Management method of ternary polymer lithium battery for high cycle number shared charging equipment

Abstract

The present invention relates to a ternary polymer lithium battery management method for a high-cycle shared charging device. The method: graph theory modeling of the battery management system in the shared charging device to construct a topological graph model; depth-first search, identification of the charge and discharge path and calculation of the composite impedance value; calculation of the dynamic voltage spectrum and impedance characteristics to obtain the time-domain current distribution; according to the time-domain current distribution, multi-factor coupling loss calculation is performed on each ternary polymer lithium battery unit to obtain dynamic loss distribution data; based on the dynamic loss distribution data, the battery state is evaluated in real time, and a nonlinear charge and discharge characteristic curve is reconstructed; based on the nonlinear charge and discharge characteristic curve, multi-time scale collaborative scheduling is performed to obtain a battery charge and discharge control strategy. The implementation of the present invention can accurately model the nonlinear characteristics of the battery, evaluate the battery state in real time, and perform intelligent scheduling based on the dynamic prediction results to maximize the battery life and the overall efficiency of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery management, and in particular to a ternary polymer lithium battery management method for a shared charging device with a high number of cycles. Background Art

[0002] As an emerging mobile energy solution, shared charging equipment is rapidly gaining popularity. However, with the increase in frequency of use, battery management faces severe challenges. Traditional battery management systems often have difficulty coping with complex operating conditions under high-frequency usage scenarios, resulting in reduced battery performance, shortened life, and even safety hazards. In addition, existing battery management methods are mostly based on simplified models, which cannot accurately characterize the nonlinear characteristics of batteries under high cycle times, resulting in deviations in battery capacity estimation and low charging and discharging efficiency.

[0003] Another prominent problem is that existing battery management systems lack the ability to adapt to multi-time scale load changes. The usage patterns of shared charging equipment are significantly random and volatile. Traditional fixed strategies are difficult to balance the contradiction between immediate needs and long-term performance maintenance, often resulting in waste of battery resources or degradation of service quality. At the same time, the inconsistency between battery cells increases with the increase in the number of uses. Existing balancing strategies are difficult to effectively handle this dynamic change, further exacerbating the performance degradation of the battery pack. Summary of the invention

[0004] The main purpose of the present invention is to provide a ternary polymer lithium battery management method for a shared charging device with a high number of cycles. The present invention can accurately model the nonlinear characteristics of the battery, evaluate the battery status in real time, and perform intelligent scheduling based on dynamic prediction results to maximize the battery life and overall system efficiency.

[0005] To achieve the above object, the present invention provides a ternary polymer lithium battery management method for a shared charging device with a high cycle number, comprising the following steps:

[0006] Conduct graph theory modeling of the battery management system in the shared charging device, and construct a topological graph model containing vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths;

[0007] Performing a depth-first search based on the topology graph model to identify all charge and discharge paths that meet the circuit rules, and calculating the composite impedance value of each charge and discharge path;

[0008] Calculating a dynamic voltage spectrum and an impedance characteristic based on the charge and discharge path and the composite impedance value to obtain a time domain current distribution of each ternary polymer lithium battery cell;

[0009] According to the time-domain current distribution, a multi-factor coupling loss calculation is performed on each ternary polymer lithium battery unit to obtain dynamic loss distribution data;

[0010] Based on the dynamic loss distribution data, a real-time evaluation of the battery state is performed, and a nonlinear charge and discharge characteristic curve is reconstructed;

[0011] Multi-time scale coordinated scheduling is performed based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy.

[0012] The present invention also provides a solid-state battery management device for a shared charging device with a high cycle number, comprising:

[0013] A modeling module, used for graph-theoretic modeling of the battery management system in the shared charging device, and constructing a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths;

[0014] An identification module, used to perform a depth-first search based on the topological graph model, identify all charge and discharge paths that meet the circuit rules, and calculate the composite impedance value of each charge and discharge path;

[0015] A calculation module, used to calculate the dynamic voltage spectrum and impedance characteristics based on the charge and discharge path and the composite impedance value, and obtain the time domain current distribution of each ternary polymer lithium battery unit;

[0016] A processing module, used to perform multi-factor coupling loss calculation on each ternary polymer lithium battery unit according to the time domain current distribution to obtain dynamic loss distribution data;

[0017] An evaluation module, used for performing real-time evaluation of the battery state based on the dynamic loss distribution data, and reconstructing a nonlinear charge and discharge characteristic curve;

[0018] The scheduling module is used to perform multi-time scale coordinated scheduling based on the nonlinear charging and discharging characteristic curve to obtain a battery charging and discharging control strategy.

[0019] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0020] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0021] In summary, the technical solution provided by the present invention can comprehensively identify effective charging and discharging paths by abstracting the battery management system into a topological graph model and combining it with a depth-first search algorithm. Considering the coupling effect of Joule heat loss, electrochemical reaction loss and cycle attenuation loss, the dynamic loss distribution of the battery can be more accurately evaluated by solving the multi-physics field model. It helps to optimize the charging and discharging strategy and extend the battery life. Based on the adaptive OCV-SOC mapping model and online parameter identification, the nonlinear characteristic equation of the battery is dynamically updated. It can adapt to the dynamic changes in battery performance, improve the accuracy of SOC estimation, and combine the day-ahead load forecast, intraday dynamic adjustment and real-time fine-tuning to achieve a full range of scheduling from long-term planning to immediate response. This multi-level scheduling strategy can effectively balance the battery efficiency and life extension and improve the overall performance of the system. Deep learning models such as LSTM, one-dimensional convolutional neural network and GRU are used to process long-term trends, periodic patterns and short-term fluctuations respectively, and the prediction results are integrated through a multi-layer perceptron fusion network. This composite prediction method can more accurately capture the characteristics of load changes, continuously record the charging and discharging process data, continuously update the battery state evaluation model and nonlinear characteristic curve, and realize the adaptive optimization of the management strategy. The present invention enables the battery management system to always maintain optimal performance and adapt to changes in the entire life cycle of the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic diagram of the steps of a ternary polymer lithium battery management method for a shared charging device with a high number of cycles in one embodiment of the present invention;

[0023] Figure 2 This is a structural block diagram of a solid-state battery management device for a shared charging device with a high number of cycles in one embodiment of the present invention;

[0024] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0025] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0027] Reference Figure 1 This embodiment provides a ternary polymer lithium battery management method for a shared charging device with a high cycle number, comprising the following steps:

[0028] S1, conduct graph theory modeling of the battery management system in the shared charging device, and construct a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths;

[0029] Among them, the ternary polymer lithium battery cell, charging circuit, discharge circuit and protection circuit of the battery management system in the shared charging device are identified, and the type and parameter information of each component, such as the capacity, internal resistance, state of charge of the battery, and the electrical characteristics of each circuit, are obtained. Based on the type and parameter information of each component, vertex modeling is performed on each ternary polymer lithium battery cell and circuit component. In this process, each battery cell and circuit component is abstracted as a vertex, and all vertices together constitute a vertex set. While constructing the vertex set, the current flow path between the components in the battery management system is analyzed to determine the interconnection relationship between these components. By analyzing the path of the current flowing from the battery cell through each circuit component, the flow direction of the current between the components is determined, and a directed edge set is established based on this, which represents the channel for the current flow. Attributes are assigned to each vertex in the vertex set. In the process of assigning vertex attributes, specific attributes of the corresponding component are specified for each vertex. For example, the attribute data of the battery cell vertex should include its rated capacity, internal resistance and current state of charge, while for the vertex of the circuit component, the attributes include the type of component and its electrical parameters. At the same time, attributes are assigned to each edge in the directed edge set. The attribute data of these edges mainly include the direction of current flow, the maximum allowable current value, and the current actual current value, so as to reflect the actual flow of current in the system. Based on the above vertex set, the directed edge set containing the attribute data of each vertex, and the edge attribute data, the topological graph model of the entire battery management system is finally constructed, which fully describes the connection relationship between each battery cell and circuit component in the system, as well as the flow path of current in the entire system.

[0030] S2, performing a depth-first search based on the topology graph model, identifying all charge and discharge paths that meet the circuit rules, and calculating the composite impedance value of each charge and discharge path;

[0031] Specifically, the vertex set in the topological graph model is encoded to obtain a unique identifier for each vertex. At the same time, an access status flag is set for each vertex to track whether the vertex has been visited during the search process. A depth-first search is implemented by constructing a recursive function, which will be responsible for starting from any vertex and gradually exploring all possible paths along the directed edges in the graph. Based on the traversal process of the recursive function, all vertices in the topological graph model will be visited one by one, and the access sequence of each vertex will be recorded. Whenever a new vertex is visited, it is verified according to the circuit rules to ensure that the path corresponding to the access sequence meets the physical and circuit constraints of current flow. Only the verified paths will be considered as valid paths, and these paths will be collected into the valid charge and discharge path set. As the recursive function is continuously executed, all charge and discharge paths that meet the conditions will be identified and recorded to form a charge and discharge path set. Based on the valid charge and discharge path set, the resistance, inductance and capacitance of each charge and discharge path are extracted. By analyzing the component characteristics on each path, a distributed parameter model related to the path is obtained, reflecting the influence of each component on the path on current transmission. The distributed parameter model is mathematically transformed to obtain the composite impedance expression of each path in the frequency domain. This expression combines the interaction of resistance, inductance and capacitance to form an equivalent model in the frequency domain, which can accurately reflect the impedance characteristics of the path at different frequencies. Based on the frequency domain composite impedance expression of each path, the equivalent impedance of the charge and discharge path within the operating frequency range is calculated. The calculation process takes into account the frequency response characteristics of each component on the path, and the composite impedance value of the path under actual working conditions is obtained by analyzing the impedance in the frequency domain. The composite impedance value of each path can accurately describe the energy loss and other related characteristics of the path during the current flow process.

