A Method and System for Estimating the State of Charge (SOC) of a Lithium Battery through Edge-Cloud Collaborative Deep Learning

Abstract

The present invention discloses a method and system for deep learning estimation of lithium battery SOC through end-cloud collaboration, which relates to the technical field of battery management and control. It includes: obtaining battery charge cycle process data from the cloud, correcting the SOC using the inverse ampere-hour integration method, and generating self-supervised SOC estimation training data through automatic tagging; constructing a CGRM model, combining the feature extraction ability of the CNN learner and the long-term dependence capture ability of the GRU learner, and realizing accurate cloud estimation of the SOC through the integration output of the meta-learner; using the sliding window method to generate the feature time series for SOC estimation and inputting it into the CGRM model to obtain the SOC estimation result; transmitting the SOC data estimated by the cloud to the battery management system at the user end to calibrate the SOC value estimated by it, and realizing accurate estimation of the battery SOC at the user end. The present invention does not require a large number of experimental data samples, reduces the data acquisition cost through the end-cloud collaboration method, solves the problem of mismatch between the experimental environment and the actual working conditions, and significantly improves the SOC estimation accuracy and the battery operation safety.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management and control, and more particularly, to a method and system for deep learning estimation of the state of charge (SOC) of a lithium battery based on automatic sample generation for end-cloud collaboration. Background Art

[0002] Lithium-ion batteries are widely used in different fields such as energy storage systems, battery power supplies, and electric vehicles. To ensure their safe use, full utilization, and intelligent operation and maintenance, it is necessary to accurately estimate the state of charge (SOC) of the battery. For this purpose, scholars at home and abroad have carried out a large number of studies on SOC estimation methods, mainly including the ampere-hour integration method, the open-circuit voltage method, model-based methods, and data-driven methods. Among them, the ampere-hour integration method is simple and easy to implement, but long-term use will produce cumulative errors and requires an accurate initial SOC. The open-circuit voltage method requires the battery to be in a long-term static state, which cannot be satisfied in actual application conditions. Model-based methods estimate the SOC by establishing an equivalent circuit model or an electrochemical mechanism model of the battery and then applying algorithms such as the extended Kalman filter. However, due to the complex internal mechanism of lithium-ion batteries and the influence of factors such as aging on the modeling accuracy, as well as measurement noise and data drift in the real environment, it is also difficult to identify the model parameters. Therefore, model-based methods are also difficult to meet the actual application requirements.

[0003] With the development of machine learning and artificial intelligence, data-driven methods have developed rapidly. By using measurement parameters such as the current, voltage, and temperature of the battery, the SOC value of the battery can be directly estimated using a machine learning model, which not only does not need to consider the internal mechanism of the battery but also avoids the problem of establishing a battery model in model-based methods. Among them, the data sample is the first problem that must be solved. However, the laboratory cannot fully simulate the operating environment of the real-world battery system. Most existing SOC estimations in previous studies are carried out in a stable experimental environment, and the application effects of most algorithms cannot be guaranteed in the real world. At the same time, the SOC of the battery management system (BMS) at the user end cannot be used as a reliable sample for model training, which restricts the performance of machine learning methods for estimating the SOC of the battery. How to accurately estimate the state of charge (SOC) of the battery remains an important problem that urgently needs to be solved in the energy storage, battery power supply, and automotive industries.

[0004] Patent document CN118858958A discloses a method for estimating the state of charge (SOC) of a lithium-ion battery based on an improved Transformer. By extracting the current, voltage, and temperature curves of the battery data set under different operating conditions and performing outlier judgment, the optimal current, voltage, and temperature curves are obtained. Then, the SOC calculated by the ampere-hour integration method is used as the label value, and a bidirectional long short-term memory network introducing a gated linear unit is introduced into the Transformer structure. The optimal current, voltage, temperature curves, state of charge, and the average value of current and voltage are input to train the Transformer for predicting the state of charge. This method can make the SOC estimation output smoother and more stable. However, this method still trains the model based on the data sets obtained from tests under different operating conditions, so it cannot adapt to the complexity and variability of the operating conditions of the real-world battery system, resulting in limited actual application effects.

