Battery capacity estimation method, train control method, device, and medium

Abstract

The application provides a battery capacity estimation method, a train control method, a device and a medium, which can be applied to the technical field of battery health management. The method comprises the following steps: obtaining a target voltage sequence related to an energy storage battery and a corresponding capacity response sequence. The target voltage sequence comprises terminal voltages corresponding to a plurality of charging time points in a charging process of the energy storage battery. The capacity response data in the capacity response sequence represent the change of the power of the energy storage battery relative to the terminal voltages in the target voltage sequence. The target voltage sequence and the capacity response sequence are processed by using a capacity estimation model to obtain a capacity estimation result. The capacity estimation result represents the power that can be stored by the energy storage battery after the charging is completed.

Description

Technical Field

[0001] This application relates to the field of battery health management technology, and more specifically to a battery capacity estimation method, train control method, device and medium. Background Technology

[0002] Battery capacity estimation methods are related to battery aging mechanisms, aging paths, and charging state of charge intervals. Due to the coupling problem between open-circuit voltage and internal resistance in the battery model, the influence of open-circuit voltage and internal resistance aging on battery capacity cannot be eliminated, resulting in low accuracy of battery capacity estimation. Summary of the Invention

[0003] In view of the above problems, this application provides a battery capacity estimation method, train control method, device and medium.

[0004] According to a first aspect of this application, a battery capacity estimation method is provided, comprising: acquiring a target voltage sequence and a corresponding capacity response sequence related to an energy storage battery, wherein the target voltage sequence includes the terminal voltages of the energy storage battery at multiple charging moments during charging, and the capacity response data in the capacity response sequence characterizes the change in the energy storage battery's charge capacity relative to the terminal voltages in the target voltage sequence; processing the target voltage sequence and the capacity response sequence using a capacity estimation model to obtain a capacity estimation result, wherein the capacity estimation result characterizes the charge capacity that the energy storage battery can store after charging is completed.

[0005] According to an embodiment of this application, the target voltage sequence is obtained based on the following operations: acquiring the voltage-capacity curve of the energy storage battery charging process; dividing the voltage-capacity curve based on a preset capacity interval window to obtain multiple voltage-capacity curve segments; determining multiple terminal voltages in the target voltage sequence based on the acquisition time attributes of each terminal voltage in the voltage-capacity curve segments; and determining the capacity response data corresponding to the terminal voltage based on the change slope of the voltage-capacity curve segments corresponding to the terminal voltages. The capacity response sequence includes multiple capacity response data.

[0006] According to an embodiment of this application, dividing the voltage-capacity curve based on a preset capacity range window includes: deleting a specified voltage-capacity curve segment from the voltage-capacity curve to obtain a processed voltage-capacity curve, wherein the specified voltage-capacity curve segment corresponds to a specified capacity range, and the specified capacity range is the capacity range between a specified capacity value and the upper limit capacity value of the energy storage battery; and dividing the processed voltage-capacity curve based on the preset capacity range window.

[0007] According to an embodiment of this application, a capacity estimation model is used to process the target voltage sequence and the capacity response sequence to obtain a capacity estimation result, including: determining voltage-capacity input features based on the target voltage sequence and the capacity response sequence, wherein the voltage-capacity input features characterize the terminal voltage at the charging moment and the change of the terminal voltage; processing the voltage-capacity input features using an activation function to obtain hidden features, wherein the hidden features characterize the interaction between the capacity of the energy storage battery and the terminal voltage; and processing the hidden features to obtain a capacity estimation result.

[0008] According to an embodiment of this application, the capacity estimation model is trained based on the following steps: obtaining sample voltage sequences, sample capacity response sequences, and tag data corresponding to multiple sets of energy storage batteries, wherein the tag data includes the actual capacity corresponding to the sample capacity response sequence; inputting the sample voltage sequences and sample capacity response sequences into the initial capacity estimation model and outputting the sample capacity estimation results; training the initial capacity estimation model based on the sample capacity estimation results and tag data to obtain the trained capacity estimation model.

[0009] According to an embodiment of this application, training an initial capacity estimation model based on sample capacity estimation results and label data to obtain a trained capacity estimation model includes: processing the sample capacity estimation results and label data according to a loss function to obtain a loss value; and training the initial capacity estimation model based on the loss value to obtain a trained capacity estimation model.

[0010] The second aspect of this application provides a train control method, comprising: determining the capacity estimation result of an energy storage battery according to the above-mentioned battery capacity estimation method; and sending an alarm message to the train control center if the capacity estimation result does not meet preset conditions.

[0011] A third aspect of this application provides a train control device, comprising: a capacity determination module, used to determine the capacity estimation result of the energy storage battery in the train power system according to the above-mentioned battery capacity estimation method; and a transmission module, used to send an alarm message to the train control center when the capacity estimation result does not meet preset conditions.

[0012] A fourth aspect of this application provides a battery capacity estimation device, comprising: an acquisition module for acquiring a target voltage sequence and a corresponding capacity response sequence related to an energy storage battery, wherein the target voltage sequence includes the terminal voltages of the energy storage battery at multiple charging moments during charging, and the capacity response sequence characterizes the change in the energy storage battery's charge capacity relative to the terminal voltages in the target voltage sequence; and a capacity estimation result module for processing the target voltage sequence and the capacity response sequence using a capacity estimation model to obtain a capacity estimation result, wherein the capacity estimation result characterizes the charge capacity that the energy storage battery can store after charging is completed.

