery voltage. Represents a small constant; The user behavior-grid state vector, the predicted remaining battery life, and the predicted battery health level are input into a preset charging prediction model to obtain the predicted charging demand and the predicted charging period. The predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period are input into a preset power prediction model to obtain the predicted charging power and predicted charging time. The vehicle is charged according to the predicted charging power and the predicted charging duration.

2. The method according to claim 1, characterized in that, After inputting the predicted remaining battery life, predicted battery health level, predicted charging demand, and predicted charging period into a preset power prediction model to obtain the predicted charging power and predicted charging duration, the method further includes: The predicted charging power is corrected based on the preset maximum sustainable safe charging power, the preset maximum battery life, the predicted remaining battery life, and the predicted charging power to obtain the corrected predicted charging power. The predicted charging duration is corrected based on the predicted charging period and the predicted charging duration to obtain the corrected predicted charging duration.

3. The method according to claim 1, characterized in that, The step of charging the vehicle based on the predicted charging power and predicted charging time includes: The predicted charging power is processed by current conversion to obtain the target charging current; The vehicle is charged according to the target charging current and the predicted charging time.

4. The method according to claim 3, characterized in that, The current state of the battery includes the battery temperature. During the process of charging the vehicle based on the predicted charging power and predicted charging duration, the method further includes: If the battery temperature exceeds the preset maximum safe temperature, and / or the target charging current exceeds the preset maximum allowable charging current, charging will stop.

5. The method according to claim 1, characterized in that, The loss function of the charging prediction model is: ； in, This represents the loss function of the charging prediction model. This indicates the predicted charging demand. Indicates the actual charging requirement. This indicates the predicted charging period. Indicates the actual charging period. This represents the adjustment coefficient. This indicates the predicted battery health level.

6. The method according to claim 1, characterized in that, The power prediction model includes an optimizer, which is: ； in, This indicates the predicted charging demand. This indicates the predicted charging power. This indicates the predicted charging time. This indicates the predicted battery health level. This indicates the current grid load threshold. Weights representing user preferences Indicates the weight of health protection. Weights representing grid-friendliness.

7. A charging and discharging intelligent management system for a charging box, characterized in that, The system includes: The acquisition module is used to acquire the vehicle's current battery status, user historical behavior characteristics, and power grid information; A construction module is used to construct a state harmonic term based on the current state of the battery, the user's historical behavior characteristics, and the power grid information; wherein, the state harmonic term includes a battery state vector and a user behavior-power grid state vector; The first prediction module is used to input the battery state vector into a preset battery prediction model to obtain the predicted remaining battery life and the predicted battery health level; wherein, the battery prediction model includes a thermal-electrical joint anomaly sensitivity regularization term and a target weight dynamic adjustment mechanism. The thermo-electric combined anomaly sensitivity regularization term is: ； in, This represents the regularization term for the combined thermal and electrical anomaly sensitivity. This indicates the actual remaining battery life. This indicates the predicted remaining battery life. Indicates the actual battery health level. This indicates the predicted battery health level. Indicates the first The rate of change of battery temperature in each training sample This represents the rate of change of the battery's internal resistance. This represents the weight of the error term in the prediction of the remaining battery life. This indicates the weight of the error term in the battery health level prediction. This indicates the weight of the thermo-electric coupling regularization term; The target weight dynamic adjustment mechanism is as follows: ； in, This represents the Sigmoid function. Indicates the voltage regulation coefficient. Indicates the nominal voltage. Indicates battery voltage. Represents a small constant; The second prediction module is used to input the user behavior-grid state vector, the predicted remaining battery life and the predicted battery health level into a preset charging prediction model to obtain the predicted charging demand and the predicted charging period. The third prediction module is used to input the predicted remaining battery life, predicted battery health level, predicted charging demand and predicted charging period into a preset power prediction model to obtain the predicted charging power and predicted charging time. A charging module is used to charge the vehicle according to the predicted charging power and the predicted charging time.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the intelligent charging and discharging management method for the charging box according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent charging and discharging management method for the charging box as described in any one of claims 1 to 6.

Images