Battery state determination method, apparatus, non-transitory storage medium, and electronic device

Abstract

The application discloses a battery state determination method and device, a nonvolatile storage medium and an electronic device. The method comprises the following steps: determining working condition information corresponding to a target load powered by a target battery pack; monitoring a plurality of battery monomers included in the target battery pack to obtain running data corresponding to the plurality of battery monomers respectively; determining a battery pack space feature based on the working condition information and the running data corresponding to the plurality of battery monomers respectively, wherein the battery pack space feature is used for representing positions of the plurality of battery monomers arranged in the target battery pack and influences on electrical performance of the target battery pack; and determining a battery health state of the target battery pack based on the battery pack space feature. The application solves the technical problem that the battery health state determined in the related art is inaccurate.

Description

Battery state determination methods, apparatus, non-volatile storage media, and electronic devices Technical Field

[0001] This invention relates to the field of battery monitoring, and more specifically, to a method, apparatus, non-volatile storage medium, and electronic device for determining battery state. Background Technology

[0002] In the field of energy storage batteries, determining the health status of batteries is of great significance. The health status of energy storage batteries directly affects the safety, reliability, and performance of energy storage systems. Therefore, accurately assessing the health status of energy storage batteries is crucial for extending battery life, improving system efficiency, and reducing operating costs. However, traditional energy storage battery health status estimation techniques have some limitations and shortcomings, restricting their effectiveness and reliability in practical applications. Related technologies generally rely on real-time monitoring of battery operation to determine operational data for assessing battery health status. However, the data dimensions used for analysis and judgment are relatively limited, leading to inaccurate determinations of battery health status.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This invention provides a method, apparatus, non-volatile storage medium, and electronic device for determining battery status, in order to at least solve the technical problem of inaccurate determination of battery health status in related technologies.

[0005] According to one aspect of the present invention, a battery state determination method is provided, comprising: determining operating condition information corresponding to a target load powered by a target battery pack; monitoring a plurality of battery cells included in the target battery pack to obtain operating data corresponding to each of the plurality of battery cells; determining battery pack spatial characteristics based on the operating condition information and the operating data corresponding to the plurality of battery cells, wherein the battery pack spatial characteristics are used to characterize the position of the plurality of battery cells arranged in the target battery pack and their impact on the electrical performance of the target battery pack; and determining the battery health state of the target battery pack based on the battery pack spatial characteristics.

[0006] Optionally, determining the operating condition information corresponding to the target load powered by the target battery pack includes: determining the load circuit model of the target load; performing equivalence based on the load circuit model to determine the equivalent circuit parameters of the target load; and determining the operating condition information based on the equivalent circuit parameters.

[0007] Optionally, the equivalent circuit parameters include equivalent voltage, equivalent current, and equivalent resistance. Determining the operating condition information based on the equivalent circuit parameters includes: determining the current power supply requirement of the target load based on the equivalent voltage and the equivalent current; determining the rotational speed information of the target load based on the equivalent resistance; and determining the operating condition information based on the current power supply requirement and the rotational speed information.

[0008] Optionally, the equivalent circuit parameters include equivalent capacitance and equivalent inductance, and determining the operating condition information based on the equivalent circuit parameters includes: determining the target load as a load type based on the equivalent capacitance and the equivalent inductance; and determining the operating condition information based on the load type.

[0009] Optionally, determining the spatial characteristics of the battery pack based on the operating condition information and the operating data corresponding to the plurality of battery cells includes: processing the operating condition information and the operating data corresponding to the plurality of battery cells using a convolutional neural network to generate a multidimensional feature map of the target battery pack, wherein the feature dimension of the multidimensional feature map is determined based on the data dimension of the operating data; and performing dimensionality reduction processing based on the multidimensional feature map to determine the spatial characteristics of the battery pack.

[0010] Optionally, the step of processing the operating condition information and the operating data corresponding to the plurality of battery cells using a convolutional neural network to generate a multidimensional feature map of the target battery pack includes: processing the operating condition information and the operating data corresponding to the plurality of battery cells using the convolutional neural network to determine the individual spatial features corresponding to the plurality of battery cells, wherein the individual spatial features are used to characterize the arrangement position of the plurality of battery cells in the target battery pack, and the impact on the electrical performance of the corresponding individual battery cells is generated based on the individual spatial features corresponding to the plurality of battery cells to generate the multidimensional feature map.