[0032] S3, calculating the dynamic voltage spectrum and impedance characteristics based on the charge and discharge path and the composite impedance value, and obtaining the time domain current distribution of each ternary polymer lithium battery unit;

[0033] It should be noted that the topological structure of the charge and discharge path is analyzed, and by analyzing the interaction between the battery cells and each circuit element on the path, a time domain differential equation group that can describe the dynamic behavior of the entire battery management system is obtained. The time domain differential equation group can reflect the time-varying relationship between current and voltage in the system. The time domain differential equation group is converted into an s-domain transfer function to obtain the frequency domain mathematical model of the battery management system. Based on the frequency domain mathematical model and the composite impedance value, the system matrix, input matrix, output matrix and feedforward matrix of the battery management system are constructed. These matrices can characterize the input-output relationship of the system and the flow of current between the components. These matrices are discretized and converted from continuous time models to discrete time system models, so that they are suitable for simulation and control in a digital environment. Parameter identification is performed on the discrete time system model. By analyzing the model, the characteristic parameters of the system are identified to accurately reflect the dynamic characteristics of the battery management system. Based on the system characteristic parameters, a broadband excitation signal is applied to the system, which can cover multiple frequency ranges so as to fully observe the response of the system at each frequency. After applying the excitation signal, the output of the system is analyzed using fast Fourier transform to calculate the dynamic voltage spectrum, which reflects the voltage response of the battery management system at different frequencies. Based on the dynamic voltage spectrum and the input current information, the impedance characteristics of the battery management system are calculated. The impedance characteristics reflect the electrical characteristics of the battery system at different frequencies, especially the energy loss and current distribution during the charging and discharging process. In order to convert the results of the frequency domain calculation back to the time domain, the impedance characteristics are inverse Fourier transformed to restore the impedance information in the frequency domain to the time domain. Combined with the topological structure information of the circuit, the time domain current distribution of each ternary polymer lithium battery unit is finally obtained.

[0034] S4, performing multi-factor coupling loss calculation on each ternary polymer lithium battery unit according to the time domain current distribution to obtain dynamic loss distribution data;

[0035] Specifically, based on the time domain current distribution, the square integral operation is performed on the current of each battery cell to preliminarily estimate the Joule heat loss of the battery cell during operation. Since the internal resistance of the battery changes dynamically, the temperature correction is performed on the initially calculated Joule heat loss value based on the real-time updated battery internal resistance parameters to obtain more accurate Joule heat loss data. Based on the charge and discharge state of the battery cell, the charge transfer impedance generated during the charge transfer process is calculated. This impedance reflects the energy loss caused by the charge transfer process in the electrochemical reaction of the battery. By calculating the charge transfer impedance, the basic data of the electrochemical reaction loss is obtained. The basic data of the electrochemical reaction loss is nonlinearly compensated to obtain the corrected electrochemical reaction loss data. At the same time, based on the number of battery cycles and the deep discharge state, a battery capacity decay model is constructed. According to the number of battery cycles and the deep discharge state, the model estimates the capacity decay degree of the battery during each charge and discharge process, and then calculates the corresponding cycle decay loss data. Based on the multi-physics field coupling model, the Joule heat loss data, electrochemical reaction loss data and cycle decay loss data are solved simultaneously. The multi-physics coupling model can comprehensively consider the combined effects of multiple factors such as electricity, heat and chemistry on battery loss. The process of joint solution correlates these factors to obtain the overall coupling loss data, which comprehensively reflects the actual loss of the battery during operation. The coupling loss data is analyzed in time series, and the law of dynamic loss change over time is revealed through the time series analysis method, and the time-varying characteristics of the dynamic loss distribution are obtained. Based on the time-varying characteristics of the dynamic loss distribution, the Joule heat loss data, electrochemical reaction loss data and cycle attenuation loss data are weighted and fused. In different time periods and working conditions, the data of the three types of losses are reasonably integrated according to their respective weights, and finally the complete dynamic loss distribution data is obtained.

[0036] S5, performs real-time evaluation of the battery status based on the dynamic loss distribution data, reconstructs the nonlinear charge and discharge characteristic curve, and performs multi-time scale coordinated scheduling based on the nonlinear charge and discharge characteristic curve to obtain the battery charge and discharge control strategy.

[0037] Among them, the time domain current distribution is processed by piecewise linearization, and the original continuous current distribution is discretized into several current sampling sequences. Based on the discretized current sampling sequence, the change in the state of charge (SOC) of the battery is calculated to obtain the initial SOC estimate. The initial SOC estimate is compensated and corrected according to the dynamic loss distribution data to obtain a more accurate corrected SOC value. In order to improve the accuracy of SOC estimation, an adaptive OCV-SOC (open circuit voltage-state of charge) mapping model is constructed based on historical charge and discharge data. The model can dynamically adapt to changes in battery performance through the accumulation and analysis of historical data. In order to ensure the real-time accuracy of the model, the adaptive OCV-SOC mapping model is online parameter identification, and the parameters of the model are continuously updated using the forgetting factor method to obtain an updated OCV-SOC mapping relationship, ensuring that the model can self-adjust to adapt to changes in the battery state during long-term operation of the battery. Based on the corrected SOC value and the updated OCV-SOC mapping relationship, the instantaneous open circuit voltage of the battery is calculated to obtain the SOC-OCV data pair, which reflects the battery open circuit voltage under different SOC states. By fitting the SOC-OCV data pair, the nonlinear characteristic equation of the charge and discharge process is obtained. This equation can accurately describe the characteristics of the battery under different charge and discharge states, and can provide a key mathematical model for predicting the battery's capacitance, internal resistance and other parameters under different SOCs. Based on the nonlinear characteristic equation, the capacity and internal resistance of the battery under different SOC states are calculated. The nonlinear charge and discharge characteristic curve reflects the working characteristics of the battery under different states and can effectively guide the battery's charge and discharge process in actual work. Through this curve, accurate guidance is provided for the battery management system to ensure that the battery's working efficiency is maximized under different charge states, and that the loss can be reduced and the battery's service life can be extended. Based on the nonlinear charge and discharge characteristic curve, multi-time scale coordinated scheduling is achieved. Multi-time scale scheduling means that in different time ranges, such as short-term real-time control and long-term energy scheduling, optimization and adjustment can be performed based on the nonlinear characteristics of the battery. Through multi-time scale coordinated scheduling, it can be ensured that the battery's charge and discharge strategy can not only meet the short-term high-efficiency energy utilization needs, but also optimize the long-term battery health management, and finally obtain a complete battery charge and discharge control strategy.

[0038] Among them, based on the nonlinear charge and discharge characteristic curve, the charge and discharge efficiency and capacity limit of the battery under different states of charge (SOC) are determined. By analyzing the working characteristics of the battery in different SOC ranges, the optimal operating conditions of the battery under each state and the performance upper limit during charge and discharge are clarified. According to the charge and discharge efficiency and capacity limit, combined with historical power consumption data, the day-ahead load forecast is carried out to obtain the day-ahead load forecast result. Based on the day-ahead load forecast result, a day-ahead charge and discharge preliminary scheduling plan is formulated to ensure that the battery charge and discharge behavior matches the future load demand and optimize the energy storage and release process. The day-ahead charge and discharge preliminary scheduling plan is rolled optimized. Through rolling optimization, the forecast data is dynamically adjusted over time to make the plan more in line with the actual load demand and obtain the intra-day dynamic adjustment strategy. As the load of the battery changes in real time during actual operation, the intra-day dynamic adjustment strategy also needs to be fine-tuned according to the actual real-time load information. By continuously adjusting the strategy, the management system can generate real-time control instructions for executing specific battery charge and discharge operations. These control instructions are based on the latest load data and the optimized charge and discharge strategy, which can ensure the efficient operation of the battery under different load conditions. At the same time, various data in the charge and discharge process are recorded during the execution of the charge and discharge operation. Based on the charging and discharging process data, the battery status assessment model needs to be updated regularly. As the assessment model is updated, the original nonlinear charging and discharging characteristic curve is optimized and adjusted so that the new curve can better reflect the current operating status and performance changes of the battery. Finally, based on the updated nonlinear charging and discharging characteristic curve, multi-time scale coordinated scheduling adjustments are performed. Combining long-term and short-term load demands, it is ensured that the operation of the battery in different time ranges not only meets the immediate energy demand, but also ensures the long-term service life of the battery, and generates the final battery charging and discharging control strategy.

