SOC estimation method and device suitable for hybrid battery
By real-time acquisition of battery current and combining the A-time integration method, gated cycle unit and Kalman filtering algorithm, the problem of inaccurate SOC estimation of hybrid batteries is solved, and higher prediction reliability and stability are achieved.
Patent Information
- Application Number
- CN202510791056.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-13
- Publication Date
- 2025-08-26
AI Technical Summary
In the prior art, the SOC estimation of hybrid batteries is not accurate enough when they are continuously operating, making it difficult to effectively reduce the impact of noise and error, resulting in unstable prediction results.
By collecting battery current in real time, calculating SOC values using the A-time integration method, and combining the gated cycle unit and Kalman filtering algorithm, battery health status characteristics are extracted and predicted, reducing the impact of noise and error, and improving the reliability and stability of prediction.
Accurate estimation of SOC in hybrid batteries is realized, reducing the impact of noise and error on the prediction results, and improving the reliability and stability of the prediction.
Smart Images
Figure CN120539601A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of hybrid batteries, and in particular to a method and device for estimating the SOC of a hybrid battery. Background Art
[0002] A battery management system (BMS) is an electrical device primarily used to manage rechargeable batteries (individual cells or battery packs). Its primary tasks are monitoring, calculation, communication, protection, and optimization. The battery's state of charge (SOC) is the ratio of a battery's current remaining capacity to its fully charged capacity.
[0003] There are many SOC estimation methods. Among them, the ampere-hour integration method is a classic method for evaluating the remaining battery charge. Its advantage is its simplicity and ease of use. However, its disadvantage is that it requires the initial value of the battery's state of charge to be known. Moreover, the estimation model relies on idealized assumptions and static parameters, making it difficult to accurately estimate battery data under dynamic conditions (nonlinear battery operation), thus requiring frequent calibration.
[0004] Therefore, how to accurately estimate the SOC when the hybrid battery continues to work is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention
[0005] In view of this, an embodiment of the present invention provides an SOC estimation method and device suitable for a hybrid battery to solve the problem in the prior art of how to accurately estimate the SOC when the hybrid battery continues to operate; that is, the embodiment of the present invention can effectively reduce the impact of noise and error on the prediction results and improve the reliability and stability of the prediction.
[0006] According to one aspect of the present invention, a SOC estimation method for a hybrid battery is provided, the method comprising: acquiring battery current in real time, and calculating an SOC value based on the battery current, wherein the battery current includes a training battery current and a non-training battery current; processing the SOC value and a gated cycle unit to obtain a battery health status characteristic; and processing the non-training battery current, the battery health status characteristic, and an SOC prediction model based on a Kalman filter algorithm to obtain a predicted SOC value.
[0007] In one embodiment, the processing based on the SOC value and the gated cycle unit to obtain the battery health status characteristics includes: preprocessing the SOC value to obtain a target SOC value; and extracting features based on the target SOC value and the gated cycle unit to obtain the battery health status characteristics.
[0008] In one embodiment, the processing based on the non-training battery current, the battery health status characteristics and the SOC prediction model based on the Kalman filter algorithm to obtain a predicted SOC value includes: training the SOC prediction model based on the Kalman filter algorithm based on the battery health status characteristics to obtain a target SOC prediction model; calling the target SOC prediction model to calculate based on the non-training battery current to obtain a predicted SOC value.
[0009] In one embodiment, the SOC value is calculated based on the battery current, and the SOC value is:
[0010]
[0011] Among them, SCO(t0) is the initial battery state of charge value, C n is the battery capacity correction value, I(τ) is the battery current, and η is the coulomb efficiency correction value.
[0012] In one embodiment, the processing is performed based on the SOC value and the gated cycle unit to obtain battery health status characteristics, and the battery health status characteristics include capacity decay rate, internal resistance growth rate, and charge and discharge time.
[0013] In one embodiment, the capacity decay rate is:
[0014]
[0015] Among them, C0 is the initial capacity of the battery, C t is the remaining capacity of the battery after time t.