[0005] Patent document CN118839114A discloses a method for predicting the SOC of a power battery of an electric mining truck based on digital twin. By using the battery Thevenin equivalent circuit model and the vehicle whole vehicle Simulink model, a digital twin model of the power battery pack of the electric mining truck is established to simulate the operating state of the power battery system. Then, a deep residual shrinkage network is used to learn and train the simulation data to achieve the prediction of the battery SOC. This method can dynamically adjust and optimize the training model according to the real-time acquired data and external conditions. However, in essence, it still depends on the battery Thevenin equivalent circuit model. Therefore, limited by the accuracy of this type of model, the accuracy of the SOC estimation of this method for the battery is limited. Summary of the Invention

[0006] The present invention aims to provide a method and system for end-cloud collaborative deep learning estimation of the SOC of a lithium battery based on automatic sample generation, aiming to automatically generate data samples for battery SOC estimation by mining real-world battery operation data, solve the problems of high cost and long cycle for obtaining sample data, especially solve the problem of difficult to comprehensively simulate the working environment conditions of real-world batteries under laboratory conditions, and at the same time propose a convolutional gated recurrent unit meta-learning model to achieve accurate end-cloud collaborative deep learning estimation of the battery SOC.

[0007] To solve the above problems, the technical solutions adopted by the present invention are as follows:

[0008] In a first aspect, the present invention provides a method for end-cloud collaborative deep learning estimation of the SOC of a lithium battery, including the following steps:

[0009] Step 1, perform data cleaning on the real-world user-side lithium battery operation data collected by the cloud to obtain effective charging cycle process data, including battery charging voltage, current, and temperature data;

[0010] Step 2: Based on the effective charge cycle process data, use the reverse ampere-hour integration method to correct the SOC in the cloud, and generate self-supervised SOC estimation training data by automatically tagging the SOC.

[0011] Step 3: Construct an SOC estimation model based on a convolutional gated recurrent unit meta-learning network.

[0012] Step 4: Use the SOC estimation training data, adopt the sliding window method to generate the characteristic time series required for SOC estimation, and input the time series and its corresponding automatically tagged SOC value into the SOC estimation model to obtain the cloud-estimated SOC data.

[0013] Step 5: Transmit the cloud-estimated SOC data to the user-side lithium battery management system to calibrate the battery SOC estimated by the user-side lithium battery management system, so as to accurately estimate the lithium battery SOC at the user side.

[0014] As a further description of the above technical solution, the Step 2 includes:

[0015] Step 2.1: Automatically search for fully charged single cells that meet the cut-off voltage in the effective charge cycle process data, and set the SOC at the fully charged moment to 100%.

[0016] Step 2.2: Calculate the SOC values at other moments in the charge cycle of the searched fully charged single cell by the reverse ampere-hour integration method to achieve automatic tagging of the SOC. After the automatic tagging is completed, the SOC estimation training data is obtained.

[0017] The SOC value calculation formula is:

[0018]

[0019] where SOC r (k) is the SOC value corrected by the reverse ampere-hour integration method at the k time step, P is the total number of steps in the charge cycle, I is the current, ΔT represents the sampling interval time, C is the nominal capacity of the battery, SOH is the health state of the battery, and η is the Coulomb efficiency.

[0020] As a further description of the above technical solution, the SOC estimation model includes a convolutional neural network independent learner, a gated regression unit independent learner, and a perceptron meta-learner; the convolutional neural network independent learner is used to extract the battery operation time series data features from the input data; the gated regression unit independent learner is used to automatically capture the long-term dependence relationship between the battery operation data and the SOC from the input data; the perceptron meta-learner is used to obtain the cloud-estimated SOC data based on the battery operation time series data features and the long-term dependence relationship between the battery operation data and the SOC.

[0021] As a further description of the above technical solution, the convolutional neural network independent learner uses a one-dimensional CNN layer to perform a convolution operation on the input data to extract first feature data, then uses a pooling layer to filter out second feature data from the first feature data, and then uses a convolutional layer to extract third feature data from the second feature data; finally, the third feature data is subjected to a non-linear transformation and dimensionality reduction through multiple fully connected layers, and the battery operation time series data features are output.

[0022] As a further description of the above technical solution, the gated regression unit independent learner includes a number of GRU layers and a number of fully connected layers.