[0039] Furthermore, the historical electricity consumption data is decomposed into three main components: trend term, seasonal term and random term. By analyzing different components, long-term trends, periodic changes and short-term fluctuations are captured respectively. The trend term is input into the long-term trend prediction model for prediction. The long-term trend prediction model adopts the LSTM (Long Short-Term Memory) network structure, which has strong memory capacity and can well capture the long-term dependencies in the time series. The LSTM network model consists of an input layer, two hidden layers and an output layer. Each hidden layer contains 128 neurons, and the tanh activation function is used to introduce nonlinear characteristics to enhance the prediction ability of the model. Through the LSTM model, the load trend of the next 24 hours can be predicted to obtain the long-term trend prediction result. At the same time, the seasonal term is input into the periodic pattern recognition model for processing. Since seasonal changes have obvious periodic characteristics, the model adopts a one-dimensional convolutional neural network (CNN) structure. The network contains 3 convolutional layers and 2 fully connected layers. The convolutional layer performs nonlinear mapping through the ReLU activation function, and batch normalization is used to accelerate the training process and improve the stability of the model. The convolutional network is used to identify periodic patterns in time series. It is used to make periodic load forecasts for the next 24 hours and obtain periodic load forecast results. The random item part is input into the short-term fluctuation forecast model for processing. The short-term fluctuation model adopts the GRU (Gated Recurrent Unit) network structure. GRU is a simplified RNN structure that can effectively capture the short-term change characteristics in time series. The model contains a bidirectional GRU layer, which contains 64 hidden units to capture the before and after relationship of random fluctuations. The model also contains a fully connected layer to output the final short-term fluctuation forecast results. Through the GRU model, the short-term fluctuation of the load in the next 24 hours is predicted. The long-term trend forecast results, periodic load forecast results and short-term fluctuation forecast results are integrated into a multi-layer perceptron (MLP) fusion network. The fusion network consists of 3 hidden layers, containing 64, 32 and 16 neurons respectively. The LeakyReLU activation function is used to process the complex nonlinear relationship of the load data, and the last layer uses a linear activation function to ensure the continuity of the forecast results. By fusing information from different sources, a more accurate and comprehensive day-ahead load forecast result is finally obtained.

[0040] In this embodiment, the long-term trend prediction results, the periodic load prediction results and the short-term fluctuation prediction results are feature fused. The prediction results of different time scales are feature spliced ​​to form a fusion input vector. The fusion input vector is input to the first hidden layer of the multilayer perceptron. The first hidden layer contains 64 neurons, and the output of each neuron is processed by the LeakyReLU activation function. The LeakyReLU activation function can avoid the "dead neuron" problem existing in the traditional ReLU activation function, thereby ensuring that the model has better performance when processing complex nonlinear data. After LeakyReLU processing, the output of the first hidden layer is obtained. In order to improve the stability of the model and accelerate the training process, the output of the first hidden layer is batch normalized. Batch normalization eliminates the changes in internal covariates by standardizing the output data and improving the convergence speed of the model. The batch-normalized output is added to the initial fusion input vector through a residual connection to obtain the residual output of the first layer. The residual connection mechanism can alleviate the gradient vanishing problem that may occur in deep network training, and retain some original information of the input layer to improve the performance of the model. The residual output of the first layer is input to the second hidden layer. The second hidden layer contains 32 neurons, and the output of each neuron is also processed nonlinearly by the LeakyReLU activation function. To prevent the model from overfitting, the output of the second hidden layer is subjected to a dropout operation, and the dropout rate is set to 0.3. This means that in each training, 30% of the neurons are randomly dropped, which forces the network to learn different feature combinations in different training cycles and enhance the generalization ability of the model. The output after the dropout operation, that is, the regularized output of the second hidden layer, is used as the input of the third hidden layer. The regularized output of the second hidden layer is input to the third hidden layer. The third hidden layer contains 16 neurons, and its output is processed by the LeakyReLU activation function. Although the third hidden layer has fewer neurons than the first two layers, it can further extract high-level features of the data. The output of the third hidden layer is input to the output layer. The output layer contains 24 neurons, representing the load forecast value for the next 24 hours. To ensure the continuity and accuracy of the forecast output, the output layer uses a linear activation function. This activation function is suitable for regression tasks and does not perform nonlinear transformation on the output value, thereby retaining the true range of the forecast value. Through this process, the fusion network combines information on long-term trends, cyclical patterns, and short-term fluctuations to ultimately produce a day-ahead load forecast.

[0041] In one example, a battery management system in a shared charging device is modeled by graph theory, and a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths is constructed, including:

[0042] Identify the ternary polymer lithium battery unit, charging circuit, discharging circuit and protection circuit of the battery management system in the shared charging device to obtain the type and parameter information of each component;

[0043] Based on the type and parameter information of each component, vertex modeling is performed on each ternary polymer lithium battery unit and circuit component to obtain a vertex set;

[0044] According to the circuit topology of the battery management system, the flow path of the current between the components is analyzed to obtain a directed edge set;

[0045] Assigning attributes to each vertex in the vertex set to obtain vertex attribute data including the rated capacity, internal resistance and current state of charge of the ternary polymer lithium battery unit and the type and parameters of the circuit component, and assigning attributes to each edge in the directed edge set to obtain edge attribute data including the current flow direction, the maximum allowable current and the current current value;

[0046] Based on vertex sets, directed edge sets and edge attribute data, a topological graph model of the battery management system is constructed.

[0047] In this example, the ternary polymer lithium battery cell, charging circuit, discharging circuit and protection circuit of the battery management system in the shared charging device are identified to obtain the type and parameter information of each component. For example, the main parameters of the ternary polymer lithium battery cell include its rated capacity, internal resistance and current state of charge (SOC), while for the charging circuit and the discharging circuit, their circuit structure, maximum allowable current and voltage limit need to be identified. The parameters of the protection circuit usually involve information such as the maximum overvoltage, undervoltage protection value and overcurrent protection threshold. Based on the type and parameter information of each component, vertex modeling is performed for each ternary polymer lithium battery cell and circuit component. Each ternary polymer lithium battery cell and circuit component can be regarded as a vertex in the graph. For example, assuming that a battery cell has a rated capacity of 5000mAh, an internal resistance of 50mΩ and a current state of charge of 80%, the battery cell can be represented as a vertex and these parameters can be used as attributes of the vertex. For the circuit component, assuming that the maximum allowable current of a charging circuit is 10A and the operating voltage range is 3.7V to 4.2V, this information will also be assigned to the corresponding vertex. After completing vertex modeling, analyze the current flow path in the battery management system. The current flow path between components can be represented as a set of directed edges in the graph. A directed edge represents the path of current flowing from one vertex (component) to another vertex. For example, when current flows from the charging circuit to the battery cell, this can be represented as a directed edge from the charging circuit vertex to the battery cell vertex. By analyzing the circuit topology, all current paths can be represented by directed edges to form a complete set of directed edges. Assign attributes to each vertex in the vertex set. These attributes include the rated capacity, internal resistance, and current state of charge of the battery cell, as well as the type and related parameter information of the circuit components. For example, for a ternary polymer lithium battery cell, the SOC can be represented as:

[0048] ;

[0049] in Represents the current remaining battery power. Represents the total capacity of the battery. Assuming that the remaining power of the current battery is 4000mAh and the total capacity is 5000mAh, the current SOC can be calculated to be 80%. For each edge in the directed edge set, attribute assignment is also performed. These attributes can include the direction of current flow, the maximum allowable current, and the current current value. Assuming that an edge represents the path from the charging circuit to the battery cell, its maximum allowable current is 10A, and the current actual current flowing is 8A, then the attribute data of the directed edge includes these current parameters. After the attributes of all vertices and edges are assigned, the topological graph model of the battery management system is constructed. The topological graph model consists of a vertex set and a directed edge set, which can describe the interconnection relationship between the components in the battery management system and the flow path of the current in the system.

[0050] In one example, a depth-first search is performed based on a topology graph model to identify all charge and discharge paths that meet the circuit rules, and the composite impedance value of each charge and discharge path is calculated, including:

[0051] Encode the vertex set in the topological graph model, obtain the unique identifier and access status flag of each vertex, and construct a recursive function of depth-first search;

[0052] The topology graph model is traversed based on a recursive function to obtain all possible vertex access sequences, and the circuit rules of each vertex access sequence are verified to obtain a set of effective charging and discharging paths;

[0053] Based on the effective charge and discharge path set, the resistance, inductance and capacitance of each charge and discharge path are extracted to obtain the distributed parameter model of the charge and discharge path, and the distributed parameter model of the charge and discharge path is mathematically transformed to obtain the frequency domain composite impedance expression of each charge and discharge path;

[0054] Based on the frequency domain composite impedance expression, the equivalent impedance of the charge and discharge path within the working frequency range is calculated to obtain the composite impedance value of each charge and discharge path.

