Battery pack energy state estimation method, device, apparatus, storage medium and product

Abstract

The application discloses a battery pack energy state estimation method, device, equipment, storage medium and product, which are used for accurately and reliably estimating the energy state of a battery pack. The method comprises the following steps: acquiring a first operation data sequence of a first battery pack, wherein the first operation data sequence comprises operation data of the first battery pack at multiple time points; inputting the first operation data sequence into a first model to output a first energy state of the first battery pack; wherein the first model is obtained by training based on a second operation data sequence and a second energy state of a second battery pack, the single cell parameter distribution of the second battery pack is similar to the single cell parameter distribution of the first battery pack, and the single cell parameter distribution of the battery pack is obtained by a second model based on an operation data sequence of the battery pack.

Description

Technical Field

[0001] This application relates to the field of battery technology, and in particular to a method, apparatus, device, storage medium and product for estimating the state of energy of a battery pack. Background Technology

[0002] In electrochemical energy storage systems, accurate estimation of the battery pack's State of Energy (SOE) is a core technological element for achieving efficient system management and reliable operation. The SOE reflects the remaining energy that the battery pack can release under its current condition. This indicator is crucial for assessing the real-time dispatch capability of the energy storage system and provides a key basis for formulating scientific and rational energy management strategies. However, current technological bottlenecks remain in battery pack SOE estimation, lacking a scheme that can accurately and reliably estimate the battery pack's SOE. Summary of the Invention

[0003] The purpose of this application is to provide a method, apparatus, device, storage medium, and product for estimating the state of energy of a battery pack, for accurately and reliably estimating the state of energy of the battery pack.

[0004] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: In a first aspect, embodiments of this application provide a method for estimating the state of energy of a battery pack, including: Obtain a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; Input the first running data sequence into the first model and output the first energy state of the first battery pack; The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

[0005] Secondly, embodiments of this application provide a battery pack state of energy estimation device, comprising: The first acquisition module is used to acquire a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; The input module is used to input the first running data sequence into the first model and output the first energy state of the first battery pack. The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

[0006] Thirdly, embodiments of this application provide an electronic device, a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the battery pack state of energy estimation method provided in the first aspect.

[0007] Fourthly, embodiments of this application provide a computer-readable storage medium that, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the steps of the battery pack state of energy estimation method provided in the first aspect.

[0008] Fifthly, embodiments of this application provide a computer program product, the computer program product including a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform steps of the battery pack state of energy estimation method provided in the first aspect.

[0009] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: The individual cell parameter distribution of the battery pack is used to quantify the inconsistencies between individual cells within the battery pack. The individual cell parameter distribution of the second battery pack is similar to that of the first battery pack, meaning that the second and first battery packs have similar individual cell inconsistencies, i.e., the differences between their individual cells are similar. Therefore, by training the first model based on the second operating data sequence and the second state of energy (SGE) of the second battery pack, the individual cell inconsistencies within the first battery pack are introduced into the estimation process. This allows the first model to deeply learn the impact of these inconsistencies on the SGE estimation, thereby eliminating their influence and accurately and reliably predicting the SGE of the first battery pack based on the first operating data sequence.

[0010] Furthermore, given that the distribution of individual parameters of a battery pack is difficult to obtain through testing, the second model learns from the overall external characteristics of the battery pack's operating data sequence and accurately infers the distribution of individual parameters, thus achieving precise quantification of this inconsistency and providing data support for accurate and reliable prediction of the battery pack's state of energy. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of an example environment in which embodiments of this application can be implemented; Figure 2 A flowchart illustrating a battery pack state of energy estimation method provided in one embodiment; Figure 3 A schematic diagram of the structure of a second battery pack provided in one embodiment; Figure 4 A schematic diagram of the equivalent circuit model of a second battery pack provided in one embodiment; Figure 5 A schematic diagram of a battery pack state of energy estimation device provided in one embodiment; Figure 6 This is a schematic diagram of the structure of an electronic device provided in one embodiment. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] The term "comprising" and its variations as used in this document are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the following description. The term "in response to" indicates that the performed operation depends on a condition or state. When the dependent condition or state is met, one or more operations may be performed in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which multiple operations are performed.

[0014] It should be noted that the concepts of "first" and "second" mentioned in this document are used only to distinguish different devices, modules or units, and are not used to restrict the order of functions performed by these devices, modules or units or their interdependencies.

[0015] It should be noted that the terms "one" and "more" used in this document are illustrative rather than restrictive, and those skilled in the art should understand that, unless explicitly stated otherwise in the context, they should be understood as "one or more".

[0016] The names of messages or information exchanged between multiple devices in this document are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0017] It should be understood that the training and prediction processes of the artificial intelligence (AI) models involved in the various embodiments of this specification all adhere to multiple legal and compliant principles, including legal data sources, compliant data content, compliant data governance, compliant training objectives and schemes, compliant training processes, compliant training environments and tools, and compliant ethical verification of training results, and comply with the requirements of Article 5 of the Patent Law. Among them: Data source legitimacy: All datasets used for AI model training were obtained through legal means, covering three categories: publicly authorized data, data authorized by partners, and self-collected compliant data. Publicly authorized data comes from compliant data sources following open-source licenses such as Apache 2.0, with complete copyright attribution and authorization scope clearly marked, and no unauthorized open-source code or data reuse. Data authorized by partners has been subject to formal data usage agreements, clearly defining the scope, duration, and confidentiality obligations, and possessing a complete authorization chain. For self-collected data involving personal information, strict informed consent procedures have been followed, and anonymization processes (including but not limited to field masking, feature anonymization, and differential privacy technology applications) have been used to remove personally identifiable information, fully complying with the requirements of the "Interim Measures for the Administration of Generative Artificial Intelligence Services," the "Personal Information Protection Law," and other relevant laws and regulations.