[0013] The fifth aspect of this application provides a train, including: a battery capacity estimation device applied according to the battery capacity estimation method described above.

[0014] A sixth aspect of this application provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the method described above.

[0015] A seventh aspect of this application also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the methods described above.

[0016] The eighth aspect of this application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0017] According to embodiments of this application, by concatenating the target voltage sequence with the corresponding capacity response sequence, and using the fused features as input to the capacity estimation model, the voltage difference characteristics in the capacity response sequence are utilized to address the interference of nonlinear changes in open-circuit voltage and internal resistance and measurement errors on the capacity estimation results during the aging process while capturing the relationship between voltage and capacity. This reduces the capacity estimation error. Based on the obtained capacity estimation results, the battery aging condition can be observed in a timely manner, thereby assessing the health status of the battery system. Attached Figure Description

[0018] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments of this application with reference to the accompanying drawings.

[0019] Figure 1 The illustration shows an application scenario of the battery capacity estimation method, train control method, device, and medium according to embodiments of this application.

[0020] Figure 2 A flowchart illustrating a battery capacity estimation method according to an embodiment of this application is shown schematically.

[0021] Figure 3 The illustration shows a training set data determined by sliding a window with a preset capacity range according to an embodiment of this application.

[0022] Figure 4 The diagram illustrates the test results for different capacity ranges according to embodiments of this application.

[0023] Figure 5 The illustration shows the test results of data determined from 0 to 85% capacity of the energy storage battery according to an embodiment of this application.

[0024] Figure 6 The diagram illustrates the transfer mechanism of the capacity estimation model according to an embodiment of this application.

[0025] Figure 7 The illustration shows the results of random training of a capacity estimation model with three different batteries as a test set and a window of 1Ah, according to an embodiment of this application.

[0026] Figure 8 The illustration schematically shows the test effect of changing the window length by modifying the capacity range according to an embodiment of this application.

[0027] Figure 9 The illustration shows the effect of different recording intervals on capacity estimation according to embodiments of this application.

[0028] Figure 10 The illustration schematically shows an error comparison of the increased capacity response sequence according to an embodiment of this application.

[0029] Figure 11 The illustration shows a comparison of errors in the battery's full-range sliding capacity increase response sequence when measurement errors exist, according to an embodiment of this application.

[0030] Figure 12 A flowchart illustrating a train control method according to an embodiment of this application is shown schematically.

[0031] Figure 13 A schematic block diagram of a battery capacity estimation device according to an embodiment of this application is shown.

[0032] Figure 14 A schematic block diagram of a train control device according to an embodiment of this application is shown.

[0033] Figure 15 A block diagram schematically illustrates an electronic device suitable for implementing a battery capacity estimation method and a train control method according to embodiments of this application. Detailed Implementation

[0034] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0035] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0036] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0037] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0038] In the technical solution of this application, the acquisition, storage, and application of user personal information comply with relevant laws and regulations, necessary confidentiality measures have been taken, and there is no violation of public order and good morals. In the technical solution of this application, user authorization or consent has been obtained before acquiring or collecting user personal information.

[0039] With the standardization of battery materials, capacity, and size, the commonly used battery types, capacities, and packaging methods are gradually becoming fixed, and fewer battery models will be used in mass production. Therefore, aging tests on batteries are crucial. Battery capacity estimation methods are related to battery aging mechanisms, aging paths, and State of Charge (SOC) ranges. Furthermore, since the degradation of open-circuit voltage and internal resistance with battery aging in battery models is not ideal, it is necessary to first address the multi-factor coupling problem. If a large amount of test data can be used to analyze the capacity degradation weights in different SOC ranges and clarify the patterns of open-circuit voltage and internal resistance changes, voltage similarity methods can also achieve high accuracy.

[0040] However, analysis of different SOC ranges of the battery reveals that battery aging depends on the relative positions of the positive and negative electrodes, the loss ratio of different materials during aging, and the aging stage. The weight of the impact on battery capacity loss varies with the aging process across different SOC ranges, making this a complex problem of variable-weight linear superposition. Furthermore, changes in the battery's internal resistance after aging significantly affect battery capacity estimation. Simultaneously, measurement errors in the Battery Management System (BMS) can introduce errors into the neural network capacity estimation results based on external voltage characteristics, leading to inaccurate battery capacity estimates.

[0041] To at least partially address the technical problems existing in related technologies, this application provides a battery capacity estimation method, apparatus, device, and medium. The method includes: acquiring a target voltage sequence and a corresponding capacity response sequence related to an energy storage battery; the target voltage sequence includes the terminal voltages of the energy storage battery at multiple charging moments during charging; and the capacity response data in the capacity response sequence characterizes the change in the energy storage battery's charge capacity relative to the terminal voltages in the target voltage sequence; processing the target voltage sequence and capacity response sequence using a capacity estimation model to obtain a capacity estimation result; and the capacity estimation result characterizes the amount of charge the energy storage battery can store after charging is complete.

[0042] In the technical solution of this invention, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0043] Figure 1 The illustration shows an application scenario of the battery capacity estimation method, train control method, device, and medium according to embodiments of this application.