[0016] According to another aspect of the present invention, an SOC estimation device for a hybrid battery is provided, the SOC estimation device for a hybrid battery comprising: a data acquisition module, a data processing module and a prediction module, wherein the data acquisition module is used to acquire battery current in real time, and calculate the SOC value based on the battery current, wherein the battery current includes a training battery current and a non-training battery current; the data processing module is used to process based on the SOC value and a gated cycle unit to obtain a battery health status characteristic; the prediction module is used to process based on the non-training battery current, the battery health status characteristic and an SOC prediction model based on a Kalman filter algorithm to obtain a predicted SOC value.
[0017] In one embodiment, the data processing module includes: a preprocessing unit and a feature extraction unit, the preprocessing unit is used to preprocess the SOC value to obtain a target SOC value; the feature extraction unit is used to perform feature extraction based on the target SOC value and the gated cycle unit to obtain the battery health status feature.
[0018] In one embodiment, the prediction module includes a model training unit and an SOC prediction unit, wherein the model training unit is used to train the SOC prediction model based on the Kalman filter algorithm based on the battery health status characteristics to obtain a target SOC prediction model; the SOC prediction unit is used to call the target SOC prediction model and calculate based on the non-training battery current to obtain a predicted SOC value.
[0019] To summarize, in an embodiment of the present invention, the battery current is collected in real time, and the SOC value is calculated based on the battery current, wherein the battery current includes the training battery current and the non-training battery current. The battery health status characteristics are obtained based on the SOC value and the gated cycle unit. The predicted SOC value is obtained based on the non-training battery current, the battery health status characteristics and the SOC prediction model based on the Kalman filter algorithm, thereby effectively reducing the impact of noise and error on the prediction results and improving the reliability and stability of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Further details, features and advantages of the present invention are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:
[0021] Figure 1 A flow chart of a method for estimating the SOC of a hybrid battery disclosed in an embodiment of the present application is shown;
[0022] Figure 2 Shown Figure 1The schematic diagram of the step flow of step S120 is shown;
[0023] Figure 3 Shown Figure 1 A schematic diagram of step S130 is shown;
[0024] Figure 4 A schematic structural diagram of an SOC estimation device for a hybrid battery disclosed in an embodiment of the present application is shown;
[0025] Figure 5 A schematic structural diagram of a vehicle disclosed in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0026] Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.
[0027] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.
[0028] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0029] It should be noted that the modifications of "one" and "multiple" mentioned in the present invention are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0030] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes and are not used to limit the scope of these messages or information.
[0031] It should be noted that the execution entity of the hybrid battery SOC estimation method provided in the embodiments of the present invention may be one or more electronic devices, which is not limited in the present invention. Specifically, the electronic device may be a terminal (i.e., a client) or a server. If the execution entity includes multiple electronic devices, and the multiple electronic devices include at least one terminal and at least one server, the hybrid battery SOC estimation method provided in the embodiments of the present invention may be jointly executed by the terminal and the server. Accordingly, the terminals mentioned herein may include, but are not limited to, smartphones, tablets, laptops, desktop computers, smartwatches, intelligent voice interaction devices, smart home appliances, in-vehicle terminals, aircraft, and the like. The servers mentioned herein may be standalone physical servers, server clusters or distributed systems consisting of multiple physical servers, or cloud servers providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDNs (Content Delivery Networks), and basic cloud computing services such as big data and artificial intelligence platforms.
[0032] Based on the above description, embodiments of the present invention provide a method for estimating the SOC of a hybrid battery. This method can be executed by the aforementioned electronic device (terminal or server); alternatively, this method can be executed jointly by the terminal and the server. For ease of illustration, the following description uses the example of an electronic device executing this method for estimating the SOC of a hybrid battery.