[0055] In this example, each vertex in the topology graph model is uniquely encoded and initialized with a status flag. The unique identifier of each vertex can be a unique ID generated based on its position or function, such as using its component name, number, or position sequence. For the status flag, whether the vertex has been visited can usually be represented by a Boolean value, such as "0" for unvisited and "1" for visited. In this way, each vertex in the graph is given an independent identifier, and the access status can be tracked during the traversal process. Based on the above encoding, a recursive function for traversing the topology graph model is constructed. Depth-first search is a graph traversal algorithm that can start from any starting vertex and go deep into each vertex in the graph along a path until it cannot continue, then backtrack and continue to search for other paths. The core of the recursive function is to recursively call itself to traverse all possible paths. At each recursive call, the visited vertex is marked as visited to avoid repeated visits. In the set of directed edges in the graph, each edge represents the direction of current flow, so the depth-first search recursive process follows the direction of these edges to traverse, starting from the starting vertex, and visiting each adjacent vertex in turn until the entire circuit topology is traversed. During the recursive traversal, the depth-first search algorithm generates all possible vertex access sequences. Each access sequence represents a potential charge and discharge path. For each vertex access sequence, circuit rule verification is performed. Circuit rules may include current direction, voltage limits of nodes, and maximum allowable current on the path. For example, some paths may exceed the allowable value of voltage or current at some nodes, in which case the path will be considered invalid. Paths that meet the circuit rules will be retained as valid charge and discharge paths. Based on the set of valid charge and discharge paths, the resistance, inductance, and capacitance of each charge and discharge path are extracted. For the resistance, inductance, and capacitance of each path, the corresponding circuit element values ​​are used for superposition. After obtaining the distributed parameter model of each charge and discharge path, these parameters are mathematically transformed and converted into a frequency domain composite impedance expression. Composite impedance is a key parameter of a circuit under alternating current conditions and is usually a function of frequency. Through mathematical transformation, the equivalent impedance expression of the charge and discharge path in the frequency domain is obtained. For a charge and discharge path containing resistance, inductance, and capacitance, its composite impedance It can be expressed as:

[0056] ;

[0057] in, represents the composite impedance, is the resistor, is the inductor, is the capacitor, is the angular frequency, is an imaginary unit. This formula shows that impedance is not only related to resistance, but also affected by inductance and capacitance, and these effects behave differently at different frequencies. Through this mathematical transformation, the performance of the charge and discharge path in the frequency domain can be clearly described, and the impedance characteristics of the path at different frequencies can be revealed. Based on the frequency domain composite impedance expression, the equivalent impedance of the charge and discharge path within the operating frequency range is calculated. Equivalent impedance is the total impedance when multiple components work together at different frequencies. By calculating the equivalent impedance, the performance of the circuit under actual working conditions can be evaluated. For example, in a charging path, as the charging frequency increases, the effect of the inductor increases, while the effect of the capacitor decreases, so the equivalent impedance of the entire path will also change. By accurately calculating the equivalent impedance, it is possible to provide a basis for subsequent circuit optimization and scheduling. For example, suppose there is a simple system consisting of a ternary polymer lithium battery cell, a charging circuit, and a discharging circuit, where the internal resistance of the battery cell is 100 m , there is an inductor of 10μH in the charging circuit and a capacitor of 100μF in the discharging circuit. In this case, after traversing through the depth-first search, an effective path from the charging circuit to the battery cell and then to the discharging circuit is identified. Based on the parameters in the circuit, the resistance, inductance, and capacitance of the path are extracted. Using the composite impedance formula, the composite impedance at different frequencies is calculated. For example, when the angular frequency When , the composite impedance of the path is:

[0058] ;

[0059] The calculations show that:

[0060] .

[0061] In one example, the dynamic voltage spectrum and impedance characteristics are calculated based on the charge and discharge path and the composite impedance value to obtain the time domain current distribution of each ternary polymer lithium battery cell, including:

[0062] The topological structure of the charge and discharge path is analyzed to obtain a set of time-domain differential equations describing the dynamic behavior of the battery management system, and the time-domain differential equations are converted into s-domain transfer functions to obtain a frequency-domain mathematical model.

[0063] Based on the frequency domain mathematical model and the composite impedance value, the system matrix, input matrix, output matrix and feedforward matrix of the battery management system are constructed and discretized to obtain a discrete time system model.

[0064] Perform parameter identification on the discrete-time system model to obtain system characteristic parameters, and based on the system characteristic parameters, apply broadband excitation signals to the battery management system, and obtain the dynamic voltage spectrum through fast Fourier transform calculation;

[0065] Based on the dynamic voltage spectrum and input current information, the impedance characteristics of the battery management system are calculated, and the impedance characteristics are inverse Fourier transformed. Combined with the circuit topology information, the time domain current distribution of each ternary polymer lithium battery unit is obtained.

[0066] In this example, the electrical connections between the components in the battery management system are analyzed to extract the time domain behavior of the system. The charge and discharge paths of the battery management system can be regarded as an electrical network containing battery cells, charging circuits, discharging circuits, and protection circuits. In this network, the dynamic behavior of each component such as resistors, inductors, and capacitors can be described by time domain differential equations. For each charge and discharge path, Kirchhoff's voltage law and Kirchhoff's current law are used to establish a set of time domain differential equations based on the relationship between voltage, current, and circuit elements. For example, suppose a charge and discharge path contains a resistor ,inductance and capacitor , in the time domain, the current is described by the following equation and voltage The relationship between:

[0067] ;

[0068] in, is the voltage of the battery management system on this path, is the current in the path, is the resistance value on the path, is the inductor, is the capacitance. The time domain differential equation can fully describe the dynamic changes of voltage and current over time during the charging and discharging process of the battery. In order to simplify the analysis and convert it to the frequency domain for processing, the Laplace transform is used to convert the time domain differential equation into a transfer function in the frequency domain. In the Laplace transform, the derivative and integral operations are converted into multiplication and division operations in the frequency domain, respectively. For the above equation, after applying the Laplace transform, the frequency domain transfer function can be obtained:

[0069] ;

[0070] in, and are the Laplace transforms of voltage and current, respectively, is a complex frequency variable, which represents the complex frequency domain characteristics of the system. This transfer function describes the relationship between voltage and current in the frequency domain, and is also called the frequency domain mathematical model of the system. Based on the frequency domain mathematical model, the system matrix, input matrix, output matrix and feedforward matrix of the battery management system are constructed. For a linear time-invariant system, the state space equation can be expressed as:

[0071] ;

[0072] ;

[0073] in, is the time derivative of the system state variable, is the state vector, is the input vector, is the output vector, is the system matrix, is the input matrix, is the output matrix, is the feedforward matrix. The system matrix Describes the internal dynamics of the battery management system, the input matrix Represents the impact of the input signal on the system, the output matrix and the feedforward matrix They respectively reflect the relationship between the output and direct input of the system. In order to make the model suitable for digital control systems, it is discretized and the continuous time system is converted into a discrete time system so that the system can be simulated and controlled by numerical calculation. The discretized system can be expressed by a difference equation:

[0074] ;

[0075] ;

[0076] in, , , and They are the system matrix, input matrix, output matrix and feedforward matrix of the discrete time system. Through the transformation, the continuous dynamic behavior of the battery management system can be described and simulated with discrete time steps. Parameter identification is performed on the discrete time system model to estimate the system's characteristic parameters based on actual data. The purpose of parameter identification is to determine the system matrix , input matrix Specific values ​​of the above. By applying a known input signal and measuring the corresponding output signal. In order to ensure the accuracy of parameter identification, a broadband excitation signal is applied to the system. The signal can cover multiple frequency ranges to ensure that the dynamic behavior of the system at all relevant frequencies can be identified. After applying the broadband excitation signal, the dynamic voltage spectrum of the system is calculated by fast Fourier transform. The Fourier transform can convert time domain signals into frequency domain signals, thereby revealing the response characteristics of the system at different frequencies. Through fast Fourier transform, the voltage response spectrum of the battery management system in a wide frequency range is obtained. Based on the dynamic voltage spectrum and input current information, the impedance characteristics of the battery management system are calculated. Impedance reflects the system's resistance to current at different frequencies. In the frequency domain, impedance is defined as the ratio of voltage to current, that is:

[0077] ;

[0078] Through this formula, the composite impedance value of the battery system is calculated at different frequencies, and the behavior of the system at different operating frequencies is analyzed. The impedance characteristics are converted back to the time domain through inverse Fourier transform, and the impedance information in the frequency domain is restored to the time domain signal. Combined with the circuit topology, the time domain current distribution of each ternary polymer lithium battery unit is obtained. For example, suppose a charge and discharge path contains a resistor , an inductor and a capacitor The dynamic behavior of the circuit is described by establishing a time domain differential equation. The time domain equation is converted into a frequency domain transfer function using the Laplace transform:

[0079] ;

[0080] Based on the transfer function, the state space equation is constructed and discretized to obtain a discrete time system model. A broadband excitation signal is applied and the voltage spectrum is calculated by fast Fourier transform, and then the impedance characteristics at different frequencies are calculated using the impedance formula. The frequency domain information is restored to the time domain by inverse Fourier transform, and the time domain current distribution of each battery cell is calculated by combining the circuit topology information.

[0081] In one example, based on the time domain current distribution, a multi-factor coupling loss calculation is performed on each ternary polymer lithium battery unit to obtain dynamic loss distribution data, including:

[0082] Based on the time domain current distribution, the current square integral operation is performed on each ternary polymer lithium battery unit to obtain the initial value of Joule heat loss, and the temperature correction is performed on the initial value of Joule heat loss according to the real-time updated battery internal resistance parameter to obtain the Joule heat loss data;

[0083] The charge transfer impedance is calculated based on the charge and discharge state of the ternary polymer lithium battery unit to obtain the basic data of electrochemical reaction loss, and the basic data of electrochemical reaction loss is nonlinearly compensated to obtain the electrochemical reaction loss data;

[0084] Based on the battery cycle number and deep discharge state, a battery capacity attenuation model is constructed to obtain the cycle attenuation loss data during a single charge and discharge process;

[0085] Based on the multi-physics field coupling model, the Joule heat loss data, electrochemical reaction loss data and cycle attenuation loss data are solved simultaneously to obtain the coupling loss data;

[0086] The coupling loss data is analyzed in time series to obtain the time-varying characteristics of the dynamic loss distribution. Based on the time-varying characteristics of the dynamic loss distribution, the Joule heat loss data, electrochemical reaction loss data and cyclic attenuation loss data are weightedly fused to obtain the dynamic loss distribution data.