[0044] like Figure 1 As shown, application scenario 100 according to this embodiment may include a vehicle 101, a network 102, and a server 103. The network 102 is used as a medium to provide a communication link between the vehicle 101 and the server 103. The network 102 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0045] Users can operate terminal devices such as computers installed in vehicle 101 to interact with server 103 via network 102 to receive or send messages. Various communication client applications can be installed on the terminal devices, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0046] Terminal devices can be various electronic devices with a display screen and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0047] Server 103 can be a server that provides various services, such as a backend management server that supports websites browsed by users using terminal devices (for example only). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0048] It should be noted that the battery capacity estimation method and train control method provided in this application embodiment can generally be executed by the vehicle 101. Accordingly, the battery capacity estimation device and train control device provided in this application embodiment can generally be installed in the vehicle 101.

[0049] Alternatively, the battery capacity estimation method and train control method provided in this application embodiment can generally also be executed by server 103. Correspondingly, the battery capacity estimation device and train control device provided in this application embodiment can generally also be located in server 103. The battery capacity estimation method and train control method provided in this application embodiment can also be executed by a server or server cluster that is different from server 103 and capable of communicating with vehicle 101 and / or server 103. Correspondingly, the battery capacity estimation device and train control device provided in this application embodiment can also be located in a server or server cluster that is different from server 103 and capable of communicating with vehicle 101 and / or server 103.

[0050] It should be understood that Figure 1 The number of vehicles, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0051] Figure 2 A flowchart illustrating a battery capacity estimation method according to an embodiment of this application is shown schematically.

[0052] like Figure 2 As shown, the battery capacity estimation method of this embodiment includes operations S210 to S220.

[0053] In operation S210, the target voltage sequence and the corresponding capacity response sequence related to the energy storage battery are obtained.

[0054] In operation S220, the target voltage sequence and capacity response sequence are processed using the capacity estimation model to obtain the capacity estimation result.

[0055] The target voltage sequence can include the terminal voltages of the energy storage battery at multiple charging moments during the charging process. The capacity response data in the capacity response sequence characterizes the change in the energy storage battery's capacity relative to the terminal voltages in the target voltage sequence. The target voltage sequence can be in the form of [V1, V2, ... V]. n-1 V n Capacity response sequence [IC1, IC2…IC] n-1 IC n Each terminal voltage V corresponds to the change in the voltage value at that terminal, IC.

[0056] The capacity estimation result characterizes the amount of electricity that a battery can store after charging. The capacity estimation result can be obtained by processing the target voltage sequence and the capacity response sequence using a capacity estimation model.

[0057] According to the embodiments of this application, by concatenating the target voltage sequence with the corresponding capacity response sequence and inputting them together as input features of the capacity estimation model, the voltage-capacity relationship can be captured while utilizing the characteristics of voltage difference in the capacity response sequence to solve the interference of nonlinear changes in open-circuit voltage and internal resistance and measurement errors on the capacity estimation results during aging, thereby reducing capacity estimation errors. Based on the obtained capacity estimation results, the battery aging condition can be observed in a timely manner, thereby assessing the health status of the battery system.

[0058] According to an embodiment of this application, the target voltage sequence is obtained based on the following operations: acquiring the voltage-capacity curve of the energy storage battery charging process; dividing the voltage-capacity curve based on a preset capacity interval window to obtain multiple voltage-capacity curve segments; determining multiple terminal voltages in the target voltage sequence based on the acquisition time attribute of each of the multiple terminal voltages in the voltage-capacity curve segments; and determining the target voltage sequence based on the multiple terminal voltages.

[0059] Due to battery aging, parameters such as battery capacity will vary, resulting in different voltage sequence lengths for each capacity test. To ensure that all input vectors have consistent lengths and that complete charging curves of all batteries in the training set can participate in the model training process, the capacity estimation model can be divided into pre-defined capacity interval windows for each voltage-capacity curve. This divides a battery's voltage-capacity curve into segments of equal length but different starting SOC points. The capacity value between each voltage-capacity curve segment can be set by the user.

[0060] Based on the acquisition time attributes of multiple terminal voltages in a voltage-capacity curve segment, multiple terminal voltages arranged according to their acquisition time attributes in the target voltage sequence can be determined. Based on these multiple terminal voltages, the target voltage sequence can be determined. Considering data transmission bandwidth, the number of terminal voltages in the target voltage sequence can be reduced by increasing the preset time interval between the multiple terminal voltages to decrease data transmission pressure. A preset capacity interval that is too small will result in a large error in the capacity estimation result; experiments show that a preset capacity interval greater than 1 amp-hour can be used. Similarly, a preset time interval that is too large will also result in a large error in the capacity estimation result; experiments show that a preset time interval less than or equal to 120 seconds can be used.

[0061] Based on the slope of the change corresponding to the terminal voltage in the voltage-capacity curve segment, i.e., dQ / dV, the capacity response data corresponding to the terminal voltage can be determined. The capacity response sequence can include multiple capacity response data.

[0062] According to embodiments of this application, by dividing the voltage-capacity curves acquired during the charging process into multiple voltage-capacity curve segments, it is possible to ensure that the vector length of the model input is the same. Based on the acquisition time attributes corresponding to each terminal voltage, the terminal voltage that meets the preset time interval can be determined, reducing data transmission pressure.