[0023] As a further description of the above technical solution, the perceptron meta-learner includes multiple fully connected layers.

[0024] As a further description of the above technical solution, step 4 includes:

[0025] Step 4.1, sample the time series of current, voltage, temperature, and cumulative ampere-hours within the sliding time window during the charging process of a fully charged single battery, where the cumulative ampere-hours is calculated according to the following formula:

[0026]

[0027] where AC(k) is the ampere-hour amount during the charging process from the charging time t1 to the k-th time step, and the charging time t1 is the time after a period of charging start.

[0028] Step 4.2, perform standardization processing on the time series of current, voltage, temperature, and cumulative ampere-hours using the Z-Score method to obtain a standardized time series;

[0029] Step 4.3, based on the SOC estimation model, predict the SOC value at the last time step of the standardized time series and use it as the SOC label of the standardized time series:

[0030]

[0031] where SOC(k) represents the SOC estimated by the SOC estimation model at the k-th time step, U(k-m+1, k-m+2, …, k), I(k-m+1, k-m+2, …, k), T(k-m+1, k-m+2, …, k), and AC(k-m+1, k-m+2, ..., k) respectively represent the time series of voltage, current, temperature, and ampere-hour amount during the charging process from the (k-m+1)-th to the k-th step, m represents the length of the sliding window, the data is sampled through a window with a fixed step size sliding along the time direction, and the SOC corresponding to the last moment of the sample is the output label of the model, and N is the length of the effective data;

[0032] Step 4.4, calculate the SOC during the battery discharge process, and the calculation formula is:

[0033]

[0034] where SOC(k + p) is the battery SOC at the p-th time step during the discharge process.

[0035] As a further description of the above technical solution, the said step 5 includes:

[0036] Step 5.1, transmit the cloud-estimated SOC data to the user-side battery management system through remote communication;

[0037] Step 5.2, the user-side lithium battery management system calibrates its estimated battery SOC by using the cloud-estimated SOC data;

[0038] Step 5.3, the user-side lithium battery management system calculates its SOC by using the ampere-hour integration method. Meanwhile, it also automatically calibrates its SOC when the full charge or full discharge condition is satisfied.

[0039] In the second aspect, the present invention provides a lithium battery SOC end-cloud collaborative deep learning estimation system. When the system works, the cloud is connected to the user-side battery management system through wireless communication. The user-side battery management system sends battery voltage, current, and temperature data to the cloud, and executes the lithium battery SOC end-cloud collaborative deep learning estimation method described in the first aspect to achieve accurate estimation of the battery SOC.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] 1) The method of the present invention realizes the automatic labeling of SOC by automatically mining real-world battery operation data, without the need for a large number of dedicated battery experiments to generate data samples, and can effectively solve the problems of high cost and long cycle for obtaining sample data, especially the problem that it is difficult to comprehensively simulate the working environment conditions of real-world batteries under laboratory conditions.

[0042] 2) The present invention introduces a convolutional gated recurrent unit meta-learning (CGRM) model, which can not only automatically and fully extract the features of battery operation time-series data through a multi-layer CNN learner, but also automatically capture the long-term dependence relationship between battery operation data and SOC through a multi-layer GRU learner. Finally, the meta-learner integrates the outputs of the two learners to achieve accurate estimation of the SOC of lithium-ion batteries in the cloud.

[0043] 3) The present invention proposes a new feature description method for characterizing battery SOC. By introducing the cumulative power feature on the basis of traditional features based on current, voltage, and temperature, it can effectively improve the accuracy of SOC estimation.

[0044] 4) Through the cooperation between the terminal and the cloud, the present invention can not only accurately estimate the battery SOC in the cloud, but also accurately estimate the battery SOC at the user terminal, effectively supporting the intelligent operation and maintenance of the battery in the cloud, and at the same time ensuring the high-efficiency and safety of the battery at the user's local end.

[0045] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following specifically gives embodiments of the present invention and, in conjunction with the accompanying drawings, provides a detailed description as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0047] Figure 1 is the implementation flowchart of the method described in the embodiments of the present invention.

[0048] Figure 2 is the framework diagram of the CGRM model in the embodiments of the present invention and a schematic diagram of its main components.

[0049] Figure 3 is the result diagram of the SOC estimation error distribution in the embodiments of the present invention.