[0087] In this example, the square of the current is integrated based on the time domain current distribution to obtain the initial value of the Joule heat loss. It can be calculated by the following formula:

[0088] ;

[0089] in, is the current distribution of the battery cell in the time domain, Is the battery in time The internal resistance at the moment. By integrating the square of the current, the initial heat loss caused by the internal resistance of the battery during the charge and discharge process is obtained. Temperature correction is performed based on the real-time updated battery internal resistance parameters. The internal resistance of the battery decreases as the temperature increases, so it is necessary to adjust the initial value of the Joule heat loss through the temperature correction coefficient. Temperature correction usually relies on empirical formulas or experimental data to reflect the nonlinear relationship between internal resistance and temperature. The corrected Joule heat loss data is more in line with the actual operating status. In addition to Joule heat loss, battery losses also include losses from electrochemical reactions. Based on the charge and discharge state of the ternary polymer lithium battery unit, the charge transfer impedance is calculated, and the basic data of the electrochemical reaction loss is obtained from it. Charge transfer impedance refers to the resistance when charge is transferred between the electrode and the electrolyte interface, which is usually obtained through impedance spectroscopy analysis. Electrochemical reaction loss Charge transfer resistance With current The relationship is expressed as:

[0090] ;

[0091] This formula represents the loss due to charge transfer impedance under a given current distribution. Nonlinear compensation is performed on the basic data of electrochemical reaction losses. By performing nonlinear compensation on the charge transfer impedance, the electrochemical loss characteristics of the battery under different charging and discharging conditions can be more accurately reflected. As the battery is repeatedly charged and discharged, the capacity will gradually decay, so a capacity decay model needs to be constructed to predict the performance degradation of the battery. Battery capacity decay is usually closely related to the number of cycles and the deep discharge state of the battery. Each deep discharge will cause certain damage to the active materials of the battery, thereby reducing its available capacity. By tracking the number of cycles N and the deep discharge parameter DOD of the battery, a capacity decay model is established:

[0092] ;

[0093] in, is the remaining capacity of the battery, is the initial capacity, and is the attenuation coefficient related to the number of cycles and the depth of discharge. This model can be used to estimate the capacity attenuation loss of the battery during each charge and discharge process, and gradually update the health status of the battery over time. After obtaining the data of Joule heat loss, electrochemical reaction loss and capacity attenuation loss, based on the multi-physics field coupling model, the Joule heat loss data, electrochemical reaction loss data and cycle attenuation loss data are solved jointly to comprehensively analyze the loss of the battery in actual operation. Through the joint solution, the mutual influence between different losses is considered to obtain more accurate coupling loss data. For example, Joule heat loss will cause the battery temperature to rise, and the temperature increase will affect the electrochemical reaction rate of the battery, so there is an interaction between the two losses. By solving these coupling effects jointly, a more complete description of battery loss is obtained. Time series analysis of the coupling loss data reveals the dynamic change characteristics of the loss over time. Time series analysis can identify the long-term trend, seasonal fluctuations and random short-term fluctuations of the loss. Through time series analysis, the loss pattern of the battery in different usage scenarios is identified, for example, the Joule heat loss of the battery is large in a high temperature environment, while the electrochemical reaction loss is dominant at low temperatures. Based on the time-varying characteristics of dynamic loss distribution, Joule heat loss, electrochemical reaction loss and cycle attenuation loss are weighted and fused to obtain dynamic loss distribution data. Different weights are assigned to different types of losses, and they are reasonably fused according to their contribution to the overall loss. Assuming that Joule heat loss accounts for the main part at a certain moment, and electrochemical reaction loss is more prominent at another moment, by dynamically adjusting the weights of each loss, the actual loss of the battery can be more accurately reflected.

[0094] In one example, the battery status is evaluated in real time based on the dynamic loss distribution data, and the nonlinear charge and discharge characteristic curve is reconstructed. Multi-time scale coordinated scheduling is performed based on the nonlinear charge and discharge characteristic curve to obtain the battery charge and discharge control strategy, including:

[0095] The time-domain current distribution is processed piecewise linearly to obtain a discrete current sampling sequence, and based on the discrete current sampling sequence, the battery state of charge change is calculated to obtain an initial SOC estimation value;

[0096] According to the dynamic loss distribution data, the initial SOC estimation value is compensated and corrected to obtain a corrected SOC value;

[0097] An adaptive OCV-SOC mapping model is constructed based on historical charge and discharge data, and online parameter identification is performed on the adaptive OCV-SOC mapping model. The model parameters are updated using the forgetting factor method to obtain an updated OCV-SOC mapping relationship.

[0098] Based on the corrected SOC value and the updated OCV-SOC mapping relationship, the instantaneous open circuit voltage of the battery is calculated to obtain the SOC-OCV data pair;

[0099] The SOC-OCV data pair is curve-fitted to obtain the nonlinear characteristic equation of the charge and discharge process, and the battery capacity and internal resistance under different SOCs are calculated based on the nonlinear characteristic equation to obtain the nonlinear charge and discharge characteristic curve;

[0100] Multi-time scale coordinated scheduling is performed based on the nonlinear charging and discharging characteristic curve to obtain the battery charging and discharging control strategy.

[0101] In this example, the time domain current distribution is processed by piecewise linearization. In actual operation of the battery, the current changes continuously over time. Through piecewise linearization, the original continuous current signal is discretized into a series of sampling points. Each sampling point corresponds to a specific current value and time period, forming a discrete current sampling sequence. Based on the discretized current sampling sequence, the change in SOC can be calculated by the coulomb counting method. The coulomb counting method estimates the change in the state of charge based on the current flowing through the battery. The specific formula is as follows:

[0102] ;

[0103] in, Indicates the change in SOC, is the total capacity of the battery, It is The current in a time period, is the duration of the time period, is the total number of time periods. The initial SOC estimate of the battery is obtained by accumulating each time period of the discretized current sampling sequence. The initial SOC estimate is compensated and corrected according to the dynamic loss distribution data. The actual SOC of the battery is affected by loss factors such as Joule heat loss, electrochemical reaction loss, and capacity attenuation loss. Through the dynamic loss distribution data, the loss of the battery under the current operating conditions is estimated, so as to correct the initial SOC. An adaptive OCV-SOC (open circuit voltage-state of charge) mapping model is constructed based on historical charge and discharge data. The OCV-SOC relationship is an important characteristic of the battery, which describes the open circuit voltage (OCV) of the battery at different SOCs. By analyzing the historical charge and discharge data of the battery, an initial OCV-SOC mapping model is constructed and updated online over time. In order to ensure that the model can adapt to changes in battery performance, the OCV-SOC model is subjected to online parameter identification. Online parameter identification can reflect the current state of the battery by updating the model parameters in real time. This process uses the forgetting factor method, which is an algorithm that assigns decreasing weights to historical data, and is used to gradually forget early historical data and pay more attention to the latest battery status data. The update formula of the forgetting factor method is:

[0104] ;

[0105] in, It is The model parameters at time t, are new parameters based on the current data, is the forgetting factor (usually between 0 and 1). The closer it is to 1, the greater the influence of historical data; The closer it is to 0, the greater the impact of the latest data. Through this update mechanism, it is possible to ensure that the OCV-SOC mapping model is adjusted in real time as the battery state changes, and an updated OCV-SOC mapping relationship is obtained. After obtaining the corrected SOC value and the updated OCV-SOC mapping relationship, the instantaneous open circuit voltage of the battery is calculated. The instantaneous open circuit voltage is based on the voltage value under the current SOC state. By substituting the corrected SOC value into the OCV-SOC mapping model, the corresponding open circuit voltage value is obtained, and a series of SOC-OCV data pairs are generated, which reflects the voltage characteristics of the battery under different SOC states. The SOC-OCV data pairs are curve fitted to obtain the nonlinear characteristic equation of the charge and discharge process. There is usually no linear relationship between SOC and OCV, and a nonlinear fitting technique is required to find an optimal equation to describe this relationship. The fitted nonlinear characteristic equation can accurately describe the dynamic behavior of the battery during the charge and discharge process. Common fitting models include polynomial fitting or exponential function fitting, which can capture the voltage change law of the battery in different SOC ranges. Based on the nonlinear characteristic equation, the capacity and internal resistance of the battery at different SOCs are calculated. The capacity and internal resistance of the battery usually change with the change of SOC. The battery parameters under different SOC conditions are estimated through the nonlinear charge and discharge characteristic equation. The change of internal resistance has an important impact on battery performance because it determines the voltage drop and heat generation of the battery under high current. Through the accurate SOC-OCV model, the optimal operating parameters of the battery at each state of charge are calculated. Based on the nonlinear charge and discharge characteristic curve, multi-time scale coordinated scheduling is carried out. Multi-time scale scheduling refers to the optimization of charging and discharging strategies in different time ranges according to the current state and future needs of the battery. For example, on a short time scale, the system can adjust the charging current in real time to maximize the battery efficiency, while on a long time scale, the system can plan the battery's charge and discharge cycle to extend the battery life. Through this scheduling method, it can be ensured that the battery is in the best working state in different time periods, which not only ensures the stability of energy supply but also optimizes the battery life.

[0106] In one example, multi-time scale coordinated scheduling is performed based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy, including:

[0107] Based on the nonlinear charge and discharge characteristic curve, determine the charge and discharge efficiency and capacity limit at different charge states;

[0108] According to the charging and discharging efficiency and capacity limit, combined with the historical electricity consumption data, the day-ahead load forecast is carried out to obtain the day-ahead load forecast result;

[0109] Based on the day-ahead load forecast results, a preliminary day-ahead charging and discharging scheduling plan is formulated, and the preliminary day-ahead charging and discharging scheduling plan is optimized on a rolling basis to obtain a dynamic adjustment strategy within the day.