[0063] According to an embodiment of this application, dividing the voltage-capacity curve based on a preset capacity range window includes: deleting a specified voltage-capacity curve segment from the voltage-capacity curve to obtain a processed voltage-capacity curve, wherein the specified voltage-capacity curve segment corresponds to a specified capacity range, and the specified capacity range is the capacity range between a specified capacity value and the upper limit capacity value of the energy storage battery; and dividing the processed voltage-capacity curve based on the preset capacity range window.

[0064] The specified capacity range is the capacity interval between a specified capacity value and the upper limit capacity value of the energy storage battery. For example, the specified capacity value is 85% of the energy storage battery capacity, and the upper limit capacity value is 100%. Deleting the specified voltage-capacity curve from the voltage-capacity curve will result in a processed voltage-capacity curve. The capacity corresponding to the processed voltage-capacity curve is 0%-85%. The processed voltage-capacity curve can be divided based on a preset capacity range window.

[0065] According to embodiments of this application, high-end charging data (i.e., specified voltage-capacity curve segments) in the voltage-capacity curve can be removed, and the processed voltage-capacity curve can be divided to reduce the error of the capacity estimation results.

[0066] Figure 3 The illustration shows a training set data determined by sliding a window with a preset capacity range according to an embodiment of this application.

[0067] like Figure 3 As shown, the voltage-capacity curve obtained by each battery during charging can be divided into segments with a fixed window size. This divides the voltage-capacity curve of a battery into voltage sequences of equal length but different starting SOC points (Qstart1, Qstart2, Qstart3). The capacity interval between each target voltage sequence can be set by the user. The figure shows three voltage-capacity curve segments, each containing a target voltage sequence and the corresponding starting capacity Qstart and actual capacity Q.

[0068] Figure 4 The diagram illustrates the test results for different capacity ranges according to embodiments of this application.

[0069] First, the dataset for this experiment can be constructed. Since the capacity estimation model requires a large number of voltage-capacity curves at various charging times for each battery as a training set, different aging paths were tested on individual batteries, and performance tests were conducted at specific aging stages. Complete charging data for 51 sets of batteries at different health states were obtained, with battery capacities ranging from 70% to 100%. Specific capacity values ​​are shown in Table 1, with a maximum capacity of 2.61 Ah and a minimum of 1.83 Ah. The capacity from the first performance test represents the battery's capacity when it first appeared at the factory; the capacity from the second performance test represents the battery's capacity after 50 charge-discharge cycles; the capacity from the third performance test represents the battery's capacity after 100 charge-discharge cycles; and the capacity from the fourth performance test represents the battery's capacity after 150 charge-discharge cycles.

[0070] Table 1. List of 51 sets of data

[0071]

[0072] Batteries No. 3, 5, and 6 can be selected from Table 1 as the test set, and the performance test data of other batteries as the training set. Since there are few battery samples with a capacity below 80%, data from batteries with a capacity below 80% can be removed. From the 37 complete charging curves of the remaining batteries, a 1Ah charging interval was extracted every 0.1Ah to obtain the input layer data for the training set, totaling 2228 sample segments. Validation was performed using 10 sets of charging data from 3 batteries. To evaluate the accuracy differences between different capacity intervals, a 1Ah sliding window was still used for extraction, resulting in 149 sample segments. The training results are as follows: Figure 4As shown in Figure (a), the estimated results of the actual capacity and the initial value of the active window for the three batteries at different aging stages are shown in Figure (b). The actual value and the estimated value are calculated after calculating the actual capacity degradation rate. After training, the model's estimation error is the largest at 4.4%. The three points with the largest errors appear in the high-capacity range of the three sets of data. Therefore, it is easy to produce a large estimation error when using charging data in the high-capacity range for capacity estimation. It may be necessary to readjust the capacity range covered when truncating the curve based on the error.

[0073] Figure 5 The illustration shows the test results of energy storage batteries from 0 to 85% capacity according to embodiments of this application.

[0074] Eleven batteries were selected as the training set, and data from batteries 3, 5, and 6 were used as the test set. The high-end 15% of charging data (i.e., charging data from 0% to 85% capacity) was removed from both the training and test sets. The input layer data for the training set was obtained by extracting charging intervals of 1Ah every 0.1Ah. Ten sets of charging data from three batteries were used for validation, with the validation battery interval also set to 1Ah, resulting in a total of 129 sample segments. The training results are as follows. Figure 5 As shown, after training, the model's estimation error is at most 2.4%, which is a significant improvement in estimation accuracy compared to the full range.

[0075] According to an embodiment of this application, a capacity estimation model is used to process the target voltage sequence and the capacity response sequence to obtain a capacity estimation result, including: determining voltage-capacity input features based on the target voltage sequence and the capacity response sequence, wherein the voltage-capacity input features characterize the terminal voltage at the charging moment and the change of the terminal voltage; processing the voltage-capacity input features using an activation function to obtain hidden features, wherein the hidden features characterize the interaction between the capacity of the energy storage battery and the terminal voltage; and processing the hidden features to obtain a capacity estimation result.