[0018] Data content compliance: The AI ​​model's dataset undergoes multiple screenings and cleaning processes to remove all content that may violate social morality or harm public interests. It contains no obscene, pornographic, violent, discriminatory, or information that endangers national or public safety, nor does it involve the illegal acquisition or use of genetic resources. For data in sensitive fields (such as healthcare and finance), an additional privacy-preserving computation module (including federated learning and secure multi-party computation technologies) ensures that the data is "usable but not visible," avoiding compliance risks during the original data transmission process and ensuring that the data application scenarios and uses comply with public order and good morals and industry regulatory requirements.

[0019] Data governance norms: A complete data traceability system is established during the AI ​​model training process to automatically record the source, collection time, annotation process, cleaning rules, and permission allocation of training data, generating traceable compliance reports to ensure that the data is verifiable throughout its entire lifecycle. The dataset annotation process for AI models is completed by a professional human R&D team, clearly defining the proportion of human creative contributions and avoiding reliance on AI-generated data that has not undergone substantial human modification, thus meeting the examination requirements for "human main contributions" in AI patent applications.

[0020] Training objectives and plans are compliant: The AI ​​model training objectives are focused on engineering design and production scenarios. The training scheme and final output results do not violate any mandatory provisions of laws and administrative regulations, do not harm the public interest or the legitimate rights and interests of others, and do not pose any potential risks of being used for illegal activities, infringing on privacy, or disrupting public safety. The training strictly adheres to the ethical principle of "intelligent for good".

[0021] Training process compliance: A closed-loop training framework is adopted to ensure compliance and controllability of the training process. The specific process is as follows: First, training samples are obtained through compliant data sources. After the aforementioned data cleaning and desensitization, they are input into the neural network model to generate preliminary training results. Second, an expert system is introduced to verify the preliminary results. Based on preset rules and human expert experience, the feasibility of the results is evaluated, and outputs that may pose ethical risks or compliance hazards are corrected (such as removing decision-making logic that violates public order and good morals, and adjusting model parameters that do not comply with safety regulations). Finally, the loss function weights are dynamically optimized based on expert system feedback to strengthen the model's learning of compliant results, avoid overfitting errors or non-compliant labels, and form a closed-loop control of "data input - model training - expert verification - parameter optimization - result feedback" to ensure that the entire training process complies with A5 ethical review requirements.

[0022] Training environment and tool compliance: AI model training is implemented using nationally licensed chips and a compliant training platform. All open-source frameworks and components used in the training process have obtained their corresponding licenses, and copyright statements and patent citation information are fully retained, with no instances of infringement or reuse. The training environment is built using virtual devices (containers / virtual machines) with fixed random seeds and initial parameter configurations to ensure the reproducibility of the training process. Furthermore, through access control and operation log recording, risks such as data leakage and parameter tampering during training are prevented, ensuring the security and compliance of the training process.

[0023] Training results ethical verification compliance: After the model is trained, it undergoes additional third-party ethical compliance assessment and algorithm filing review to verify that the model output does not violate social morality or harm public interests. For potentially sensitive scenarios (such as public services and intelligent decision-making), a special result verification mechanism is established to ensure that the model always complies with Article 5 of the Patent Law and relevant laws and regulations in practical applications.

[0024] In summary, the data and training process used in the AI ​​model of this specification strictly comply with the relevant provisions of Article 5 of the Patent Law and the Patent Examination Guidelines. There are no violations of laws, social ethics, public interests, or illegal use of genetic resources, and the model fully meets the compliance requirements for patent authorization.

[0025] The energy state of a battery pack reflects the remaining energy that can be released under its current condition. This indicator is crucial for assessing the real-time dispatch capability of an energy storage system and also provides a key basis for formulating scientific and reasonable energy management strategies.

[0026] Compared to individual cells, estimating the state of energy (SGE) of battery packs presents greater challenges. Effective SGE methods and models developed and validated for individual cells cannot be directly and simply transferred to SGE estimation for battery packs due to the unique structural characteristics and complex operating mechanisms of battery packs.

[0027] Some related technologies attempt to evaluate the overall state of energy (SGE) of a battery pack by selecting representative individual cells from the pack and using their parameters as a benchmark. However, such methods have certain drawbacks. Specifically, the accuracy of SGE estimation is highly dependent on the consistency between individual cells. When differences exist between individual cells, ignoring these differences during the estimation process will introduce estimation bias, severely weakening the accuracy and reliability of the estimation results. Especially when facing complex operating conditions such as uneven aging of individual cells, significant differences in temperature field distribution, and drastic dynamic changes in operating conditions, the representative individual cells determined according to existing selection rules often fail to accurately reflect the actual SGE of the battery pack, leading to significant estimation errors.