[0011] Optionally, determining the battery health status of the target battery pack based on the spatial characteristics of the battery pack includes: processing the input into a gated loop unit based on the spatial characteristics of the battery pack to determine the temporal characteristics of the battery pack, wherein the temporal characteristics of the battery pack are used to represent the dependency relationship between the electrical performance of the multiple battery cells and the monitoring time within a predetermined time period; and determining the battery health status based on the temporal characteristics of the battery pack.

[0012] According to another aspect of the present invention, a battery state determination device is provided, comprising: an operating condition monitoring module, configured to determine operating condition information corresponding to a target load powered by a target battery pack; a single cell monitoring module, configured to monitor a plurality of battery cells included in the target battery pack and obtain operating data corresponding to each of the plurality of battery cells; a spatial feature acquisition module, configured to determine battery pack spatial features based on the operating condition information and the operating data corresponding to the plurality of battery cells, wherein the battery pack spatial features characterize the position of the plurality of battery cells in the target battery pack and its impact on the electrical performance of the target battery pack; and a battery health status monitoring module, configured to determine the battery health status of the target battery pack based on the battery pack spatial features.

[0013] According to another aspect of the present invention, a non-volatile storage medium is provided, the non-volatile storage medium storing a plurality of instructions adapted for loading by a processor and executing any one of the battery state determination methods described herein.

[0014] According to another aspect of the present invention, an electronic device is provided, comprising: one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement any of the battery state determination methods described above.

[0015] In this embodiment of the invention, the operating condition information corresponding to the target load powered by the target battery pack is determined; multiple battery cells included in the target battery pack are monitored to obtain operating data corresponding to each of the multiple battery cells; based on the operating condition information and the operating data corresponding to each of the multiple battery cells, the spatial characteristics of the battery pack are determined, wherein the spatial characteristics of the battery pack are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack; based on the spatial characteristics of the battery pack, the battery health status of the target battery pack is determined. This achieves the goal of determining the battery health status by combining operating condition information and battery pack spatial characteristics, realizing the technical effect of comprehensively considering the impact of external operating condition changes and battery spatial arrangement characteristics on the battery health status, improving the accuracy of battery health status determination, and thus solving the technical problem of inaccurate battery health status determination in related technologies. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0017] Figure 1 is a flowchart of an optional battery state determination method provided according to an embodiment of the present invention;

[0018] Figure 2 is a block diagram of an optional battery state determination method provided according to an embodiment of the present invention;

[0019] Figure 3 is a schematic diagram of an optional battery state determination device provided according to an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0022] In the field of energy storage batteries, accurately determining the health status of batteries is a crucial step in battery monitoring. The health status of energy storage batteries directly affects the safety, reliability, and performance of the energy storage system. Therefore, accurately assessing the health status of energy storage batteries is of great significance for extending battery life, improving system efficiency, and reducing operating costs. However, traditional energy storage battery health status estimation techniques have some limitations and shortcomings. Among related technologies, methods for determining the health status of energy storage batteries based on mathematical models rely on specific electrochemical models. A single mathematical model often cannot comprehensively and accurately describe the dynamic characteristics of the battery, leading to deviations between the estimated results and actual conditions. Especially under conditions such as battery aging, increased cycle count, and changes in the external environment, mathematical models often need continuous correction and adjustment, limiting their reliability and stability in practical applications. Data-driven methods for determining the health status of energy storage batteries estimate the battery's state using a large amount of historical data. However, due to the numerous uncertainties and nonlinear factors that exist during battery operation, traditional data-driven methods struggle to accurately predict the battery's long-term health status. Furthermore, the aforementioned technologies only monitor the battery's voltage, current, and temperature data to determine its health, without considering the impact of changes in the external operating conditions on the battery's health, or the spatial characteristics of the battery cell arrangement, leading to inaccurate monitoring of the battery's health status.

[0023] To address the aforementioned problems, this invention provides a method embodiment for determining battery state. It should be noted that the steps shown in the flowcharts can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be executed in a different order than that presented here.