[0076] Based on the target voltage sequence and capacity response sequence, the voltage-capacity input characteristics can be determined. These characteristics can be 2×n dimensional data, where n represents the number of terminal voltages in the target voltage sequence. The voltage-capacity input characteristics characterize the terminal voltage at each charging moment and its changes.

[0077] By processing the voltage and capacity input features using activation functions, hidden features can be obtained. These hidden features characterize the interaction between the capacity and terminal voltage of the energy storage battery. The hidden layer learns the correspondence between terminal voltage and capacity from the target voltage sequence and the sensitivity of voltage to capacity changes from the capacity response sequence. Specifically, the larger the capacity response data, the more drastic the voltage change with capacity, and the easier it is for the model to capture the corresponding feature intervals (such as inflection points). This is used to determine which stage of aging the battery is currently in, thereby correcting the capacity estimation. Finally, mapping the hidden features yields the capacity estimation result.

[0078] According to embodiments of this application, voltage and capacity input features can be determined based on the target voltage sequence and capacity response sequence. By processing the voltage and capacity input features using an activation function, hidden features can be obtained. This allows the voltage-capacity relationship to be captured while the characteristics of voltage difference in the capacity response sequence are utilized to address the interference of nonlinear changes in open-circuit voltage and internal resistance during aging and measurement errors on the capacity estimation results, thereby reducing capacity estimation errors.

[0079] Figure 6 The diagram illustrates the transfer mechanism of the capacity estimation model according to an embodiment of this application.

[0080] Capacity estimation models, as a type of multilayer feedforward neural network, use backpropagation to train the model and minimize the error between the actual output and the expected output. A capacity estimation model consists of an input layer, hidden layers, and an output layer, where there can be one or more hidden layers. Each layer contains several nodes connected by weights. The signal enters from the input layer, is processed by the hidden layers, and is finally output by the output layer.

[0081] By employing a commonly used three-layer network structure, including one hidden layer, the activation function of the hidden layer can be the tansig function, and the output layer can use the purelin function, enabling the model to fit nonlinearities. The input to the capacity estimation model consists of 30 sets of processed target voltage sequences and capacity response sequences of batteries. The actual usable capacity and initial capacity of the battery corresponding to each target voltage sequence are the output variables of the capacity estimation model. After experimentation, a hidden layer node count of 3 was selected. Therefore, the established capacity estimation model is as follows: Figure 6 As shown, the input layer contains 30×m neurons, where m represents the number of equal-length windows intercepted on a complete charging voltage-capacity curve. The intermediate hidden layer contains 3 neurons, which are output to the output layer through an activation function. Each neuron in the layer is weighted and accumulated with the value passed from the previous layer, plus its own threshold b, before being passed to the next node.

[0082] Based on the basic principles of the capacity estimation model, the expression is shown in formula (1).

[0083] (1);

[0084] Among them, W 11 ...W 1n W 12 ...W 2n W 13 ...W 3n All are weights, b1 and b2 are biases, x1...x n The input data consists of the target voltage sequence and the capacity response sequence.

[0085] After determining the preset capacity range window of the target voltage sequence, the number of terminal voltages input in each matrix is ​​the same, and the charging capacity corresponding to the window is the same. The training set includes both the capacity starting point and the actual battery capacity as the prediction result; the test set uses the target voltage sequence and the capacity response sequence as input.

[0086] According to an embodiment of this application, the capacity estimation model is trained based on the following steps: obtaining sample voltage sequences, sample capacity response sequences, and tag data corresponding to multiple sets of energy storage batteries, wherein the tag data includes the actual capacity corresponding to the sample capacity response sequence; inputting the sample voltage sequences and sample capacity response sequences into the initial capacity estimation model and outputting the sample capacity estimation results; training the initial capacity estimation model based on the sample capacity estimation results and tag data to obtain the trained capacity estimation model.

[0087] The labeled data can include the initial capacity in the voltage-capacity curve segment and the actual capacity corresponding to the sample capacity response sequence. Inputting the sample voltage sequence and sample capacity response sequence into the initial capacity estimation model outputs the sample capacity estimation result. Training the initial capacity estimation model based on the sample capacity estimation result and the labeled data yields the trained capacity estimation model.

[0088] According to embodiments of this application, by training an initial capacity estimation model using sample voltage sequences, sample capacity response sequences, and tag data corresponding to multiple sets of energy storage batteries, the problem of open-circuit voltage and internal resistance degradation with aging in battery modeling for single-cell capacity estimation methods in electric vehicles and new energy rail transit battery systems—a problem stemming from multiple coupled factors—can be solved. This addresses the issue of low estimation accuracy caused by nonlinear changes in open-circuit voltage and internal resistance during aging and BMS detection errors, providing a solution for battery performance evaluation in current service.

[0089] According to an embodiment of this application, training an initial capacity estimation model based on sample capacity estimation results and label data to obtain a trained capacity estimation model includes: processing the sample capacity estimation results and label data according to a loss function to obtain a loss value; and training the initial capacity estimation model based on the loss value to obtain a trained capacity estimation model.

[0090] By processing the sample capacity estimation results and label data using the loss function, a loss value can be obtained. Using this loss value, an initial capacity estimation model can be trained to obtain a trained capacity estimation model.

[0091] Figure 7 The illustration shows the results of random training of a capacity estimation model with three different batteries as a test set and a window of 1Ah, according to an embodiment of this application.