[0050] Figure 4 is the comparison diagram of the SOC estimation error indexes between the embodiments of the present invention and three existing methods. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will, in conjunction with the drawings in the embodiments of the present invention, clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention.

[0052] Refer to Figure 1 As shown, the embodiments of the present invention provide a method for estimating the SOC of a lithium battery through deep learning with the cooperation between the terminal and the cloud, including the following steps:

[0053] Step 1: Clean the operation data of the lithium battery at the user terminal in the real world collected in the cloud to obtain effective charging cycle process data, including battery charging voltage, current, and temperature data.

[0054] Step 2: Search for fully charged cells in the cloud, and use the reverse ampere-hour integration method to correct the SOC in the cloud based on the effective charging cycle process data to obtain self-supervised SOC estimation training data, and realize automatic SOC labeling of battery charging process data, as follows:

[0055] Step 2.1, automatically searching for a fully charged single battery that meets the cut-off voltage in the effective charging cycle process data, and setting the SOC of the fully charged single battery to 100%;

[0056] Step 2.2, calculate the SOC value at other times in the charging cycle of the searched fully charged single battery by the reverse ampere-hour integration method, realize automatic labeling of SOC, and obtain SOC estimation training data after the automatic labeling is completed;

[0057] The SOC value calculation formula is:

[0058]

[0059] Among them, SOC r (k) is the SOC value corrected by the reverse ampere-hour integration method at k time steps, P is the total number of charging cycle steps, I is the current (positive for charging and negative for discharging), ΔT represents the sampling interval, C is the nominal capacity of the battery, SOH is the health state of the battery, and η is the coulombic efficiency. Since the current of the battery charging process is stable and the SOC at the full charge point is accurate, the cumulative error of the SOC calculated by the ampere-hour integration method in the current charging cycle can be ignored, and the corrected SOC can be used as the real SOC of the charging process, thereby realizing the automatic labeling of the SOC of the charging process data.

[0060] Step 3: Construct a SOC estimation model based on a convolutional gated recurrent unit meta-learning network (CGRM).

[0061] like Figure 2 As shown, the SOC estimation model includes a convolutional neural network independent learner, a gated regression unit (GRU) independent learner, and a perceptron meta-learner.

[0062] The convolutional neural network independent learner is used to extract battery operation sequence data features from the input data. The convolutional neural network independent learner uses a one-dimensional CNN layer to perform convolution operations on the input data to extract features from the input data layer; then uses a pooling layer to filter out important features; then uses a convolution layer to further extract important features; finally, through two layers of fully connected layers, further nonlinear transformation and dimensionality reduction are performed to output the final SOC prediction result of the learner, that is, the battery operation sequence data features.

[0063] The GRU independent learner is used to automatically capture the long-term dependencies between battery operating data and SOC from input data. The GRU independent learner includes two GRU layers and two fully connected neural networks. Among them, the latter GRU layer uses the prediction results of the former GRU layer to further predict trends, so as to capture more complex and deep time-domain features.

[0064] The perceptron meta-learner is used to obtain the cloud-estimated SOC data based on the characteristics of battery operating time-series data and the long-term dependencies between battery operating data and SOC. The perceptron meta-learner is composed of two fully connected neural networks, which realizes the integration of the outputs of the two independent learners.

[0065] Step 4: Using the SOC estimation training data, adopt the sliding window method to generate the characteristic time series required for SOC estimation, and input the time series and its corresponding automatic label SOC value into the SOC estimation model to obtain the cloud-estimated SOC data, specifically as follows:

[0066] Step 4.1: Sample the time series of current, voltage, temperature, and cumulative ampere-hours within the sliding time window during the charging process of a fully charged single battery. Among them, the cumulative ampere-hours is calculated by the following formula:

[0067]

[0068] where AC(k) is the ampere-hour amount during the charging process from the charging time t1 to the k-th time step. To eliminate the influence of the SOC difference at the start of charging and the battery polarization problem and improve the regularity of the features, the charging time t1 is defined as the time after a certain period (such as 10 minutes) from the start of charging.

[0069] Step 4.2: Perform standardization processing on the time series of current, voltage, temperature, and cumulative ampere-hours using the Z-Score method to obtain the standardized time series.