[0028] To overcome the aforementioned technical bottlenecks, the inconsistencies between individual cells are proactively introduced and quantified during the estimation process. These inconsistencies are then used as constraints to generate training data that more closely reflects the actual state of the battery pack. Based on this training data, the first model (i.e., the model used to estimate the state of energy) is trained, enabling the first model to effectively eliminate the negative impact of inconsistencies. Even under complex and variable inconsistency conditions, the battery pack's state of energy estimation can achieve higher accuracy and stronger robustness.

[0029] It should be understood that the battery pack state of energy estimation method provided in the embodiments of this specification can be executed by an electronic device. The electronic device referred to herein can include any type of mobile terminal, fixed terminal, or portable terminal, specifically including but not limited to: smartphones, tablets, laptops, desktop computers, smart voice interaction devices, smart wearable devices, etc.; or, the electronic device can also include a server, such as a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0030] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0031] Figure 1 A schematic diagram of an example environment in which embodiments of this application can be implemented is shown. This example environment includes a terminal device 100 and a server 200.

[0032] The terminal device 100 and the server 200 establish a communication connection, which may include, but is not limited to, at least one of the following: wired connection and wireless connection.

[0033] Terminal device 100 includes, but is not limited to, smartphones, tablets, laptops, desktop computers, smart voice interaction devices, smart home appliances, smartwatches, vehicle terminals, and aircraft. Server 200 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0034] Terminal device 100 can receive relevant parameters input by the user and send these parameters to server 200. Server 200 can obtain the processing result based on the received parameters and return the processing result to terminal device 100.

[0035] It should be understood that in some embodiments, the terminal device 100 may obtain the processing result itself based on the relevant parameters input by the user, without needing to interact with the server 200. This application embodiment does not limit this.

[0036] Based on the example environment described above, this application provides a method for estimating the state of energy of a battery pack. Please refer to... Figure 2 This is a flowchart illustrating a battery pack energy state estimation method provided in an embodiment of this application. The method includes the following steps: S202, Obtain the first operating data sequence of the first battery pack.

[0037] The first battery pack can be understood as the battery pack to which the state of energy estimation operation is to be performed, and it can be a real, existing battery pack entity. The first battery pack may include multiple individual cells. These individual cells can be connected in various ways, and there is no limitation on this. In some embodiments, these individual cells are connected in series.

[0038] The first operational data sequence includes operational data of the first battery pack at multiple points in time. These multiple points in time can refer to multiple specific time nodes prior to the current time.

[0039] Battery pack operating data at a given moment refers to data that reflects the overall operating status of the battery pack at that moment. For example, operating data may include, but is not limited to, at least one of the following: total voltage, maximum single-cell voltage, minimum single-cell voltage, current, etc. The maximum single-cell voltage can be understood as the highest value among the terminal voltages of all individual cells in the battery pack, and the minimum single-cell voltage can be understood as the lowest value among the terminal voltages of all individual cells in the battery pack.

[0040] Accordingly, the first operating data sequence may include, but is not limited to, at least one of the following: the total voltage sequence of the first battery pack, the maximum single-cell voltage sequence, the minimum single-cell voltage sequence, and the current sequence. The total voltage sequence includes the total voltage at multiple times, the maximum single-cell voltage sequence includes the maximum single-cell voltage at multiple times, the minimum single-cell voltage sequence includes the minimum single-cell voltage at multiple times, and the current sequence includes the current at multiple times.

[0041] The first operational data sequence can be obtained by testing the first battery pack. Specific testing methods can be adopted using existing technologies and will not be elaborated further.

[0042] S204, input the first running data sequence into the first model and output the first energy state of the first battery pack.

[0043] The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual cell parameter distribution of the second battery pack is similar to that of the first battery pack. The individual cell parameter distribution is predicted by the second model based on the operating data sequence of the battery packs. That is, the individual cell parameter distribution of the first battery pack is predicted by the second model based on the first operating data sequence, and the individual cell parameter distribution of the second battery pack is predicted by the second model based on the second operating data sequence.

[0044] Specifically, the second running data sequence can be input into the first model for prediction, and the parameters of the first model can be adjusted based on the difference between the energy state output by the first model and the second energy state. In this way, the first model can output the corresponding energy state based on any input running data sequence.

[0045] In practical applications, the structure of the second battery pack can be similar to that of the first battery pack. For example, if the first battery pack consists of n individual cells connected in series, the second battery pack also consists of n individual cells connected in series. The second battery pack can be an actual, physical battery pack entity, or it can be a virtualized battery pack constructed based on an equivalent circuit model.

[0046] The second operating data sequence includes operating data of the second battery pack at multiple times. In some embodiments, the second operating data sequence may include, but is not limited to, at least one of the following: the total voltage sequence of the second battery pack, the maximum single-cell voltage sequence, the minimum single-cell voltage sequence, the current sequence, etc. The second operating data sequence can be obtained by testing the second battery pack or by simulating an equivalent circuit model of the second battery pack.

[0047] The second energy state corresponds to the second operating data. The second energy state can be obtained through simulation of the equivalent circuit model of the second battery pack.