[0033] See also Figure 1 , which is a flow chart of a SOC estimation method applicable to a hybrid battery disclosed in an embodiment of the present application. A SOC estimation method applicable to a hybrid battery solves the problem of how to provide a reasonable uncertainty estimation while ensuring the prediction accuracy, thereby improving the accuracy of collision prediction and enhancing the robustness of the system in complex traffic scenarios. It should be noted that a SOC estimation method applicable to a hybrid battery in the embodiment of the present application is not limited to Figure 1 The steps and order in the flowchart shown. According to different needs, the steps in the flowchart shown can be added, removed, or changed in order. In the embodiment of the present application, Figure 1 As shown, the original video frame is input to the SOC estimation model suitable for hybrid batteries (such as Figure 4 As shown), a frame-level risk score is obtained, and the SOC estimation model applicable to the hybrid battery executes the following SOC estimation method applicable to the hybrid battery. A process of the SOC estimation method applicable to the hybrid battery includes at least the following steps.
[0034] S110 , collecting battery current in real time, and calculating an SOC value based on the battery current, wherein the battery current includes a training battery current and a non-training battery current.
[0035] In the embodiment of the present invention, the battery current is collected in real time, and the ampere-hour integration method is used to accumulate the battery's charge and discharge power to obtain the SOC value. The SOC value is expressed as follows:
[0036]
[0037] Among them, SOC(t0) is the initial battery state of charge value, C n is the battery capacity correction value, I(τ) is the battery current, and η is the coulomb efficiency correction value.
[0038] Specifically, the current is integrated to measure the energy charged into or discharged from the battery in real time, and the battery power is recorded and monitored for a long time, which can provide the remaining power of the battery at any time. The current data of the battery during charging and discharging is obtained in real time through the ammeter, and then the current data is integrated in real time. The ampere-hour capacity charged or discharged by the battery is obtained to represent the change in battery capacity over a period of time. Furthermore, the quotient of the ampere-hour capacity and the theoretical available capacity is used as the theoretical SOC change. For example, the theoretical SOC change can be obtained by accumulating the capacity through the ampere-hour integration method during the charging and discharging process.
[0039] S120 : Processing is performed based on the SOC value and the gated cycle unit to obtain a battery health status characteristic.
[0040] like Figure 2 As shown, in an embodiment of the present invention, Figure 2 The step S120 at least includes the following steps:
[0041] S121 . Preprocess the SOC value to obtain a target SOC value.
[0042] In the embodiment of the present invention, the SOC value is preprocessed, specifically, abnormal values in the SOC value are removed and noise is filtered.
[0043] S122 : Perform feature extraction based on the target SOC value and the gated cycle unit to obtain the battery health status feature.
[0044] In an embodiment of the present invention, a gated recurrent unit (GRU) model is called to extract the characteristics of the target battery state of charge value to obtain the battery state of health (SOH) characteristics, where the SOH characteristics may include at least the capacity decay rate, the internal resistance growth rate, and the charge and discharge time. Specifically, the GRU model is used to capture the correlation between different decay characteristics. The capacity decay rate is as follows:
[0045]
[0046] Among them, C0 is the initial capacity of the battery, C t is the remaining capacity of the battery after time t.
[0047] The internal resistance growth rate is as follows:
[0048]
[0049] Among them, R0 is the initial internal resistance, R t is the internal resistance after time t.
[0050] S130 , performing processing based on the non-training battery current, the battery health status characteristics, and an SOC prediction model based on a Kalman filter algorithm to obtain a predicted SOC value.
[0051] like Figure 3 As shown, in an embodiment of the present invention, Figure 3 The step S130 at least includes the following steps:
[0052] S131 . Training an SOC prediction model based on a Kalman filter algorithm based on the battery health status characteristics to obtain a target SOC prediction model.
[0053] In an embodiment of the present invention, the SOC prediction model based on the Kalman filter algorithm is trained based on the battery health status characteristics. Specifically, the initialization state variable is calculated. and error covariance P0, initialize state variables The sum error covariance P0 is as follows:
[0054]
[0055] in, is the estimated value of the initial state variable X0, E() is the mathematical expectation, P0 is the initial error covariance matrix, is the transpose of the error vector.