[0070] Step 4.3: Based on the SOC estimation model, predict the SOC value at the last time step of the standardized time series and use it as the SOC label of the standardized time series:

[0071]

[0072] Among them, SOC(k) represents the SOC estimated by the SOC estimation model at the k-th time step. U(k - m + 1, k - m + 2, …, k), I(k - m + 1, k - m + 2, …, k), T(k - m + 1, k - m + 2, …, k), and AC(k - m + 1, k - m + 2, …, k) respectively represent the time series of voltage, current, temperature, and ampere-hour during charging from the (k - m + 1)-th step to the k-th step. m = 90 represents the length of the sliding window. Data is sampled through a window with a fixed step size sliding along the time direction. The SOC corresponding to the last moment of the sample is the output label of the model, and N is the length of the valid data;

[0073] Step 4.4, during the battery discharging process, the SOC starts from the SOC estimated during the charging process, and the SOC during the battery discharging process is calculated using the ampere-hour integration method. The calculation formula is:

[0074]

[0075] Among them, SOC(k + p) is the battery SOC at the p-th time step during the discharging process, and SOC(k) is the SOC during the charging process estimated by the CGRM.

[0076] Step 5, transmit the cloud-estimated SOC data to the user-side lithium battery management system, and calibrate the battery SOC estimated by the user-side lithium battery management system to achieve accurate estimation of the lithium battery SOC at the user side, specifically as follows:

[0077] Step 5.1, use remote communication to transmit the cloud-estimated SOC to the user-side battery management system BMS;

[0078] Step 5.2, the user-side BMS calibrates the battery SOC estimated by it using the cloud-estimated SOC;

[0079] Step 5.3, the user-side BMS calculates its SOC using the ampere-hour integration method. At the same time, it also automatically calibrates its SOC when the full charge or full discharge condition is satisfied.

[0080] The present invention realizes the end-cloud collaborative deep learning estimation of the lithium battery SOC with automatic sample generation. This method shows excellent lithium-ion battery SOC effects in tests. The estimated SOC is highly consistent with the true SOC in different situations, as shown in the appendix. Figure 3 Further tests show that the lithium-ion battery SOC estimation results of the method and system of the present invention are significantly better than other current state-of-the-art methods (for example, the method using a convolutional neural network (CNN), the method using a gated recurrent unit (GRU), and even the method using a cascaded network of CNN and GRU (CNN-GRU)). The detailed performance comparison is shown in the appendix. Figure 4。This embodiment utilizes deep learning and has the characteristics of simultaneously achieving accurate estimation of the state of charge (SOC) of lithium-ion batteries in the cloud and at the user side for the real world, without the need for dedicated experiments on a large number of batteries to generate data samples, effectively solving the problems of high cost and long cycle for obtaining sample data, especially solving the problem of difficult comprehensive simulation of the working environment conditions of batteries in the real world under laboratory conditions.

[0081] Given the crucial role of battery SOC estimation in energy storage systems, battery power supplies, electric vehicles, etc., the present invention has important practical value, especially applicable to ensuring the use safety and improving the energy utilization rate of lithium-ion battery systems, supporting the safe, efficient application and intelligent operation and maintenance of lithium-ion batteries.