[0024] Figure 1 is a flowchart of an optional battery state determination method provided according to an embodiment of the present invention. As shown in Figure 1, the method includes the following steps:

[0025] Step S102: Determine the operating condition information corresponding to the target load powered by the target battery pack;

[0026] It is understandable that the output of the target battery pack is directly related to the operating conditions of the target load it powers. The battery pack's output affects the battery's health status. Therefore, to accurately determine the battery's health status, it is essential to first clarify the operating conditions of the target load powered by the target battery pack. The above processing method allows us to obtain the operating conditions of the target load powered by the target battery pack, thus enabling a more accurate determination of the battery pack's health status.

[0027] In one optional embodiment, determining the operating condition information corresponding to the target load powered by the target battery pack includes: determining the load circuit model of the target load; performing equivalence based on the load circuit model to determine the equivalent circuit parameters of the target load; and determining the operating condition information based on the equivalent circuit parameters.

[0028] It is understandable that, in order to accurately obtain the operating condition information corresponding to the target load powered by the target battery pack, the target circuit of the target load needs to be modeled. To facilitate operating condition analysis of the load circuit model of the target load, the load circuit model can be equivalent, for example, using the Thevenin Equivalent Circuit Model. By analyzing the equivalent circuit model, the equivalent circuit parameters of the target load are determined. Analyzing the obtained equivalent circuit parameters allows us to determine the operating condition information corresponding to the target load. Through the above processing, the process of obtaining the operating condition information corresponding to the target load powered by the target battery pack is simplified, improving the efficiency and accuracy of obtaining this information. This is beneficial for obtaining real-time operating condition information and timely considering the impact of changes in operating condition information on battery health.

[0029] Optionally, the Thevenin equivalent circuit model refers to a model that simplifies complex circuits using Thevenin's theorem and the superposition theorem. By applying the Thevenin equivalent circuit model, the load circuit model can be simplified and equivalently processed, which is beneficial to the understanding and analysis of complex circuits.

[0030] In one optional embodiment, the equivalent circuit parameters include equivalent voltage, equivalent current, and equivalent resistance. Based on the equivalent circuit parameters, the operating condition information is determined, including: determining the current power supply requirement of the target load based on the equivalent voltage and equivalent current; determining the speed information of the target load based on the equivalent resistance; and determining the operating condition information based on the current power supply requirement and speed information.

[0031] It is understandable that by employing an equivalent circuit model, such as the Thevenin equivalent circuit model, and performing equivalent analysis, the equivalent circuit parameters of the target load can be obtained, including equivalent voltage, equivalent current, and equivalent resistance. The equivalent voltage and equivalent current reflect the current power supply requirements of the target load. The equivalent resistance reflects the load's rotational speed. When the load (e.g., a motor) is static, i.e., not energized or not rotating, its resistance is mainly determined by factors such as the material, length, cross-sectional area, and connection method of the motor windings. The resistance value measured at this time can be considered as the motor's basic resistance or static resistance. However, when the motor rotates, the electromagnetic field distribution and current distribution inside the motor change, leading to a change in the resistance value. Furthermore, the heat generated during motor operation also affects the winding temperature, thus altering the resistance value. Therefore, the resistance value measured when the motor is rotating is often different from the resistance value measured when it is static. Thus, analyzing the equivalent resistance can determine the rotational speed information of the target load. The power supply requirements and rotational speed of the target load reflect its operating condition. By determining the power supply requirements and rotational speed of the target load, its operating condition information can be determined. Through the above processing, the operating condition information corresponding to the target load powered by the target battery pack is clarified. Accurate analysis of the operating condition information can improve the accuracy of determining the battery health status of the target battery pack.

[0032] In one optional embodiment, the equivalent circuit parameters include equivalent capacitance and equivalent inductance. Based on the equivalent circuit parameters, determining the operating condition information includes: determining the target load as a load type based on the equivalent capacitance and equivalent inductance; and determining the operating condition information based on the load type.

[0033] It is understandable that by employing an equivalent circuit model and performing equivalent analysis, the equivalent circuit parameters of the target load can be obtained, including, for example, equivalent capacitance and equivalent inductance. Equivalent capacitance and equivalent inductance reflect the load type of the target load, such as capacitive or inductive loads. Due to its characteristics, an inductive load generates a back electromotive force when there are sudden changes in voltage or current, leading to drastic changes in voltage and current. This change may cause instability in the output voltage of the battery pack, affecting its power supply performance. Inductive loads also affect the charging and discharging process of the battery pack, thus shortening its lifespan. Simultaneously, frequent changes in the load type of the target load may damage the internal structure and materials of the battery pack, further shortening its lifespan. Therefore, capacitive and inductive loads have different effects on the battery health status of the target battery pack. Through the above processing, the load type of the target load is determined, considering the different characteristics of the output voltage and current of different load types on the target battery pack, as well as the impact of frequent changes in load type on the battery health status of the target battery pack, thereby improving the accuracy of determining the battery health status of the target battery pack.