[0048] Each battery pack has a corresponding individual cell parameter distribution. Specifically, the individual cell parameter distribution of the battery pack encompasses the parameter distribution of the individual cells within the battery pack. This distribution is a quantitative representation of the inconsistencies (i.e., differences) between the individual cells within the battery pack. In some embodiments, the individual cell parameter distribution of the battery pack may include the mean and standard deviation of the individual cell parameters of all individual cells within the battery pack, the correlation coefficient between individual cell parameters, etc. The individual cell parameters may include, but are not limited to, at least one of the following performance parameters: capacity, state of charge (SOC), internal resistance, etc.

[0049] In this embodiment, the individual cell parameter distribution of the battery pack is used to quantify the inconsistency between individual cells within the battery pack. The individual cell parameter distribution of the second battery pack is similar to that of the first battery pack, meaning that the second and first battery packs have similar inconsistencies between individual cells, i.e., the differences between their individual cells are similar. Therefore, by training the first model based on the second operating data sequence and the second state of energy (SGE) of the second battery pack, the inconsistency between individual cells within the first battery pack is introduced into the estimation process. This allows the first model to deeply learn the impact of such inconsistencies on SGE estimation, thereby eliminating the impact of these inconsistencies on SGE estimation and accurately and reliably predicting the SGE of the first battery pack based on the first operating data sequence.

[0050] Furthermore, due to performance differences among individual cells within a battery pack, the resulting "weakest link" effect makes it difficult to accurately obtain the parameter distribution of individual cells using only overall charge-discharge testing. Moreover, battery packs typically consist of numerous individual cells; conducting charge-discharge tests on each cell individually to obtain its parameters would significantly extend the testing cycle. Therefore, obtaining the individual parameter distribution of a battery pack using conventional testing methods faces considerable challenges. In light of this, by using a second model to learn from the overall external characteristics of the battery pack's operational data sequence and accurately infer the individual parameter distribution, this inconsistency can be precisely quantified, providing data support for accurate and reliable prediction of the battery pack's state of energy.

[0051] In the embodiments of this application, the second operating data sequence and the second energy state of the second battery pack can be obtained in various ways, and there is no limitation thereto.

[0052] In some embodiments, the second running data sequence and the second energy state can be obtained in the following manner: S302, based on the individual parameter distribution of the first battery pack, determine the state space corresponding to the second battery pack.

[0053] The state space is used to represent the constraints on the individual cell parameters of the second battery pack. These constraints can include the following two types: Statistical constraints: Individual cell parameters must conform to the individual cell parameter distribution of the first battery pack. For example, the deviation between the capacity and the mean must be within a preset range, and the internal resistance variance must be consistent with the resistance variance of the individual cells in the first battery pack. Physical constraints: Individual cell parameters must conform to the physical laws of individual cells, such as capacity > 0, internal resistance > 0, and SOC between 0 and 100%.

[0054] For example, the first battery pack consists of 100 individual cells, and the individual cell parameters of the first battery pack are distributed as follows: Capacity distribution: mean 30 Ah, standard deviation 1 Ah; Internal resistance distribution: mean 5mΩ, standard deviation 0.5mΩ; Covariance between capacitance and internal resistance: 0.2.

[0055] Based on the distribution of the individual parameters, the constraints can be determined as follows: capacity ~ N(30,1), internal resistance ~ (5, 0.25), covariance between capacity and internal resistance 0.2, and capacity > 0 and inner group > 0.

[0056] S304, sampling is performed in the state space to obtain the individual cell parameters of the second battery pack.

[0057] Monte Carlo sampling can be used to randomly generate multiple sets of individual cell parameters similar to the distribution of individual cell parameters in the first battery pack, under constraints. Each set of individual cell parameters corresponds to a virtual battery pack (i.e., the second battery pack), which includes the values ​​of multiple characteristic parameters for each individual cell within the battery pack. This can be used to simulate the inconsistencies between individual cells within the battery pack. These multiple characteristic parameters may include, but are not limited to, capacity, internal resistance, and state of charge (SOC).

[0058] For example, suppose the first battery pack has 100 individual cells, and 1000 sets of individual cell parameters are randomly selected. Each set of individual cell parameters includes the values ​​of characteristic parameters such as capacity, internal resistance, and SOC of 100 individual cells. That is, each set of individual cell parameters can include a ternary group of 100 individual cells <capacity, SOC, internal resistance>.

[0059] S306, based on the individual cell parameters of the second battery pack and the equivalent circuit model of the second battery pack, simulation is performed to obtain the second operating data sequence and the second energy state of the second battery pack.

[0060] The equivalent circuit model of the second battery pack is a circuit composed of basic circuit components such as voltage sources, resistors, and capacitors, used to simulate the external electrical characteristics of the second battery pack.

[0061] The individual cell parameters of the second battery pack are assigned to the basic circuit elements in its equivalent circuit model. For example, based on the internal resistance and capacity of each individual cell in the second battery pack, the resistance, capacitance, etc. in the equivalent circuit model are assigned values ​​to form a virtual battery pack that is equivalent to the second battery pack. Then, a "charge and discharge current sequence" (such as constant current charging and dynamic operating condition discharging) is input into the assigned equivalent circuit model to calculate the individual cell voltage, total voltage, maximum individual cell voltage, minimum individual cell voltage, current, etc. at each moment, thus obtaining the second operating data sequence of the second battery pack.

[0062] For example, such as Figure 3 As shown, the second battery pack comprises n individual cells connected in series. Accordingly, as... Figure 4 As shown, the equivalent circuit model of the second battery pack may include n first-order RC equivalent circuit models 40 connected in series, each first-order RC equivalent circuit model representing a single cell.