[0092] To ensure the model's performance on different datasets, cross-validation was used to verify its generalization ability. Data from three other batteries were randomly selected as the training set, and data from another battery was used as the test set. This process of training and validating the model was repeated twice. Figure 7 As shown, when the test sets are batteries 5, 8, and 14 (i.e., Figure a) and batteries 2, 3, and 4 (i.e., Figure c), the errors are all within 3%, thus verifying the model's generalization ability.

[0093] Figure 8 The illustration schematically shows the test effect of changing the window length by modifying the capacity range according to an embodiment of this application.

[0094] Further consideration is to modify the range of the preset capacity interval window. This could involve changing the window size, setting the preset capacity interval window to 0.52 Ah for both training and validation. Figure 8 The error shown is significantly increased, especially in the high-capacity range where the maximum estimation error occurs. Therefore, the selected sliding window should not be too short.

[0095] Figure 9 The illustration shows the effect of different time intervals on capacity estimation according to embodiments of this application.

[0096] Considering the data transmission bandwidth issue, to reduce data transmission pressure and model training time, the data interval was further increased, with a data point taken every 120 seconds and 180 seconds respectively. The results are as follows. Figure 9 As shown, when the time interval is extended from 60s to 120s, the capacity estimation result does not change much and has no significant impact, but the model training speed is improved. When the data interval is 180s, the error increases significantly.

[0097] Table 2 Summary of errors for different input data

[0098]

[0099] In summary, the model training was conducted using the voltage characteristics of the interval charging curve, and the results are listed in Table 2. It can be concluded that the voltage data used needs to cover approximately 45% of the SOC range (1.2Ah), and the maximum sampling interval can be extended to 2 minutes during 0.5C charging. Therefore, for the background battery capacity estimation work, only the charging process data at 2-minute intervals needs to be recorded, which will greatly reduce the storage pressure on the background.

[0100] Figure 10 The illustration schematically shows an error comparison of the increased capacity response sequence according to an embodiment of this application.

[0101] Figure 10 In (a), the model's input data consists only of the target voltage sequence (i.e., the charging voltage curve alone). dQ / dV is calculated every 5 points (i.e., 5 minutes) on the defined voltage-capacity curve to calculate the capacity response data, which, together with the target voltage sequence (i.e., the charging voltage), serves as feature input to train the model and verify its accuracy. Figure 10 (b) shows that when batteries 2, 3 and 4 are used as the validation set, it can be found that after adding capacity response data (i.e. IC), the maximum capacity estimation error decreased from 2.5% to 2%, indicating that the model accuracy improved after adding capacity response data. This shows that using capacity is beneficial to reduce error.

[0102] Figure 11 The illustration shows a comparison of errors after a full-range sliding capacity increase response sequence of two batteries when measurement errors exist, according to an embodiment of this application.

[0103] Using capacity response data can reduce the influence of internal resistance on the one hand, and on the other hand, by differentiating the terminal voltages, it can effectively reduce model errors caused by BMS measurement errors. To verify this, a measurement error of 0~3mV was added to the voltage curves of 51 sample batteries, i.e., 51 data points of -3~3mV were randomly generated and superimposed on the charging curves of different batteries. The model errors after training the model using only voltage curves and after adding IC curve data are as follows: Figure 11 As shown, when considering BMS measurement error, the battery capacity estimation error using only charging voltage data reaches 4.3%, while the error is reduced to 2.2% after adding capacity response data.

[0104] Figure 12 A flowchart illustrating a train control method according to an embodiment of this application is shown schematically.

[0105] like Figure 12 As shown, the train control method of this embodiment includes operations S1210 to S1220.

[0106] In operation S1210, the capacity estimation result of the energy storage battery is determined according to the battery capacity estimation method.

[0107] When operating S1220, if the capacity estimation result does not meet the preset conditions, an alarm message is sent to the train control center.

[0108] The capacity estimation result of the energy storage battery can be determined based on the battery capacity estimation method. If the capacity estimation result is less than a preset threshold, an alarm message can be sent to the train control center.

[0109] Based on the above-described battery capacity estimation method, this application also provides a battery capacity estimation device. The following will be combined with... Figure 13 The device is described in detail.

[0110] Figure 13 A schematic block diagram of a battery capacity estimation device according to an embodiment of this application is shown.

[0111] like Figure 13 As shown, the battery capacity estimation device 1300 of this embodiment includes an acquisition module 1310 and a capacity estimation result module 1320.

[0112] The acquisition module 1310 is used to acquire a target voltage sequence and a corresponding capacity response sequence related to the energy storage battery. The target voltage sequence includes the terminal voltage of the energy storage battery at multiple charging moments during the charging process, and the capacity response sequence characterizes the change of the energy storage battery's capacity relative to the terminal voltage in the target voltage sequence.

[0113] The capacity estimation result module 1320 is used to process the target voltage sequence and capacity response sequence using the capacity estimation model to obtain the capacity estimation result, which characterizes the amount of electricity that the energy storage battery can store after charging.

[0114] The acquisition module 1310 includes: an acquisition unit, a segment acquisition unit, a terminal voltage acquisition unit, and a capacity response data determination unit.