[0110] According to the real-time load changes, fine-tune the intraday dynamic adjustment strategy, generate real-time control instructions, and execute battery charging and discharging operations based on the real-time control instructions, and record the charging and discharging process data;

[0111] Based on the charge and discharge process data, the battery state evaluation model is updated, and the nonlinear charge and discharge characteristic curve is optimized to obtain an updated nonlinear charge and discharge characteristic curve;

[0112] Multi-time scale coordinated scheduling adjustment is performed according to the updated nonlinear charging and discharging characteristic curve to obtain the battery charging and discharging control strategy.

[0113] In this example, the charge and discharge efficiency and capacity limit of the battery at different states of charge (SOC) are determined based on the nonlinear charge and discharge characteristic curve. The charge and discharge efficiency of a battery is its ability to effectively store or release energy under different SOC conditions, while the capacity limit reflects the maximum energy that the battery can provide in each SOC range. The nonlinear charge and discharge characteristic curve is generated through historical data and real-time evaluation, which can describe the performance of the battery at different SOCs, including factors such as voltage, internal resistance and efficiency. By analyzing these curves, the optimal operating conditions and limit capacity of the battery in a specific SOC range are obtained. For example, suppose a battery has a charging efficiency of 95% at an SOC of 90%, and the charging efficiency drops to 80% at an SOC of 20%. The nonlinear change is due to the fact that it is more difficult to charge the battery at a high SOC and the efficiency is reduced. Through the data, the charge and discharge efficiency and capacity limit at different SOCs are reasonably determined. Based on the charge and discharge efficiency and capacity limit, combined with historical power consumption data, a day-ahead load forecast is performed. Based on previous power consumption patterns and trends, the power demand in the next 24 hours is predicted. Through time series analysis, it is assumed that the power consumption pattern over the past period of time shows a relatively regular trend and fluctuation. Using these historical data, the model predicts future load demand. Based on the predicted load demand, the battery management system formulates a charge and discharge plan in advance to ensure that the battery can release enough energy when the load is high, and to charge to restore the SOC when the load is low. The preliminary scheduling plan is mainly based on the current state of the battery and future demand, but since the prediction results may be uncertain, rolling optimization is required. Rolling optimization is a dynamic adjustment strategy that corrects the scheduling plan in real time according to the actual load changes during the execution process to make the plan more accurate. For example, if the actual load does not match the forecast, the system will automatically adjust the charge and discharge rate to ensure that the load demand is met. When the system enters the actual operation stage, the battery management system fine-tunes the intraday dynamic adjustment strategy according to the real-time load changes. This means that the system generates real-time control instructions based on the current power demand and the state of the battery. These control instructions include information such as the rate, time, and power output of charging or discharging, aiming to ensure that the battery operates within the optimal efficiency range. For example, if the system detects a sudden increase in the current load, it will increase the discharge rate to meet the demand, otherwise it will reduce the discharge or start charging. The execution of real-time control instructions is achieved through a specific battery management system (BMS), which adjusts parameters such as current and voltage to ensure the safety and efficiency of system operation. During the charging and discharging operation, the charging and discharging process data is recorded, including the changes in current, voltage, temperature and SOC. Through this information, the performance of the battery under current conditions is evaluated. This data can also be used to optimize the battery management strategy. For example, the system may find that the charging efficiency within a specific SOC range is lower than expected, and this information can be used to adjust future charging and discharging strategies.The battery state assessment model is updated based on the charge and discharge process data. The state of the battery changes with time and usage, so through continuous state assessment, the health status and capacity degradation of the battery can be accurately tracked. As the battery is used, its capacity will gradually decrease, while the internal resistance will gradually increase. The battery state of health (SOH) is estimated through the charge and discharge data. At the same time, these data are used to optimize the nonlinear charge and discharge characteristic curve to ensure that the curve can accurately reflect the actual working state of the battery. For example, if the system finds through data that the discharge efficiency of the battery is greatly reduced when the SOC is below 10%, the system will incorporate this change into the new nonlinear charge and discharge characteristic curve. As time goes by and data accumulates, the accuracy of the curve will continue to improve, thus better guiding future battery management. Multi-time scale coordinated scheduling adjustment is carried out through the optimized nonlinear charge and discharge characteristic curve. Multi-time scale scheduling performs coordinated optimization of charge and discharge in both short-term and long-term time ranges according to the current state and future needs of the battery. For example, on the short time scale, the charge and discharge rate is adjusted in real time to cope with instantaneous load fluctuations; on the long time scale, the overall charge and discharge cycle of the battery is planned to avoid frequent deep charge and discharge and extend the service life of the battery.

[0114] In one example, based on the charging and discharging efficiency and capacity limit, combined with historical electricity consumption data, a day-ahead load forecast is performed to obtain a day-ahead load forecast result, including:

[0115] Perform time series decomposition on historical electricity consumption data to obtain trend items, seasonal items, and random items;

[0116] The trend item is input into the long-term trend prediction model. The long-term trend prediction model adopts the LSTM network structure, which includes an input layer, two hidden layers and an output layer. Each hidden layer contains 128 neurons and uses the tanh activation function. The load trend of the next 24 hours is predicted through the long-term trend prediction model to obtain the long-term trend prediction result.

[0117] The seasonal term is input into the periodic pattern recognition model. The periodic pattern recognition model adopts a one-dimensional convolutional neural network structure, which includes 3 convolutional layers and 2 fully connected layers. Each convolutional layer uses the ReLU activation function and batch normalization. The periodic pattern recognition model is applied to the next 24 hours to obtain the periodic load forecast results.

[0118] The random items are input into the short-term fluctuation prediction model. The short-term fluctuation prediction model adopts a gated recurrent unit network structure, which includes a bidirectional GRU layer and a fully connected layer. The GRU layer contains 64 hidden units. The load fluctuation in the next 24 hours is predicted by the short-term fluctuation prediction model to obtain the short-term fluctuation prediction result.

[0119] The long-term trend forecast results, periodic load forecast results and short-term fluctuation forecast results are input into a multi-layer perceptron fusion network. The multi-layer perceptron fusion network contains 3 hidden layers, each containing 64, 32 and 16 neurons respectively. The LeakyReLU activation function is used, and the last layer uses a linear activation function to obtain the day-ahead load forecast result.

[0120] In this example, the historical electricity consumption data is decomposed into trend items, seasonal items, and random items. Through time series analysis, the long-term trend, cyclical fluctuations, and short-term random fluctuations in the data can be effectively separated. Decomposition into trend terms , Seasonal items and random items , the formula is as follows:

[0121] ;

[0122] in, Reflects long-term electricity consumption trends, such as whether electricity consumption shows an upward or downward trend over time; represents the periodic fluctuations in electricity consumption, which may be related to daily, weekly or annual patterns of electricity consumption; It is a short-term random fluctuation that cannot be explained by trend and seasonal patterns. Once the decomposition of the time series is completed, each part is modeled and predicted accordingly. The trend term reflects the long-term changes in electricity consumption and is suitable for prediction using a long-term trend prediction model. The long-term trend prediction model can adopt the LSTM (Long Short-Term Memory) network structure. The LSTM network can remember historical information over a long period of time when processing time series data, and is suitable for predicting long-term trends. The LSTM network contains an input layer, two hidden layers, and an output layer. Each hidden layer contains 128 neurons and uses the tanh activation function for nonlinear transformation. The tanh activation function can compress the data into the range of [-1, 1] and effectively handle the nonlinear characteristics in the time series. By inputting the trend term into the LSTM network, the model can predict the long-term electricity consumption trend in the next 24 hours and obtain the long-term trend prediction result. The seasonal term reflects the periodic pattern in the electricity consumption data. Since seasonal fluctuations usually have fixed periodic characteristics, such as daily morning and evening peak electricity consumption or weekend electricity consumption changes, it is suitable to use a one-dimensional convolutional neural network to capture this periodic pattern. The periodic pattern recognition model can adopt a structure including 3 convolutional layers and 2 fully connected layers. Each convolutional layer uses the ReLU activation function and batch normalization operation. The ReLU activation function can introduce nonlinearity after the convolution operation, while batch normalization can accelerate the training of the network and improve the stability of the prediction. By recognizing the periodic pattern of the seasonal item, the model can accurately predict the periodic power consumption fluctuations in the next 24 hours and obtain the periodic load prediction results. For the random item, since it reflects short-term random fluctuations, these fluctuations cannot be explained by long-term trends or periodic patterns, it is suitable to use the gated recurrent unit (GRU) network to predict short-term fluctuations. GRU is a simplified recurrent neural network structure that can efficiently capture short-term dependencies in time series. The short-term fluctuation prediction model can adopt a bidirectional GRU layer containing 64 hidden units and equipped with a fully connected layer to output the prediction results. The bidirectional GRU can consider both the past and future information of the time series, so it is more accurate in capturing short-term fluctuations. By inputting the random item into the short-term fluctuation prediction model, the short-term fluctuations in the next 24 hours are predicted to obtain the short-term fluctuation prediction results. The long-term trend forecast results, periodic load forecast results, and short-term fluctuation forecast results are input into a multi-layer perceptron fusion network. The fusion network contains 3 hidden layers, containing 64, 32, and 16 neurons respectively, and each layer uses the LeakyReLU activation function. LeakyReLU is an activation function that can effectively avoid the "death" of neurons and is suitable for solving tasks with complex data distribution. The last layer uses a linear activation function to ensure that the output is a continuous load forecast result. The three types of forecast results, long-term trend, periodic pattern, and short-term fluctuation, are input into the multi-layer perceptron network as input.The fusion network can generate a more accurate comprehensive load forecast by learning the complex relationship between these inputs. The final output layer contains 24 neurons, corresponding to the load forecast value for each hour in the next 24 hours. Assume that the output of the fusion network is. ,So:

[0123] ;

[0124] in, is the final predicted load value, is the long-term trend forecast result. is the periodic load forecast result, is the short-term volatility forecast result. , , are the weights learned by the fusion network. These weights will be dynamically adjusted according to the situation in different time periods to ensure the accuracy of the final prediction value.