[0063] Based on this, the terminal voltage of a single cell at each moment is The total voltage is Furthermore, based on the terminal voltage of each individual cell at each moment, the maximum and minimum individual cell voltages at each moment can be determined. This indicates the open-circuit voltage of a single cell. Indicates the state of charge of a single cell. The curve representing the relationship between the open circuit voltage (OCV) and the state of charge (i.e., the OCV-SOC curve) of the first-order RC equivalent circuit model corresponding to a single cell. Represents current. This indicates the internal resistance of a single cell. The polarization voltage of a single cell is represented by the polarization resistance. Polarized capacitors And determined by the current integral.

[0064] The second state of energy refers to the ratio of the remaining usable energy of the second battery pack to the total usable energy. The second state of energy can be calculated as follows: First, for each individual cell, the total energy from full charge (SOC=100%) to discharge (SOC=0%) is the area integral under the OCV curve: ,in, This indicates the capacity of a single battery cell; further, the total energy of the second battery pack is... .

[0065] Secondly, for each individual cell, since the state of charge (SOC) of the individual cell is known, the remaining energy of the individual cell is... ,in, This represents the State of Energy (SOC) of a single cell. Since the state of energy of the battery pack is not the average of the SOCs of all individual cells, but rather limited by the worst-performing cell, the SOE of each cell within the second battery is calculated. .

[0066] Finally, the second energy state is .

[0067] Understandably, obtaining the state of energy (SGE) of a real battery pack is difficult and costly, especially for samples with specific consistency distribution characteristics (such as those inconsistent with the first battery pack), where measured data is extremely scarce and may be inaccurate. Therefore, this approach first uses a second model to accurately predict the individual cell parameter distribution of the first battery pack. Then, using this distribution as a priori constraint, it generates "individual cell parameters within a virtual battery pack" that conform to this distribution. Finally, simulation is performed based on the equivalent circuit model of the first battery pack, outputting the corresponding second operating data sequence and SGE, forming a dataset for training the first model. This allows for the efficient and low-cost provision of massive, full-lifecycle training samples with consistent physical characteristics for the first model without the need for destructive or long-term physical testing. This effectively overcomes the bottleneck of measured data, ensures the comprehensiveness and diversity of training samples, and significantly improves the generalization ability and estimation accuracy of the first model.

[0068] The equivalent circuit model of the second battery pack can be constructed in various ways, and the embodiments of this application do not limit this.

[0069] In some embodiments, the second battery pack includes multiple individual cells, and the equivalent circuit model of the second battery pack can be obtained by: constructing an equivalent circuit model of each individual cell based on the structure of each individual cell in the second battery pack; and connecting the equivalent circuit models of each individual cell based on the connection method of the multiple individual cells to obtain the equivalent circuit model of the second battery pack.

[0070] For example, the individual cells in the second battery pack are connected in series, and the equivalent circuit models of all the individual cells are connected in series in sequence to obtain the equivalent circuit model of the second battery pack.

[0071] The equivalent circuit model for each individual battery cell can employ various battery dynamic models commonly used in engineering. For example, given that the first-order RC equivalent circuit model is a commonly used battery dynamic model in engineering and achieves a good balance between computational complexity and accuracy, the equivalent circuit model for each individual battery cell can include a first-order RC equivalent circuit model. In the first-order RC equivalent circuit model, the resistance R0 represents the internal resistance of the individual battery cell, the capacitance Cp represents the polarization capacitance of the individual battery cell, and the resistance Rp represents the polarization capacitance of the individual battery cell.

[0072] By constructing the equivalent circuit model of the second battery pack using the above-mentioned "individual cell modeling and topology connection" method, the second battery pack can be accurately reproduced. This model can clearly characterize the nonlinear impact of inconsistencies between individual cells on the overall performance of the battery pack. As a result, the generated virtual dataset not only contains macroscopic operating data sequences, but also implicitly contains microscopic individual cell parameter distributions.

[0073] In this embodiment, the first model can employ various artificial intelligence models with time-series data processing and prediction capabilities, such as neural network models, and is not limited thereto. The first model can output the corresponding energy state based on any input sequence of running data.

[0074] In some embodiments, the first model may include a two-layer fully connected neural network. This architecture has fewer model parameters, lower computational complexity, and is easy to implement for hardware embedded deployment and upgrades.

[0075] Optionally, the first model can be trained offline and then deployed to reduce computational complexity and improve computational efficiency.

[0076] In this embodiment, the second model can employ various artificial intelligence models with time-series data processing and prediction capabilities, such as neural network models, and is not limited thereto. The second model can output a corresponding individual parameter distribution based on any input sequence of running data.

[0077] For example, the input to the second model is The output is The inputs include the total voltage sequence, maximum single-cell voltage sequence, minimum single-cell voltage sequence, and charging current sequence of the battery pack during the charging segment. This represents the average capacity of the individual cells in the battery pack. This represents the average SOC of each cell in the battery pack. This represents the average internal resistance of a single cell in the battery pack. This represents the standard deviation of the individual cell capacity of the battery pack. This represents the standard deviation of the state of charge (SOC) of a single cell in the battery pack. This represents the standard deviation of the internal resistance of each cell in the battery pack. This represents the correlation coefficient between the cell capacity, cell internal resistance, and cell state of charge (SOC).