[0115] According to the embodiments of this application, by concatenating the target voltage sequence with the corresponding capacity response sequence and inputting them together as input features of the capacity estimation model, the voltage-capacity relationship can be captured while the characteristics of the voltage difference in the capacity response sequence are used to solve the interference of measurement error on the capacity estimation result, thereby reducing the capacity estimation error. Based on the obtained capacity estimation result, the battery aging condition can be observed in a timely manner.

[0116] The acquisition unit is used to acquire the voltage-capacity curve during the charging process of the energy storage battery.

[0117] The segment acquisition unit is used to divide the voltage capacity curve based on a preset capacity range window to obtain multiple voltage capacity curve segments.

[0118] The terminal voltage acquisition unit is used to determine multiple terminal voltages arranged based on the acquisition time attributes in the target voltage sequence, based on the acquisition time attributes of each terminal voltage in the voltage capacity curve segment.

[0119] The capacity response data determination unit is used to determine the capacity response data corresponding to the terminal voltage based on the slope of the change in the voltage-capacity curve segment corresponding to the terminal voltage. The capacity response sequence includes multiple capacity response data.

[0120] The segments obtained include: deleting sub-units and dividing sub-units.

[0121] The deletion sub-unit is used to delete a specified voltage-capacity curve segment from the voltage-capacity curve to obtain a processed voltage-capacity curve. The specified voltage-capacity curve segment corresponds to a specified capacity range, which is the capacity range between a specified capacity value and the upper limit capacity value of the energy storage battery.

[0122] Sub-units are used to divide the processed voltage-capacity curves based on a preset capacity range window.

[0123] The capacity estimation result module 1320 includes: an input feature determination unit, a hidden feature acquisition unit, and a processing unit.

[0124] The input feature determination unit is used to determine the voltage and capacity input features based on the target voltage sequence and the capacity response sequence. The voltage and capacity input features characterize the terminal voltage at the charging moment and the changes in the terminal voltage.

[0125] The hidden feature acquisition unit is used to process the voltage and capacity input features using an activation function to obtain hidden features. The hidden features characterize the interaction between the capacity and terminal voltage of the energy storage battery.

[0126] The processing unit is used to process the hidden features to obtain the capacity estimation result.

[0127] The capacity estimation result module 1320 includes: a sample acquisition unit, an output unit, and a training unit.

[0128] The sample acquisition unit is used to acquire the sample voltage sequence, sample capacity response sequence and tag data corresponding to each of the multiple energy storage batteries. The tag data includes the actual capacity corresponding to the sample capacity response sequence.

[0129] The output unit is used to input the sample voltage sequence and sample capacity response sequence into the initial capacity estimation model and output the sample capacity estimation result.

[0130] The training unit is used to train an initial capacity estimation model based on the sample capacity estimation results and label data, resulting in a trained capacity estimation model.

[0131] The training unit includes: the loss value sub-unit and the training sub-unit.

[0132] The loss value sub-unit is used to process the sample size estimation results and label data according to the loss function to obtain the loss value.

[0133] The training sub-unit is used to train the initial capacity estimation model based on the loss value, resulting in the trained capacity estimation model.

[0134] Based on the above-described train control method, this application also provides a train control device. The following will be combined with... Figure 14 The device is described in detail.

[0135] Figure 14 A schematic block diagram of a train control device according to an embodiment of this application is shown.

[0136] like Figure 14 As shown, the train control device 1400 of this embodiment includes a capacity determination module 1410 and a sending module 1420.

[0137] The capacity determination module 1410 is used to determine the capacity estimation result of the energy storage battery in the train power system according to the battery capacity estimation method.

[0138] The sending module 1420 is used to send alarm information to the train control center when the capacity estimation result does not meet the preset conditions.

[0139] According to embodiments of this application, any multiple modules among the acquisition module 1310, capacity estimation result module 1320, capacity determination module 1410, and transmission module 1420 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the acquisition module 1310, capacity estimation result module 1320, capacity determination module 1410, and transmission module 1420 can be at least partially implemented as hardware circuitry, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system-on-a-chip, a system-on-a-substrate, a system-on-package, an application-specific integrated circuit (ASIC), or implemented in hardware or firmware by any other reasonable means of integrating or packaging the circuitry, or implemented in software, hardware, or firmware, or in any suitable combination of any of these three implementation methods. Alternatively, at least one of the acquisition module 1310, capacity estimation result module 1320, capacity determination module 1410, and transmission module 1420 may be implemented at least partially as a computer program module, which can perform corresponding functions when the computer program module is run.

[0140] Figure 15 A block diagram schematically illustrates an electronic device suitable for implementing a battery capacity estimation method according to an embodiment of this application.

[0141] like Figure 15 As shown, an electronic device 1500 according to an embodiment of this application includes a processor 1501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1502 or a program loaded from a storage portion 1508 into a random access memory (RAM) 1503. The processor 1501 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1501 may also include onboard memory for caching purposes. The processor 1501 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0142] RAM 1503 stores various programs and data required for the operation of electronic device 1500. Processor 1501, ROM 1502, and RAM 1503 are interconnected via bus 1504. Processor 1501 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 1502 and / or RAM 1503. It should be noted that the programs may also be stored in one or more memories other than ROM 1502 and RAM 1503. Processor 1501 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0143] According to embodiments of this application, the electronic device 1500 may further include an input / output (I / O) interface 1505, which is also connected to a bus 1504. The electronic device 1500 may also include one or more of the following components connected to the I / O interface 1505: an input section 1506 including a keyboard, mouse, etc.; an output section 1507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1508 including a hard disk, etc.; and a communication section 1509 including a network interface card such as a LAN card, modem, etc. The communication section 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to the I / O interface 1505 as needed. A removable medium 1511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1510 as needed so that computer programs read from it can be installed into the storage section 1508 as needed.