[0034] Optionally, a rectifier and an inverter are provided between the target load and the target battery pack. The rectifier is used to convert AC power to DC power. When the load recovers energy to generate AC power, the rectifier can convert the AC power back to DC power to replenish the target battery pack. The inverter is used to convert DC power to AC power. It can convert the DC power output from the target battery pack back to AC power to power loads that require AC power.

[0035] Alternatively, a capacitive load refers to a load with a capacitance parameter, in which the current phase leads the voltage phase.

[0036] Optionally, an inductive load refers to a load with inductive parameters, whose current phase is followed by its voltage phase, such as a transformer or a motor.

[0037] Optionally, current phase and voltage phase are physical quantities that reflect the state of alternating current at any given moment. Current phase refers to the time offset of the current waveform relative to a reference signal, and can be expressed in phase angles. Voltage phase refers to the time offset of the voltage waveform relative to a reference signal; voltage phase can also be expressed in phase angles.

[0038] Step S104: Monitor the multiple battery cells included in the target battery pack to obtain the operating data corresponding to each of the multiple battery cells.

[0039] It is understandable that the target battery pack consists of multiple battery cells. Because the quality of these cells—that is, their uniformity and spatial distribution—is not entirely the same, the operating data corresponding to each cell will differ. By monitoring multiple battery cells through the above process, accurately obtaining their corresponding operating data can provide a basis for subsequent analysis.

[0040] Optionally, the operating data of a single battery cell includes, for example, output voltage, charging voltage, output current, charging current, and battery temperature. Output voltage, charging voltage, output current, charging current, and battery temperature are crucial operating data for a single battery cell. The magnitude of the output voltage determines the amount of electrical energy the battery provides to the external load, the magnitude of the output current determines the amount of power the battery provides within a certain time, and battery temperature has a significant impact on battery performance and lifespan. The charging voltage affects the rate of chemical reactions within the battery cell, thus affecting the cell's temperature and performance. Excessive charging current leads to overheating of the battery cell, while insufficient charging current leads to metal deposition within the cell, affecting its performance and overall health. Monitoring output voltage, charging voltage, output current, charging current, and battery temperature is essential for accurately assessing battery performance and health.

[0041] Step S106: Based on the operating condition information and the operating data corresponding to each of the multiple battery cells, determine the spatial characteristics of the battery pack. The spatial characteristics of the battery pack are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack.

[0042] It is understandable that the different arrangements of individual battery cells within a battery pack will have varying impacts on their electrical performance. For example, a battery cell located in the center of the pack will have its heat dissipation capacity affected, resulting in a higher operating temperature compared to cells located at the edges, thus affecting its battery health. The wiring at the edges of the pack may experience uneven voltage distribution, further impacting the battery health of cells in different positions. The spatial characteristics of a battery pack refer to the impact of the different arrangements of multiple battery cells on its electrical performance. These spatial characteristics can be determined by analyzing the operating conditions of the target load powered by the battery pack and the operating data of each individual battery cell. By combining the above processing with the operating condition information and the operating data of each individual battery cell to determine the battery health status, a more accurate assessment of battery health can be obtained.

[0043] In one optional embodiment, the spatial characteristics of the battery pack are determined based on operating condition information and the operating data corresponding to each of the multiple battery cells, including: processing the operating condition information and the operating data corresponding to each of the multiple battery cells using a convolutional neural network to generate a multidimensional feature map of the target battery pack, wherein the feature dimension of the multidimensional feature map is determined based on the data dimension of the operating data; and performing dimensionality reduction processing based on the multidimensional feature map to determine the spatial characteristics of the battery pack.