[0078] In some embodiments, the second model may include a first feature extraction network, a second feature extraction network, a fusion network, and a prediction network. The first feature extraction network is used to extract spatial features of the battery pack's operating data sequence, and the second feature extraction network is used to extract temporal features of the operating data sequence; the fusion network is used to fuse the spatial and temporal features to obtain fused features; and the prediction network predicts the energy state of the battery pack based on the fused features.

[0079] For example, the first feature extraction network may include a convolutional neural network, and the second feature extraction network may include a long short-term memory network. The fusion network may employ a cross-attention mechanism to achieve feature fusion.

[0080] In this case, the individual parameter distribution of the first battery pack can be obtained as follows: extract the spatial and temporal features of the first running data sequence; fuse the spatial and temporal features to obtain fused features; and make predictions based on the fused features to obtain the individual parameter distribution of the first battery pack.

[0081] It is evident that by extracting spatial features, the imbalance within the battery pack (i.e., the inconsistency between individual cells) can be captured. The aging of the battery pack is a dynamic process, and the current state is often strongly correlated with the historical state. By extracting event features, the dynamic evolution pattern of the battery pack can be captured. Furthermore, through the dual-dimensional fusion analysis of "space + time", the inconsistency between individual cells can be more accurately quantified into the distribution of individual cells in the battery pack.

[0082] In this embodiment of the application, the second model can be trained in various ways, and no limitation is made thereto.

[0083] In some embodiments, the second model is trained based on a third operating data sequence of multiple third battery packs and the individual cell parameter distributions. The individual cell parameter distributions of the multiple third battery packs are different.

[0084] The third battery pack here can be an actual, real battery pack entity, or a virtual battery pack constructed based on an equivalent circuit model.

[0085] The third operating data sequence for the third battery pack may include operating data of the third battery pack at multiple times. In some embodiments, the third operating data sequence may include, but is not limited to, at least one of the following: the total voltage sequence of the third battery pack, the maximum single-cell voltage sequence, the minimum single-cell voltage sequence, the current sequence, etc. The third operating data sequence can be obtained by testing the third battery pack or by simulating an equivalent circuit model of the third battery pack.

[0086] During the training of the second model, the third running data sequence of each third battery pack can be input into the second model for prediction, and the parameters of the second model can be adjusted based on the difference between the individual parameter distribution output by the second model and the individual parameter distribution of each third battery pack.

[0087] It is evident that the different distributions of individual parameters across multiple third battery packs indicate varying inconsistencies among the individual cells within these third battery packs. Training the second model based on the third operational data sequences and individual cell distribution parameters of these third battery packs allows the second model to learn and infer the individual cell parameter distributions of the battery pack from the overall external characteristics of the battery pack (i.e., the third operational data sequences), rather than relying on the selection of a single representative cell. This enables the second model to adapt to different levels of inconsistency and exhibit stronger robustness.

[0088] In practical applications, the second model can be trained offline and then deployed to reduce computational complexity and improve computational efficiency.

[0089] The third operating data sequence and individual cell parameter distribution of the aforementioned multiple third battery packs can be obtained in various ways, and this application embodiment does not limit this.

[0090] In some embodiments, the third operating data sequence and individual cell parameter distribution of the above-mentioned plurality of third battery packs can be obtained in the following manner: S402, based on the preset single-cell parameter distribution model, samples to obtain the single-cell parameters of each third battery pack.

[0091] The single-cell parameter distribution model describes the constraints that the single-cell parameters of each third battery pack must follow.

[0092] In some implementations, the individual cell parameters of each third battery pack include the values ​​of multiple characteristic parameters for each individual cell in that third battery pack. For example, these multiple characteristic parameters may include, but are not limited to, internal capacity, capacitance, and state of charge (SOC).

[0093] In this case, the single-parameter distribution model can include a multivariate Gaussian distribution model. The random variables in the multivariate Gaussian distribution model include the aforementioned multiple feature parameters. The mean vector in the multivariate Gaussian distribution model includes the mean of each feature parameter. The covariance matrix in the multivariate Gaussian distribution model includes the correlation coefficients between multiple feature parameters.

[0094] For example, the functional form of the multivariate Gaussian distribution model is as follows:

[0095] in, This represents a random variable, which includes multiple characteristic parameters; This represents the mean vector, which can include the mean of each feature parameter; The covariance matrix represents the inconsistency of the battery pack, using the mean and standard deviation of each characteristic parameter. Specifically, the diagonal elements of the covariance matrix represent the variance of a single characteristic parameter, while the off-diagonal elements represent the covariance (i.e., correlation coefficient) between two characteristic parameters.

[0096] Understandably, the multivariate Gaussian distribution model is a generalization of the univariate normal distribution in a high-dimensional space, often used to describe the joint probability distribution of a set of interconnected continuous random variables. The multiple characteristic parameters of a single battery cell are not independent; for example, aging single batteries often exhibit characteristics such as "capacity decrease, internal resistance increase, and SOC lower." The inconsistency of battery packs is essentially due to the dispersion of multiple characteristic parameters and the abnormal correlations between them. The multivariate Gaussian distribution model, as a statistical model that can simultaneously describe the mean, dispersion, and correlation between these characteristic parameters, perfectly matches the "correlation" and "randomness" of these characteristic parameters.