[0125] In one example, the long-term trend forecast results, periodic load forecast results, and short-term fluctuation forecast results are input into a multi-layer perceptron fusion network. The multi-layer perceptron fusion network contains three hidden layers, each containing 64, 32, and 16 neurons, respectively. The LeakyReLU activation function is used, and the last layer uses a linear activation function to obtain the day-ahead load forecast results, including:

[0126] Perform feature splicing on the long-term trend forecast results, the periodic load forecast results and the short-term fluctuation forecast results to obtain a fused input vector;

[0127] The fused input vector is input into the first hidden layer, which contains 64 neurons. The output of each neuron is processed by the LeakyReLU activation function to obtain the output of the first hidden layer.

[0128] Perform batch normalization on the output of the first hidden layer, and add the batch normalized output to the fusion input vector through a residual connection to obtain the residual output of the first layer;

[0129] The residual output of the first layer is input into the second hidden layer. The second hidden layer contains 32 neurons. The output of each neuron is processed by the LeakyReLU activation function to obtain the output of the second hidden layer.

[0130] Perform a dropout operation on the output of the second hidden layer, and set the dropout rate to 0.3 to obtain the regularized output of the second hidden layer;

[0131] The regularized output of the second hidden layer is input into the third hidden layer. The third hidden layer contains 16 neurons. The output of each neuron is processed by the LeakyReLU activation function to obtain the output of the third hidden layer.

[0132] The output of the third hidden layer is input into the output layer. The output layer contains 24 neurons, corresponding to the load forecast value for the next 24 hours. A linear activation function is used to obtain the day-ahead load forecast result.

[0133] In this example, feature concatenation is performed on the long-term trend forecast results, periodic load forecast results, and short-term fluctuation forecast results, and multiple feature vectors are merged into a fusion vector. The long-term trend forecast results, periodic load forecast results, and short-term fluctuation forecast results represent the load change trend, periodic fluctuation, and random fluctuation information in the next 24 hours, respectively. Each prediction result can be regarded as a vector of length 24, representing the predicted value at each moment in 24 hours. Through feature concatenation, these three vectors are combined into a fusion input vector with a length of 72 (24×3). The fusion input vector is input to the first hidden layer. The first hidden layer contains 64 neurons, each of which performs a linear transformation on the input data and then applies a nonlinear activation function. The activation function used is LeakyReLU (Leaky Rectified LinearUnit), which can maintain a small activation output when the input is negative, thereby avoiding the "neuron death" phenomenon. The formula of LeakyReLU is:

[0134] ;

[0135] in, is the input, is a small constant (usually 0.01) used to control the activation amplitude of the negative part. Through nonlinear activation, the first hidden layer can effectively extract complex features in the input data and obtain 64 outputs. The output of the first hidden layer is batch normalized to make the data have a more stable distribution when passing through the neural network, avoid the problem of gradient disappearance or explosion, accelerate network training and improve the stability of the model. Batch normalization adjusts the data to a standard normal distribution, that is, the mean is 0 and the variance is 1, ensuring that the output of each layer is in a similar numerical range. After batch normalization, the batch normalized output is added to the initial fused input vector through a residual connection to obtain the residual output of the first layer. Residual connection is a technology used in deep neural networks that can solve the problem of gradient disappearance caused by increasing network depth. Residual connection skips one or more hidden layers and directly adds the input to the output to ensure that the network does not rely too much on the learning results of the intermediate layer. The formula is:

[0136] Residual output ;

[0137] in, is the output of the hidden layer, is the input. Through residual connection, the network retains the information of the original input, so that the features will not be lost in the deep network, thereby improving the performance of the model. The residual output of the first layer is input to the second hidden layer. The second hidden layer contains 32 neurons, and the LeakyReLU activation function is used to perform nonlinear processing on the output of each neuron. Compared with the first layer, the number of neurons in the second hidden layer is reduced, with the purpose of gradually compressing the feature space and extracting higher-level feature representations. Through the activation of LeakyReLU, the second hidden layer can process the fused features of the input and extract key patterns and information. In order to prevent the model from overfitting, the dropout operation is applied after the output of the second hidden layer. Dropout is a regularization technique that enhances the generalization ability of the model by randomly discarding a part of the neurons. During the training process, dropout randomly sets the output of some neurons to 0, forcing the network not to rely on certain specific neurons, thereby improving the performance of the model on different data. The dropout dropout rate is set to 0.3, which means that 30% of the neuron outputs will be randomly discarded in each training. This operation can effectively prevent the model from overfitting on the training data and improve its performance on unseen data. The output of the second hidden layer after dropout regularization is input to the third hidden layer. The third hidden layer contains 16 neurons with a relatively simple structure, which is used to compress features and extract the most representative patterns. The output of each neuron is also processed by the LeakyReLU activation function to ensure that the model maintains nonlinear characteristics. Through the step-by-step processing of the three hidden layers, the input data is converted into a higher-level feature representation. The output of the third hidden layer is input to the output layer. The output layer contains 24 neurons, each of which corresponds to the load forecast value for the next 24 hours. Since load forecasting is a continuous regression task, the output layer uses a linear activation function. The linear activation function does not perform nonlinear transformation on the output value and can directly output the original forecast value. In this way, the output layer can generate accurate load forecast results.

[0138] Reference Figure 2 This embodiment provides a solid-state battery management device for a shared charging device with a high number of cycles, including:

[0139] Modeling module 1, used for graph theory modeling of the battery management system in the shared charging device, and constructing a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths;

[0140] Identification module 2, used to perform depth-first search based on the topology model, identify all charging and discharging paths that meet the circuit rules, and calculate the composite impedance value of each charging and discharging path;

[0141] Calculation module 3, used to calculate the dynamic voltage spectrum and impedance characteristics based on the charge and discharge path and the composite impedance value, and obtain the time domain current distribution of each ternary polymer lithium battery unit;

[0142] Processing module 4 is used to calculate the multi-factor coupling loss of each ternary polymer lithium battery unit according to the time domain current distribution to obtain dynamic loss distribution data;

[0143] Evaluation module 5 is used to evaluate the battery status in real time based on dynamic loss distribution data, reconstruct the nonlinear charge and discharge characteristic curve, and perform multi-time scale coordinated scheduling based on the nonlinear charge and discharge characteristic curve to obtain the battery charge and discharge control strategy.

[0144] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0145] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0146] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0147] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0148] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0149] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0150] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation based on the contents of the present invention description and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A ternary polymer lithium battery management method for a high cycle number shared charging device, characterized in that: The following steps are involved: Conduct graph theory modeling of the battery management system in the shared charging device, and construct a topological graph model containing vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths; Performing a depth-first search based on the topology graph model to identify all charge and discharge paths that meet the circuit rules, and calculating the composite impedance value of each charge and discharge path; Calculating a dynamic voltage spectrum and an impedance characteristic based on the charge and discharge path and the composite impedance value to obtain a time domain current distribution of each ternary polymer lithium battery cell; According to the time-domain current distribution, a multi-factor coupling loss calculation is performed on each ternary polymer lithium battery unit to obtain dynamic loss distribution data; Based on the dynamic loss distribution data, the battery state is evaluated in real time, and a nonlinear charge and discharge characteristic curve is reconstructed. Based on the nonlinear charge and discharge characteristic curve, multi-time scale collaborative scheduling is performed to obtain a battery charge and discharge control strategy.

2. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 1, characterized in that: The method of performing graph theory modeling on the battery management system in the shared charging device and constructing a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths includes: Identify the ternary polymer lithium battery unit, charging circuit, discharging circuit and protection circuit of the battery management system in the shared charging device to obtain the type and parameter information of each component; Based on the type and parameter information of each component, vertex modeling is performed on each ternary polymer lithium battery unit and circuit component to obtain a vertex set; According to the circuit topology of the battery management system, the flow path of the current between the components is analyzed to obtain a directed edge set; Assigning attributes to each vertex in the vertex set to obtain vertex attribute data including the rated capacity, internal resistance and current state of charge of the ternary polymer lithium battery unit and the type and parameters of the circuit component, and assigning attributes to each edge in the directed edge set to obtain edge attribute data including the current flow direction, the maximum allowable current and the current current value; Based on the vertex set, the directed edge set and the edge attribute data, a topological graph model of the battery management system is constructed.

3. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 2, characterized in that: The depth-first search is performed based on the topology graph model to identify all charge and discharge paths that meet the circuit rules, and the composite impedance value of each charge and discharge path is calculated, including: Encoding the vertex set in the topological graph model, obtaining a unique identifier and access status flag of each vertex, and constructing a recursive function of depth-first search; Traversing the topology graph model based on the recursive function to obtain all possible vertex access sequences, and performing circuit rule verification on each vertex access sequence to obtain a set of effective charging and discharging paths; Based on the effective charge and discharge path set, the resistance, inductance and capacitance of each charge and discharge path are extracted to obtain a distributed parameter model of the charge and discharge path, and the distributed parameter model of the charge and discharge path is mathematically transformed to obtain a frequency domain composite impedance expression of each charge and discharge path; Based on the frequency domain composite impedance expression, the equivalent impedance of the charge and discharge path within the operating frequency range is calculated to obtain the composite impedance value of each charge and discharge path.

4. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 1, characterized in that: The method of calculating the dynamic voltage spectrum and impedance characteristics based on the charge and discharge path and the composite impedance value to obtain the time domain current distribution of each ternary polymer lithium battery cell includes: Analyzing the topological structure of the charge and discharge path to obtain a time-domain differential equation group describing the dynamic behavior of the battery management system, and converting the time-domain differential equation group into an s-domain transfer function to obtain a frequency-domain mathematical model; Based on the frequency domain mathematical model and the composite impedance value, a system matrix, an input matrix, an output matrix and a feedforward matrix of the battery management system are constructed, and discretized to obtain a discrete time system model; Performing parameter identification on the discrete-time system model to obtain system characteristic parameters, applying a broadband excitation signal to the battery management system based on the system characteristic parameters, and obtaining a dynamic voltage spectrum through fast Fourier transform calculation; Based on the dynamic voltage spectrum and input current information, the impedance characteristics of the battery management system are calculated, and the impedance characteristics are inverse Fourier transformed, and combined with the circuit topology information, the time domain current distribution of each ternary polymer lithium battery unit is obtained.

5. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 1, characterized in that: According to the time domain current distribution, multi-factor coupling loss calculation is performed on each ternary polymer lithium battery unit to obtain dynamic loss distribution data, including: Based on the time domain current distribution, the current square integral operation is performed on each ternary polymer lithium battery unit to obtain the initial value of Joule heat loss, and the temperature correction is performed on the initial value of Joule heat loss according to the real-time updated battery internal resistance parameter to obtain Joule heat loss data; Calculating the charge transfer impedance based on the charge and discharge state of the ternary polymer lithium battery unit to obtain electrochemical reaction loss basic data, and performing nonlinear compensation on the electrochemical reaction loss basic data to obtain electrochemical reaction loss data; Based on the battery cycle number and deep discharge state, a battery capacity attenuation model is constructed to obtain the cycle attenuation loss data during a single charge and discharge process; Based on a multi-physics field coupling model, the Joule heat loss data, the electrochemical reaction loss data, and the cycle attenuation loss data are solved simultaneously to obtain coupling loss data; The coupling loss data is subjected to a time series analysis to obtain a time-varying characteristic of a dynamic loss distribution, and based on the time-varying characteristic of the dynamic loss distribution, the Joule heat loss data, the electrochemical reaction loss data and the cycle attenuation loss data are weightedly fused to obtain dynamic loss distribution data.

6. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 1, characterized in that: The real-time evaluation of the battery state based on the dynamic loss distribution data, and reconstruction of the nonlinear charge and discharge characteristic curve, and multi-time scale coordinated scheduling based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy include: Performing piecewise linearization processing on the time-domain current distribution to obtain a discrete current sampling sequence, and calculating a battery state of charge change based on the discrete current sampling sequence to obtain an initial SOC estimation value; According to the dynamic loss distribution data, the initial SOC estimation value is compensated and corrected to obtain a corrected SOC value; An adaptive OCV-SOC mapping model is constructed based on historical charge and discharge data, and online parameter identification is performed on the adaptive OCV-SOC mapping model, and the model parameters are updated using a forgetting factor method to obtain an updated OCV-SOC mapping relationship; Based on the corrected SOC value and the updated OCV-SOC mapping relationship, calculating the instantaneous open circuit voltage of the battery to obtain a SOC-OCV data pair; Performing curve fitting on the SOC-OCV data pair to obtain a nonlinear characteristic equation of the charge and discharge process, and calculating the battery capacity and internal resistance at different SOCs based on the nonlinear characteristic equation to obtain a nonlinear charge and discharge characteristic curve; Multi-time scale coordinated scheduling is performed based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy.

7. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 6, characterized in that: The multi-time scale coordinated scheduling is performed based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy, including: Based on the nonlinear charge-discharge characteristic curve, determining the charge-discharge efficiency and capacity limit at different states of charge; According to the charging and discharging efficiency and the capacity limit, combined with historical electricity consumption data, a day-ahead load forecast is performed to obtain a day-ahead load forecast result; Based on the day-ahead load forecast result, a day-ahead charge-discharge preliminary scheduling plan is formulated, and the day-ahead charge-discharge preliminary scheduling plan is optimized on a rolling basis to obtain a daily dynamic adjustment strategy; According to the real-time load change, the intraday dynamic adjustment strategy is fine-tuned to generate real-time control instructions, and based on the real-time control instructions, the battery charging and discharging operations are executed, and the charging and discharging process data are recorded; Based on the charging and discharging process data, a battery state evaluation model is updated, and the nonlinear charging and discharging characteristic curve is optimized to obtain an updated nonlinear charging and discharging characteristic curve; Multi-time scale coordinated scheduling adjustment is performed according to the updated nonlinear charging and discharging characteristic curve to obtain the battery charging and discharging control strategy.

8. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 7, characterized in that: The step of performing a day-ahead load forecast based on the charge-discharge efficiency and the capacity limit in combination with historical power consumption data to obtain a day-ahead load forecast result includes: Perform time series decomposition on historical electricity consumption data to obtain trend items, seasonal items, and random items; The trend item is input into a long-term trend prediction model, which adopts an LSTM network structure, including an input layer, two hidden layers and an output layer, each hidden layer includes 128 neurons, uses a tanh activation function, and predicts the load trend in the next 24 hours through the long-term trend prediction model to obtain a long-term trend prediction result; The seasonal term is input into a periodic pattern recognition model, which adopts a one-dimensional convolutional neural network structure, including 3 convolutional layers and 2 fully connected layers, each convolutional layer uses a ReLU activation function and batch normalization, and the periodic pattern recognition model is applied to the next 24 hours to obtain a periodic load forecast result; The random item is input into a short-term fluctuation prediction model, which adopts a gated recurrent unit network structure, including a bidirectional GRU layer and a fully connected layer, and the GRU layer includes 64 hidden units. The load fluctuation in the next 24 hours is predicted by the short-term fluctuation prediction model to obtain a short-term fluctuation prediction result; The long-term trend forecast result, the periodic load forecast result and the short-term fluctuation forecast result are input into a multi-layer perceptron fusion network, wherein the multi-layer perceptron fusion network comprises three hidden layers, each layer comprises 64, 32 and 16 neurons respectively, and uses a LeakyReLU activation function, and the last layer uses a linear activation function to obtain the day-ahead load forecast result.

9. The ternary polymer lithium battery management method for a shared charging device with a high number of cycles according to claim 8, characterized in that: The long-term trend prediction result, the periodic load prediction result and the short-term fluctuation prediction result are input into a multi-layer perceptron fusion network, wherein the multi-layer perceptron fusion network comprises three hidden layers, each layer comprises 64, 32 and 16 neurons respectively, uses a LeakyReLU activation function, and the last layer uses a linear activation function to obtain a day-ahead load prediction result, including: Performing feature splicing on the long-term trend prediction result, the periodic load prediction result and the short-term fluctuation prediction result to obtain a fusion input vector; Input the fused input vector into a first hidden layer, where the first hidden layer includes 64 neurons, and the output of each neuron is processed by a LeakyReLU activation function to obtain the output of the first hidden layer; Performing a batch normalization operation on the first hidden layer output, and adding the batch normalized output to the fused input vector through a residual connection to obtain a first layer residual output; The residual output of the first layer is input into the second hidden layer, where the second hidden layer includes 32 neurons, and the output of each neuron is processed by the LeakyReLU activation function to obtain the output of the second hidden layer; Perform a dropout operation on the second hidden layer output, with the dropout rate set to 0.3, to obtain the regularized second hidden layer output; Input the regularized second hidden layer output into the third hidden layer, wherein the third hidden layer comprises 16 neurons, and the output of each neuron is processed by the LeakyReLU activation function to obtain the third hidden layer output; The output of the third hidden layer is input into the output layer, and the output layer includes 24 neurons corresponding to the load forecast value of the next 24 hours. A linear activation function is used to obtain the day-ahead load forecast result.

10. A solid-state battery management device for a high-cycle shared charging device, characterized in that: For implementing the steps of the method according to any one of claims 1 to 9, the device comprises: A modeling module, used for graph-theoretic modeling of the battery management system in the shared charging device, and constructing a topological graph model including vertices representing ternary polymer lithium battery cells and management circuits and directed edges representing current flow paths; An identification module, used to perform a depth-first search based on the topological graph model, identify all charge and discharge paths that meet the circuit rules, and calculate the composite impedance value of each charge and discharge path; A calculation module, used to calculate the dynamic voltage spectrum and impedance characteristics based on the charge and discharge path and the composite impedance value, and obtain the time domain current distribution of each ternary polymer lithium battery unit; A processing module, used to perform multi-factor coupling loss calculation on each ternary polymer lithium battery unit according to the time domain current distribution to obtain dynamic loss distribution data; An evaluation module is used to perform real-time battery status evaluation based on the dynamic loss distribution data, reconstruct a nonlinear charge and discharge characteristic curve, and perform multi-time scale collaborative scheduling based on the nonlinear charge and discharge characteristic curve to obtain a battery charge and discharge control strategy.

Images