[0082] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for estimating the state of charge (SOC) of a lithium battery through end-cloud collaborative deep learning, characterized in that, It includes the following steps: Step 1: Clean the real-world user-side lithium battery operation data collected in the cloud to obtain effective charging cycle process data, including battery charging voltage, current, and temperature data; Step 2: According to the effective charging cycle process data, correct the SOC using the reverse ampere-hour integration method in the cloud, and generate self-supervised SOC estimation training data by automatically tagging the SOC; Step 3: Construct an SOC estimation model based on a convolutional gated recurrent unit meta-learning network; Step 4: Use the SOC estimation training data, adopt the sliding window method to generate the characteristic time series required for SOC estimation, and input the time series and its corresponding automatically tagged SOC value into the SOC estimation model to obtain the cloud-estimated SOC data; Step 5: Transmit the cloud-estimated SOC data to the user-side lithium battery management system to calibrate the battery SOC estimated by the user-side lithium battery management system, realizing the accurate estimation of the lithium battery SOC at the user side; The said Step 2 includes: Step 2.1: Automatically search for fully charged single cells that meet the cut-off voltage in the effective charging cycle process data, and set the SOC at the fully charged moment to 100%; Step 2.2: Calculate the SOC values at other moments in the charging cycle of the searched fully charged single cells through the reverse ampere-hour integration method to achieve automatic tagging of the SOC. After the automatic tagging is completed, the SOC estimation training data is obtained; The formula for calculating the SOC value is: ； Among them, is the SOC value corrected by the reverse ampere-hour integration method at the k time step, P is the total number of steps in the charging cycle, I is the current, represents the sampling interval time, C is the nominal capacity of the battery, SOH is the health state of the battery, and η is the Coulomb efficiency; The said SOC estimation model includes a convolutional neural network independent learner, a gated regression unit independent learner, and a perceptron meta-learner; the convolutional neural network independent learner is used to extract battery operation time series data features from the input data; the gated regression unit independent learner is used to automatically capture the long-term dependence relationship between the battery operation data and the SOC from the input data; the perceptron meta-learner is used to obtain the cloud-estimated SOC data based on the battery operation time series data features and the long-term dependence relationship between the battery operation data and the SOC; The said Step 4 includes: Step 4.1: Sample the time series of current, voltage, temperature, and cumulative ampere-hour in the sliding time window during the charging process of the fully charged single cell, where the cumulative ampere-hour is calculated according to the following formula: ； Among them, is the ampere-hour amount of the charging process from the charging time t1 to the k-th time step. The charging time t1 is the time after a period of time since the start of charging; Step 4.2: Perform standardization processing on the time series of current, voltage, temperature, and cumulative ampere-hour using the Z-Score method to obtain the standardized time series; Step 4.3: Based on the said SOC estimation model, predict the SOC value at the last time step of the standardized time series and use it as the SOC label of the standardized time series; ； Among them, represents the SOC estimated by the SOC estimation model at the k-th time step, , , and respectively represent the time series of voltage, current, temperature, and ampere-hour during charging from the -th step. m represents the length of the sliding window. The data is sampled through a window with a fixed step size sliding along the time direction. The SOC corresponding to the last moment of the sample is the output label of the model, and N is the length of the valid data; Step 4.4: Calculate the SOC during the battery discharge process, and the formula is: ； Among them, is the battery SOC at the p-th time step of the discharge process.

2. The method for estimating the state of charge (SOC) of a lithium battery through end-cloud collaborative deep learning according to claim 1, wherein, The said convolutional neural network independent learner uses a one-dimensional CNN layer to perform convolutional operations on the input data to extract the first feature data, then uses a pooling layer to filter out the second feature data from the first feature data, and then uses a convolutional layer to extract the third feature data from the second feature data; finally, perform non-linear transformation and dimension reduction on the third feature data through multiple fully connected layers, and output to obtain the battery operation time series data features.

3. The method for estimating the state of charge (SOC) of a lithium battery through collaborative deep learning between the device and the cloud according to claim 1, wherein, The gated regression unit independent learner includes a number of GRU layers and a number of fully connected layers.

4. The method for estimating the state of charge (SOC) of a lithium battery through cloud-edge collaborative deep learning according to claim 1, wherein The perceptron meta-learner includes multiple fully connected layers.

5. The method for estimating the state of charge (SOC) of a lithium battery through cloud-edge collaborative deep learning according to claim 1, characterized in that Step 5 includes: Step 5.1, transmitting the cloud-estimated SOC data to the user-side battery management system through telecommunication; Step 5.2, the user-side lithium battery management system calibrates its estimated battery SOC using the cloud-estimated SOC data; Step 5.3, the user-side lithium battery management system calculates its SOC using the ampere-hour integration method, and at the same time, automatically calibrates its SOC when the full charge or full discharge condition is satisfied.

6. A cloud-edge collaborative deep learning estimation system for the state of charge (SOC) of a lithium battery, characterized in that, When the system is working, the cloud is connected to the user-side battery management system through wireless communication. The user-side battery management system sends battery voltage, current, and temperature data to the cloud, and executes the lithium battery SOC cloud-edge collaborative deep learning estimation method according to any one of claims 1 to 5, so as to accurately estimate the battery SOC.

Images