[0097] In other implementations, the individual parameter distribution model may include a Gaussian mixture model, etc., and the embodiments of this application do not limit this.

[0098] In step S402 above, individual cell parameters are randomly sampled based on the individual cell parameter distribution model to obtain multiple sets of individual cell parameters that conform to the distribution law described by the individual cell parameter distribution model. Each set of individual cell parameters corresponds to a virtual battery pack (i.e., the third battery pack), which includes the individual cell parameters of each cell in the battery pack and can be used to simulate the inconsistencies between individual cells in the battery pack.

[0099] S404 determines the distribution of individual cell parameters for each third battery pack based on the individual cell parameters of each third battery pack.

[0100] For each third battery pack, statistical analysis of the individual cell parameters yields the distribution of those parameters. The statistical analysis can be performed using existing techniques and will not be elaborated further.

[0101] S406, based on the individual cell parameters of each third battery pack and the equivalent circuit model of each third battery pack, performs simulation to obtain the third operating data sequence of each third battery pack.

[0102] The equivalent circuit model of the third battery pack is similar to that of the second battery pack. It is also a circuit composed of basic circuit elements such as voltage sources, resistors, and capacitors, used to simulate the external electrical characteristics of the third battery pack.

[0103] The individual cell parameters of the third battery pack are assigned to the basic circuit elements in its equivalent circuit model. For example, based on the internal resistance and capacity of each individual cell in the third battery pack, the resistance and capacitance in the equivalent circuit model are assigned values ​​to form a virtual battery pack equivalent to the third battery pack. Then, a "charge and discharge current sequence" (such as constant current charging and dynamic operating condition discharging) is input into the assigned equivalent circuit model to calculate the individual cell voltage, total voltage, maximum individual cell voltage, minimum individual cell voltage, current, etc. at each moment, thus obtaining the third operating data sequence of the third battery pack.

[0104] The construction method of the equivalent circuit model of the third battery pack is similar to that of the equivalent circuit model of the second battery pack mentioned above, and will not be repeated here.

[0105] Understandably, in practical applications, it is difficult to obtain a large amount of real-world battery pack data with varying inconsistencies. By sampling using a single-cell parameter distribution model, the differences in single-cell parameters within a real battery pack can be simulated. Combined with simulation using an equivalent circuit model of the battery pack, a large amount of training data can be generated for the second model efficiently and with low overhead. This training data is diverse and comprehensive, helping the second model to better adapt to different levels of inconsistency and exhibiting stronger robustness.

[0106] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0107] Based on the same inventive concept, embodiments of this application also provide a battery pack state of energy estimation device. Please refer to... Figure 5 The image shows a schematic diagram of a battery pack state of energy estimation device 500 provided in one embodiment. The device 500 may include a first acquisition module 510 and an input module 520.

[0108] The first acquisition module 510 is used to acquire a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times.

[0109] The input module 520 is used to input the first running data sequence into the first model and output the first energy state of the first battery pack.

[0110] The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

[0111] In some embodiments, the second operational data sequence and the second energy state are obtained as follows: Based on the individual parameter distribution of the first battery pack, the state space corresponding to the second battery pack is determined, and the state space is used to represent the constraints on the individual parameters of the second battery pack. Sampling is performed within the state space to obtain the individual cell parameters of the second battery pack; Simulations were performed based on the individual cell parameters and the equivalent circuit model of the second battery pack to obtain the second operating data sequence and the second energy state of the second battery pack.

[0112] In some embodiments, the second battery pack includes a plurality of individual cells, and the equivalent circuit model of the second battery pack is obtained as follows: Based on the structure of each individual cell in the second battery pack, an equivalent circuit model of each individual cell is constructed. Based on the connection method of the multiple individual cells, the equivalent circuit models of each individual cell are connected to obtain the equivalent circuit model of the second battery pack.

[0113] In some embodiments, the second model is trained based on the third operating data sequence of multiple third battery packs and the individual parameter distributions; the individual parameter distributions of the multiple third battery packs are different.

[0114] In some embodiments, the third operating data sequence and individual cell parameter distribution of the plurality of third battery packs are obtained in the following manner: Based on a pre-defined single-cell parameter distribution model, the single-cell parameters of each third battery pack are obtained by sampling. Based on the individual cell parameters of each third battery pack, the distribution of individual cell parameters of each third battery pack is determined; Simulations were performed based on the individual cell parameters of each third battery pack and the equivalent circuit model of each third battery pack to obtain the third operating data sequence of each third battery pack.

[0115] In some embodiments, the individual cell parameters of each third battery pack include the values ​​of multiple characteristic parameters of the individual cells in each third battery pack; The single-parameter distribution model includes a multivariate Gaussian distribution model, in which the random variables include the multiple feature parameters, the mean vector in the multivariate Gaussian distribution model includes the mean of each feature parameter, and the covariance matrix in the multivariate Gaussian distribution model includes the correlation coefficients between the multiple feature parameters.

[0116] In some embodiments, the individual cell parameter distribution of the first battery pack is obtained in the following manner: Extract the spatial and temporal features of the first running data sequence; The spatial features and the temporal features are fused to obtain fused features; Based on the fusion features, the individual parameter distribution of the first battery pack is obtained through prediction.