[0056] The state prediction value is as follows:
[0057]
[0058] in, is the predicted value of the state variable at time k based on the information at time k-1, f(,) is the state transfer function, is the estimated value of SOC at time k-1, u k-1 is the battery current at time k-1.
[0059] The output observation value is as follows:
[0060]
[0061] in, is the predicted value of the system output at time k based on the information at time k-1, and g(,) is the output function.
[0062] For state prediction Perform a Taylor expansion linearization process to obtain the state transfer matrix A K and the observation matrix C K As shown in the following formula:
[0063]
[0064] in, is the partial derivative.
[0065] The error covariance matrix is calculated as follows:
[0066]
[0067] Among them, A K is the Jacobian matrix, P K-1 is the error covariance matrix at the previous moment, is the transpose of the Jacobian matrix, is the process noise covariance.
[0068] The Kalman gain is calculated as follows:
[0069]
[0070] in, is the observation matrix C K The transpose of R K-1 is the observation noise covariance matrix.
[0071] The error vector at time K is calculated as follows:
[0072]
[0073] Among them, H K is the target matrix, Ei is the error vector at the i-th moment, For E i The transpose of .
[0074] The updated state is as follows:
[0075]
[0076] The corrected and updated error covariance matrix is as follows:
[0077] P K =(IL K C K )P K-1 Formula (15)
[0078] The covariance of the corrected and updated adaptive noise and observation noise is as follows:
[0079]
[0080] S132: Call the target SOC prediction model and calculate based on the non-training battery current to obtain a predicted SOC value.
[0081] In an embodiment of the present invention, the target SOC prediction model is called to calculate the non-training battery current to obtain a predicted SOC value.
[0082] In summary, it can be seen that in an SOC estimation method suitable for a hybrid battery of the present application, the battery current is collected in real time, and the SOC value is calculated based on the battery current, wherein the battery current includes a training battery current and a non-training battery current, and is processed based on the SOC value and a gated cycle unit to obtain a battery health status characteristic, and is processed based on the non-training battery current, the battery health status characteristic and an SOC prediction model based on a Kalman filter algorithm to obtain a predicted SOC value, thereby effectively reducing the impact of noise and error on the prediction results and improving the reliability and stability of the prediction.
[0083] See also Figure 4 , which is a schematic diagram of the structure of a SOC estimation device for a hybrid battery disclosed in an embodiment of the present application. In one embodiment, Figure 4 As shown, the present application provides an SOC estimation device 100 for a hybrid battery, which may include at least: a data acquisition module 110, a data processing module 130, and a prediction module 160. Information exchange occurs between the data acquisition module 110 and the data processing module 130, and information exchange occurs between the data processing module 130 and the prediction module 160.
[0084] The data acquisition module 110 is used to collect battery current in real time and calculate the SOC value based on the battery current, wherein the battery current includes the training battery current and the non-training battery current. In an embodiment of the present invention, the battery current is collected in real time, and the ampere-hour integration method is used to accumulate the battery's charge and discharge to obtain the SOC value. The SOC value is expressed as follows:
[0085]
[0086] Among them, SOC(t0) is the initial battery state of charge value, C n is the battery capacity correction value, I(τ) is the battery current, and η is the coulomb efficiency correction value.
[0087] Specifically, the current is integrated to measure the energy charged into or discharged from the battery in real time, and the battery power is recorded and monitored for a long time, which can provide the remaining power of the battery at any time. The current data of the battery during charging and discharging is obtained in real time through the ammeter, and then the current data is integrated in real time. The ampere-hour capacity charged or discharged by the battery is obtained to represent the change in battery capacity over a period of time. Furthermore, the quotient of the ampere-hour capacity and the theoretical available capacity is used as the theoretical SOC change. For example, the theoretical SOC change can be obtained by accumulating the capacity through the ampere-hour integration method during the charging and discharging process.