[0044] It is understandable that the spatial characteristics of the battery pack can be determined by analyzing the operating conditions of the target load powered by the target battery pack and the operational data of each individual battery cell obtained from monitoring the individual cells. A Convolutional Neural Network (CNN) is used to process the operating conditions and operational data of each battery cell to generate a multidimensional feature map of the target battery pack. The feature dimension of the multidimensional feature map is determined by the dimension of the operational data; for example, when the operational data includes battery temperature, output current, and output voltage, the feature dimension of the multidimensional feature map is three. This multidimensional feature map is then mapped to a lower-dimensional space for dimensionality reduction, extracting key feature information to obtain the spatial characteristics of the battery pack. Through this processing, the spatial characteristics of the battery pack are clearly defined based on the operating conditions and operational data of each individual battery cell. The impact of the operating conditions and the arrangement of the battery cells on the battery health status is fully considered, improving the accuracy of determining the battery health status.

[0045] Optionally, a convolutional neural network (CNN) is a type of feedforward neural network that includes convolutional computation and has a deep structure. CNNs can perform translation-invariant classification and are commonly used to analyze visual images. In the embodiments of this invention, the CNN can effectively capture static feature information in the operating data of individual battery cells, improving the expressive power of static feature information.

[0046] In one optional embodiment, based on operating condition information and the operating data corresponding to multiple battery cells, a convolutional neural network is used for processing to generate a multidimensional feature map of the target battery pack. This includes: based on operating condition information and the operating data corresponding to multiple battery cells, a convolutional neural network is used for processing to determine the individual spatial features corresponding to each of the multiple battery cells, wherein the individual spatial features are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the corresponding individual battery cells; and based on the individual spatial features corresponding to each of the multiple battery cells, a multidimensional feature map is generated.

[0047] It is understandable that by using convolutional neural networks to process operating condition information and the operational data corresponding to multiple individual battery cells, multi-dimensional feature maps of individual battery cells can be obtained. These multi-dimensional feature maps reflect the impact of the arrangement of multiple battery cells in the target battery pack on the battery performance of that individual cell. For example, when a battery cell is arranged in the center of the battery pack, its heat dissipation capacity will be affected, and the operating temperature of that battery cell will be higher than that of battery cells arranged at the edge of the battery pack, thus affecting its battery health. The circuitry at the edge of the battery pack may exhibit uneven voltage distribution, which will also have different impacts on the battery health of battery cells arranged in different positions. Summarizing and combining the multi-dimensional feature maps of multiple individual battery cells yields the multi-dimensional feature map of the battery pack. Through the above processing, the multi-dimensional feature map of the battery pack is obtained. Mapping this multi-dimensional feature map to a low-dimensional space for dimensionality reduction yields the spatial features of the battery pack. Based on the analysis of operating condition information and the spatial features of the battery pack, the accuracy of determining the battery health status is improved.

[0048] Step S108: Determine the battery health status of the target battery pack based on the spatial characteristics of the battery pack.

[0049] It is understandable that the spatial characteristics of the battery pack—that is, the arrangement of multiple battery cells within the target battery pack—affect the electrical performance of the target battery pack, thereby determining the battery health status. The above processing fully considers the impact of the battery pack's spatial characteristics on battery health status, improving the accuracy of determining battery health status.

[0050] In one optional embodiment, determining the battery health status of a target battery pack based on its spatial characteristics includes: processing the input into a gated loop unit based on the spatial characteristics of the battery pack to determine the battery pack temporal characteristics, wherein the battery pack temporal characteristics are used to represent the dependency relationship between the electrical performance of multiple individual battery cells and the monitoring time within a predetermined time period; and determining the battery health status based on the battery pack temporal characteristics.

[0051] It can be understood that the spatial characteristics of the battery pack—that is, the arrangement of multiple battery cells within the target battery pack—and their impact on the electrical performance of the target battery pack are input into a gated loop unit for processing. This process determines the battery pack's temporal characteristics, i.e., the dependencies between the electrical performance of each battery cell and the monitoring time within a predetermined time period. The gated loop unit can process the sequence data, capture the dependencies between the electrical performance of each battery cell and the monitoring time, and determine the changes in the battery health status of the battery pack at different time points, thereby determining the battery pack's temporal characteristics. Based on these temporal characteristics, the battery health status of the battery pack can be determined. Through the above processing, utilizing the gated loop unit's ability to process dynamic changes in sequence data, understanding the data's evolution patterns, and obtaining the battery pack's temporal characteristics, the battery health status of the battery pack can be accurately determined.