[0117] Obviously, the battery pack state of energy estimation device provided in this application embodiment can be used as the above-mentioned... Figure 2 The illustrated battery pack state of energy estimation method is the main execution body, thus enabling the battery pack state of energy estimation device to perform... Figure 2 The functions implemented are the same, so they will not be described in detail here.

[0118] Figure 6 This is a schematic diagram of the structure of an electronic device provided in one embodiment. Please refer to it. Figure 6At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or non-volatile memory, such as at least one disk drive. Of course, the electronic device may also include other hardware required for other business operations.

[0119] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0120] Memory is used to store programs. Specifically, programs may include program code, which includes computer operation instructions. Memory may include main memory and non-volatile memory, and provides instructions and data to the processor.

[0121] The processor reads the corresponding computer program from non-volatile memory into main memory and then executes it, forming a battery pack state of energy estimation device at the logical level. The processor executes the program stored in memory and specifically performs the following operations: Obtain a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; Input the first running data sequence into the first model and output the first energy state of the first battery pack; The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

[0122] The above is as stated in this application. Figure 2The method executed by the battery pack state of energy estimation device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0123] The electronic device can also perform Figure 2 The method, and the implementation of the battery pack state of energy estimation device in Figure 2 The functions of the embodiments shown are not described in detail here.

[0124] Of course, in addition to software implementation, the electronic device of this application does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0125] This application also proposes a computer-readable storage medium that stores one or more programs, the programs including instructions that, when executed by an electronic device including multiple applications, enable the electronic device to perform... Figure 2 The method of the illustrated embodiment is specifically used to perform the following operations: Obtain a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; Input the first running data sequence into the first model and output the first energy state of the first battery pack; The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

[0126] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform the steps of the battery pack state of energy estimation method provided in this application.

[0127] In summary, the above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

[0128] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0129] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0130] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0131] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

Claims

1. A method for estimating the state of energy of a battery pack, characterized in that, include: Obtain a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; Input the first running data sequence into the first model and output the first energy state of the first battery pack; The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

2. The method according to claim 1, characterized in that, The second running data sequence and the second energy state are obtained in the following manner: Based on the individual parameter distribution of the first battery pack, the state space corresponding to the second battery pack is determined, and the state space is used to represent the constraints on the individual parameters of the second battery pack. Sampling is performed within the state space to obtain the individual cell parameters of the second battery pack; Simulations were performed based on the individual cell parameters and the equivalent circuit model of the second battery pack to obtain the second operating data sequence and the second energy state of the second battery pack.

3. The method according to claim 2, characterized in that, The second battery pack comprises multiple individual cells, and its equivalent circuit model is obtained as follows: Based on the structure of each individual cell in the second battery pack, an equivalent circuit model of each individual cell is constructed. Based on the connection method of the multiple individual cells, the equivalent circuit models of each individual cell are connected to obtain the equivalent circuit model of the second battery pack.

4. The method according to claim 1, characterized in that, The second model is trained based on the third operating data sequence and individual cell parameter distribution of multiple third battery packs; the individual cell parameter distributions of the multiple third battery packs are different.

5. The method according to claim 4, characterized in that, The third operating data sequence and individual cell parameter distribution of the multiple third battery packs are obtained in the following way: Based on a pre-defined single-cell parameter distribution model, the single-cell parameters of each third battery pack are obtained by sampling. Based on the individual cell parameters of each third battery pack, the distribution of individual cell parameters of each third battery pack is determined; Simulations were performed based on the individual cell parameters of each third battery pack and the equivalent circuit model of each third battery pack to obtain the third operating data sequence of each third battery pack.

6. The method according to claim 5, characterized in that, The individual cell parameters of each third battery pack include the values ​​of multiple characteristic parameters of the individual cells in each third battery pack; The single-parameter distribution model includes a multivariate Gaussian distribution model, in which the random variables include the multiple feature parameters, the mean vector in the multivariate Gaussian distribution model includes the mean of each feature parameter, and the covariance matrix in the multivariate Gaussian distribution model includes the correlation coefficients between the multiple feature parameters.

7. The method according to claim 1, characterized in that, The individual cell parameter distribution of the first battery pack is obtained in the following way: Extract the spatial and temporal features of the first running data sequence; The spatial features and the temporal features are fused to obtain fused features; Based on the fusion features, the individual parameter distribution of the first battery pack is obtained through prediction.

8. A battery pack state of energy estimation device, characterized in that, include: The first acquisition module is used to acquire a first operating data sequence of the first battery pack, the first operating data sequence including operating data of the first battery pack at multiple times; The input module is used to input the first running data sequence into the first model and output the first energy state of the first battery pack. The first model is trained based on the second operating data sequence and the second state of energy of the second battery pack. The individual parameter distribution of the second battery pack is similar to that of the first battery pack. The individual parameter distribution of the battery pack is predicted by the second model based on the operating data sequence of the battery pack.

9. An electronic device, characterized in that, A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the battery pack state of energy estimation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is able to perform the steps of the battery pack state of energy estimation method as described in any one of claims 1 to 7.

11. A computer program product, characterized in that, The computer program product includes a non-transitory computer-readable storage medium storing a computer program operable to cause a computer to perform the steps of the battery pack state of energy estimation method as described in any one of claims 1 to 7.

Images