[0088] The data processing module 130 is used to process the SOC value and the gated cycle unit to obtain the battery health status characteristics. The data processing module 130 may include at least a pre-processing unit 131 and a feature extraction unit 133.
[0089] The preprocessing unit 131 is used to preprocess the SOC value to obtain a target SOC value. In the embodiment of the present invention, the preprocessing of the SOC value specifically includes removing abnormal values in the SOC value and filtering noise.
[0090] The feature extraction unit 133 is used to extract features based on the target SOC value and the gated recurrent unit to obtain the battery health status feature. In an embodiment of the present invention, the gated recurrent unit (GRU) model is called to extract features of the target battery state of charge value to obtain the battery health status (SOH) feature, wherein the SOH feature may at least include capacity decay rate, internal resistance growth rate, and charge and discharge time. Specifically, the GRU model is used to capture the correlation between different decay features, and the capacity decay rate is as follows:
[0091]
[0092] Among them, C0 is the initial capacity of the battery, C t is the remaining capacity of the battery after time t.
[0093] The internal resistance growth rate is as follows:
[0094]
[0095] Among them, R0 is the initial internal resistance, R t is the internal resistance after time t.
[0096] The prediction module 160 is configured to process the non-trained battery current, the battery health status characteristics, and the SOC prediction model based on the Kalman filter algorithm to obtain a predicted SOC value. The prediction module 160 may include at least a model training unit 161 and an SOC prediction unit 163.
[0097] The model training unit 161 is used to train the SOC prediction model based on the Kalman filter algorithm based on the battery health status characteristics to obtain a target SOC prediction model. In an embodiment of the present invention, the SOC prediction model based on the Kalman filter algorithm is trained based on the battery health status characteristics. Specifically, the initialization state variable is calculated. and error covariance P0, initialize state variables The sum error covariance P0 is as follows:
[0098]
[0099] in, is the estimated value of the initial state variable X0, E() is the mathematical expectation, P0 is the initial error covariance matrix, is the transpose of the error vector.
[0100] The state prediction value is as follows:
[0101]
[0102] in, is the predicted value of the state variable at time k based on the information at time k-1, f(,) is the state transfer function, is the estimated value of SOC at time k-1, u k-1 is the battery current at time k-1.
[0103] The output observation value is as follows:
[0104]
[0105] in, is the predicted value of the system output at time k based on the information at time k-1, and g(,) is the output function.
[0106] For state prediction Perform a Taylor expansion linearization process to obtain the state transfer matrix A K and the observation matrix C K As shown in the following formula:
[0107]
[0108] in, is the partial derivative.
[0109] The error covariance matrix is calculated as follows:
[0110]
[0111] Among them, A K is the Jacobian matrix, P K-1 is the error covariance matrix at the previous moment, is the transpose of the Jacobian matrix, is the process noise covariance.
[0112] The Kalman gain is calculated as follows:
[0113]
[0114] in, is the observation matrix C K The transpose of R K-1 is the observation noise covariance matrix.
[0115] The error vector at time K is calculated as follows:
[0116]
[0117] Among them, H K is the target matrix, E i is the error vector at the i-th moment, For E i The transpose of .
[0118] The updated state is as follows:
[0119]
[0120] The corrected and updated error covariance matrix is as follows:
[0121] P K =(IL K C K )P K-1Formula (15)
[0122] The covariance of the corrected and updated adaptive noise and observation noise is as follows:
[0123]
[0124] The SOC prediction unit 163 is used to call the target SOC prediction model and calculate the predicted SOC value based on the non-training battery current. In an embodiment of the present invention, the target SOC prediction model is called to calculate the non-training battery current to obtain the predicted SOC value.
[0125] In summary, in an SOC estimation device suitable for a hybrid battery of the present application, the battery current is collected in real time by the data acquisition module 110, and the SOC value is calculated based on the battery current, wherein the battery current includes a training battery current and a non-training battery current. The data processing module 130 performs processing based on the SOC value and the gated cycle unit to obtain the battery health status characteristics. The prediction module 160 performs processing based on the non-training battery current, the battery health status characteristics and the SOC prediction model based on the Kalman filter algorithm to obtain a predicted SOC value, thereby effectively reducing the impact of noise and error on the prediction results and improving the reliability and stability of the prediction.