[0144] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0145] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 1502 and / or RAM 1503 and / or one or more memories other than ROM 1502 and RAM 1503 described above.

[0146] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the battery capacity estimation method provided in the embodiments of this application.

[0147] When the computer program is executed by the processor 1501, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0148] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 1509, and / or installed from the removable medium 1511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0149] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1509, and / or installed from the removable medium 1511. When the computer program is executed by the processor 1501, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0150] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0151] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0152] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this application can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments and / or claims of this application can be combined or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

[0153] The embodiments of this application have been described above. However, these embodiments are merely illustrative and not intended to limit the scope of this application. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this application is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this application, and all such substitutions and modifications should fall within the scope of this application.

Claims

1. A battery capacity estimation method, characterized in that, include: Obtain a target voltage sequence and a corresponding capacity response sequence related to the energy storage battery. The target voltage sequence includes the terminal voltage of the energy storage battery at multiple charging moments during the charging process. The capacity response data in the capacity response sequence characterizes the change in the energy storage battery's charge capacity relative to the terminal voltage in the target voltage sequence. The target voltage sequence and the capacity response sequence are processed using a capacity estimation model to obtain a capacity estimation result, which characterizes the amount of electricity that the energy storage battery can store after charging.

2. The method according to claim 1, characterized in that, The target voltage sequence is obtained based on the following operation. Obtain the voltage-capacity curve of the energy storage battery during the charging process; The voltage capacity curve is divided into multiple voltage capacity curve segments based on a preset capacity range window. Based on the acquisition time attributes of each terminal voltage in the voltage capacity curve segment, determine the multiple terminal voltages in the target voltage sequence arranged according to the acquisition time attributes; Based on the multiple terminal voltages, a target voltage sequence is determined.

3. The method according to claim 2, characterized in that, The voltage capacity curve is divided based on a preset capacity range window, including: A specified voltage capacity curve segment is deleted from the voltage capacity curve to obtain a processed voltage capacity curve. The specified voltage capacity curve segment corresponds to a specified capacity range, which is the capacity range between a specified capacity value and the upper limit capacity value of the energy storage battery. The processed voltage-capacity curve is divided based on a preset capacity range window.

4. The method according to claim 1, characterized in that, The process of using a capacity estimation model to process the target voltage sequence and the capacity response sequence to obtain the capacity estimation result includes: Based on the target voltage sequence and the capacity response sequence, voltage capacity input characteristics are determined, wherein the voltage capacity input characteristics characterize the terminal voltage at the charging moment and the change of the terminal voltage. The voltage-capacity input features are processed using an activation function to obtain hidden features, which characterize the interaction between the capacity and terminal voltage of the energy storage battery. The hidden features are processed to obtain the capacity estimation result.

5. The method according to claim 1, characterized in that, The capacity estimation model was trained based on the following steps: Acquire sample voltage sequences, sample capacity response sequences, and tag data corresponding to each of multiple energy storage batteries, wherein the tag data includes the actual capacity corresponding to the sample capacity response sequence; The sample voltage sequence and the sample capacity response sequence are input into the initial capacity estimation model, and the sample capacity estimation result is output. The initial capacity estimation model is trained based on the sample capacity estimation results and the label data to obtain the trained capacity estimation model.

6. The method according to claim 5, characterized in that, The step of training an initial capacity estimation model based on the sample capacity estimation result and the label data to obtain a trained capacity estimation model includes: The loss value is obtained by processing the sample size estimation result and the label data according to the loss function; The initial capacity estimation model is trained based on the loss value to obtain the trained capacity estimation model.

7. A train control method, comprising: The capacity estimation result of the energy storage battery is determined by the method according to any one of claims 1 to 6; If the capacity estimation result does not meet the preset conditions, an alarm message is sent to the train control center.

8. A train control device, comprising: A capacity determination module is used to determine the capacity estimation result of the energy storage battery in the train power system by the method according to any one of claims 1 to 6; The sending module is used to send alarm information to the train control center when the capacity estimation result does not meet the preset conditions.

9. A battery capacity estimation device, comprising: The acquisition module is used to acquire a target voltage sequence and a corresponding capacity response sequence related to the energy storage battery. The target voltage sequence includes the terminal voltage of the energy storage battery at multiple charging moments during the charging process, and the capacity response sequence characterizes the change of the energy storage battery's capacity relative to the terminal voltage in the target voltage sequence. The capacity estimation result module is used to process the target voltage sequence and the capacity response sequence using a capacity estimation model to obtain a capacity estimation result, which characterizes the amount of electricity that the energy storage battery can store after charging.

10. A train, comprising: A battery capacity estimation device applied to the battery capacity estimation method according to any one of claims 1 to 6.

11. An electronic device, comprising: One or more processors; Memory, used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1 to 6.

12. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 6.

13. A computer program product comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images