[0052] Optionally, a Gated Recurrent Unit (GRU) is a type of recurrent neural network that uses a gating mechanism to control the flow of information. It contains a reset gate and an update gate, which control the amount of information passed to the next time step. Gated recurrent units help solve the vanishing gradient problem in traditional recurrent neural networks (RNNs) by allowing the model to selectively retain or forget information from previous time steps.

[0053] Optionally, the convolutional neural network and gated recurrent units need to be pre-trained. First, the training dataset needs to be constructed. This involves cleaning, aligning, and normalizing the collected battery pack operating data. Data cleaning removes outliers, noise, and interference from the raw data, including output voltage, output current, battery temperature, sampling time, and operating conditions. Kalman filtering is used to smooth the data and remove noise. Data alignment aligns the output voltage, charging voltage, output current, charging current, battery temperature, and operating conditions that have time differences, ensuring temporal consistency. Interpolation or resampling methods are used for data alignment. Data normalization standardizes the extracted features to eliminate the influence of different data magnitudes and units, using min-max scaling. The resulting dataset is then fed into the convolutional neural network for training. Feature extraction begins by using output voltage, output current, battery temperature, and load condition data as input to the convolutional neural network. Convolutional and pooling layers capture features at different spatial locations, generating corresponding multidimensional feature maps. The multidimensional feature map output by the convolutional neural network (CNN) is flattened and converted into a one-dimensional vector. This one-dimensional feature vector contains static feature information from the input data. This one-dimensional feature vector is then input into a gated recurrent unit (GRU) for training, learning the dynamic changes, temporal relationships, and evolutionary patterns of the sequence data. The GRU, through its adaptive gating mechanism, effectively captures long-term dependencies in the sequence. Iterative training is performed on the training set at time steps t, using RMSE (Root Mean Square Error) as the loss function for gradient inversion to adjust and obtain optimal weight values, biases, and learning rates. This completes the pre-training process of the CNN and the GRU, resulting in a trained fusion model based on the CNN and the GRU.

[0054] Optionally, time step t represents a data observation at a point in time. Iterative training at time step t will treat the data at each time step as a training sample and use it to update the model parameters in turn, which helps the model learn the time dependence of the sequence data.

[0055] Alternatively, Kalman filtering is a highly efficient recursive filter that takes into account the joint distribution of each measurement at different times and then generates an estimate of the position variable.

[0056] Through step S102, the operating condition information corresponding to the target load powered by the target battery pack is determined; in step S104, multiple battery cells included in the target battery pack are monitored to obtain the operating data corresponding to each battery cell; in step S106, based on the operating condition information and the operating data corresponding to each battery cell, the spatial characteristics of the battery pack are determined, wherein the spatial characteristics of the battery pack are used to characterize the position of multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack; in step S108, based on the spatial characteristics of the battery pack, the battery health status of the target battery pack is determined. This achieves the goal of determining the battery health status by combining the operating condition information and the spatial characteristics of the battery pack, and realizes the technical effect of improving the accuracy of determining the battery health status by comprehensively considering the impact of external operating condition changes and battery spatial arrangement characteristics on the battery health status, thereby solving the technical problem of inaccurate determination of battery health status in related technologies.

[0057] Based on the above embodiments and optional embodiments, this invention proposes an optional implementation method. Figure 2 is a block diagram of an optional battery state determination method provided by an embodiment of this invention. As shown in Figure 2, offline training is first performed, requiring the construction of a training dataset. The collected battery pack operating data is cleaned, aligned, and normalized. The dataset obtained from the above processing, containing battery pack operating data such as voltage, current, battery temperature, and operating condition information, is fed into a convolutional neural network (CNN) for training. Feature extraction is first performed, capturing features at different spatial locations of the data through convolutional layers, and retaining the main features while reducing parameters and computational load through pooling layers. There are multiple sets of convolutional and pooling layers. The multidimensional feature map obtained from the above processing is flattened and converted into a one-dimensional vector. The one-dimensional feature vector contains static feature information from the input data. The one-dimensional feature vector is input into a gated recurrent unit for training to learn the dynamic changes, temporal relationships, and evolutionary patterns of the sequence data. The training is iteratively trained on the training set at time step t. RMSE (Root Mean Square Error) is used as the loss function for gradient inversion. The optimal weight values, bias, and learning rate are adjusted to obtain the optimal parameters, thereby completing the pre-training process of the convolutional neural network and the gated recurrent unit, and obtaining the trained fusion model based on the convolutional neural network and the gated recurrent unit.