[0126] See also Figure 5 , which is a structural diagram of a vehicle disclosed in an embodiment of the present application. The present application also provides a vehicle 10, which includes Figure 4 The SOC estimation device 100 in the illustrated embodiment is applicable to a hybrid battery.
[0127] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "example," "specific example," "one implementation," "a preferred implementation," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
[0128] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the claims and their equivalents.
Claims
1. A method for estimating the SOC of a hybrid battery, characterized in that: The SOC estimation method applicable to a hybrid battery includes: Collecting battery current in real time and calculating an SOC value based on the battery current, wherein the battery current includes a training battery current and a non-training battery current; Processing based on the SOC value and the gated cycle unit to obtain a battery health status characteristic; Processing is performed based on the non-training battery current, the battery health status characteristics and an SOC prediction model based on a Kalman filter algorithm to obtain a predicted SOC value.
2. The SOC estimation method for a hybrid battery according to claim 1, characterized in that: The processing based on the SOC value and the gated cycle unit to obtain the battery health status characteristics includes: Preprocessing the SOC value to obtain a target SOC value; Feature extraction is performed based on the target SOC value and the gated cycle unit to obtain the battery health status feature.
3. The SOC estimation method for a hybrid battery according to claim 1, characterized in that: The processing based on the non-training battery current, the battery health status characteristics and the SOC prediction model based on the Kalman filter algorithm to obtain the predicted SOC value includes: Training an SOC prediction model based on a Kalman filter algorithm based on the battery health status characteristics to obtain a target SOC prediction model; The target SOC prediction model is called to calculate and obtain a predicted SOC value based on the non-training battery current.
4. The SOC estimation method for a hybrid battery according to claim 3, characterized in that: The SOC value is calculated based on the battery current, and the SOC value is: Among them, SOC(t0) is the initial battery state of charge value, C n is the battery capacity correction value, I(τ) is the battery current, and η is the coulomb efficiency correction value.
5. The SOC estimation method for a hybrid battery according to claim 4, characterized in that: The processing is performed based on the SOC value and the gated cycle unit to obtain battery health status characteristics, and the battery health status characteristics include capacity decay rate, internal resistance growth rate and charge and discharge time.
6. The SOC estimation method for a hybrid battery according to claim 5, characterized in that: The capacity decay rate is: Among them, C0 is the initial capacity of the battery, C t is the remaining capacity of the battery after time t.
7. A SOC estimation device for a hybrid battery, characterized in that: The SOC estimation device for a hybrid battery includes: a data acquisition module, a data processing module and a prediction module, wherein: The data acquisition module is used to collect battery current in real time and calculate the SOC value based on the battery current, wherein the battery current includes the training battery current and the non-training battery current; The data processing module is used to process the SOC value and the gated cycle unit to obtain the battery health status characteristics; The prediction module is used to process the non-training battery current, the battery health status characteristics and the SOC prediction model based on the Kalman filter algorithm to obtain a predicted SOC value.
8. The SOC estimation device for a hybrid battery according to claim 7, characterized in that: The data processing module includes: a pre-processing unit and a feature extraction unit, The preprocessing unit is used to preprocess the SOC value to obtain a target SOC value; The feature extraction unit is used to perform feature extraction based on the target SOC value and the gated cycle unit to obtain the battery health status feature.
9. The SOC estimation device for a hybrid battery according to claim 8, characterized in that: The prediction module includes a model training unit and an SOC prediction unit, wherein: The model training unit is used to train the SOC prediction model based on the Kalman filter algorithm based on the battery health state characteristics to obtain a target SOC prediction model; The SOC prediction unit is used to call the target SOC prediction model and calculate based on the non-training battery current to obtain a predicted SOC value.