[0058] After offline training, the battery health status determination process begins. First, circuit modeling is performed on the target load to determine its load circuit model. An equivalent circuit model, such as the Thevenin equivalent circuit model, is used to equivalence the load circuit model, determining the equivalent circuit parameters of the target load. These parameters include, for example, equivalent voltage, equivalent current, and equivalent resistance. Based on the equivalent voltage and equivalent current, the current power supply requirements of the target load are determined. Based on the equivalent resistance, the rotational speed information of the target load is determined. The equivalent resistance reflects the load's rotational speed. When the load (e.g., a motor) is static (i.e., not energized or not rotating), its resistance is mainly determined by factors such as the material, length, cross-sectional area, and connection method of the motor windings. The resistance value measured at this time can be considered the motor's basic resistance or static resistance. However, when the motor rotates, the electromagnetic field and current distribution inside the motor change, leading to a change in the resistance value. Furthermore, the heat generated during motor operation also affects the winding temperature, thus altering the resistance value. Therefore, the resistance value measured when the motor is rotating is often different from the resistance value measured when it is static. Therefore, the rotational speed of the target load can be determined by analyzing the equivalent resistance. Based on the current power supply requirements and rotational speed information, the operating condition information is determined.

[0059] The process involves monitoring multiple individual battery cells within the target battery pack to obtain their respective operational data, including output voltage, charging voltage, output current, charging current, and battery temperature. Based on this operational data and the operating conditions of each individual cell, a convolutional neural network is used to determine their individual spatial features. These features characterize the arrangement of the individual cells within the target battery pack and their impact on the electrical performance of each cell. A multidimensional feature map is then generated based on these individual spatial features. The feature dimension of the multidimensional feature map is determined by the data dimension of the operational data. Dimensionality reduction is then performed on the multidimensional feature map to determine the spatial characteristics of the battery pack. These spatial characteristics characterize the arrangement of the individual battery cells within the target battery pack and their impact on the electrical performance of the target battery pack. Based on the spatial characteristics of the battery pack, the input is processed by a gated loop unit to determine the temporal characteristics of the battery pack. The temporal characteristics of the battery pack are used to represent the dependence between the electrical performance of multiple battery cells and the monitoring time within a predetermined time period. Based on the temporal characteristics of the battery pack, the extracted temporal characteristics are integrated in a fully connected layer to determine the battery health status.

[0060] The above-mentioned optional implementation methods achieve at least the following effects: The equivalent circuit model in this invention provides operating condition information corresponding to the target load powered by the target battery pack, i.e., the energy storage battery. Considering that the arrangement of individual battery cells in the battery pack has an inner and outer distinction, the heat dissipation capacity of individual battery cells is uneven, or the voltage distribution at the corners of the circuit is uneven, the spatial characteristics of the battery pack are obtained by combining the fusion model based on convolutional neural networks and gated cyclic units. Combining the operating condition information and the spatial characteristics of the battery pack, the battery health status of the battery pack can be accurately determined. At the same time, the process of determining the battery health status is performed in real time. By using the equivalent circuit model and the fusion model based on convolutional neural networks and gated cyclic units for online estimation, the model can be updated using real-time battery operating data and the operating condition information corresponding to the load, which is conducive to more accurate determination of the battery health status.

[0061] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0062] This embodiment also provides a battery state determination device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the terms "module" and "device" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0063] According to an embodiment of the present invention, an apparatus embodiment for implementing the battery state determination method is also provided. FIG3 is a schematic diagram of an optional battery state determination apparatus provided according to an embodiment of the present invention. As shown in FIG3, the above-mentioned battery state determination apparatus includes an operating condition monitoring module 302, a single cell monitoring module 304, a spatial feature acquisition module 306, and a battery health status monitoring module 308. The apparatus will be described below.

[0064] Operating condition monitoring module 302 is used to determine the operating condition information corresponding to the target load powered by the target battery pack;

[0065] The single cell monitoring module 304 is connected to the operating condition monitoring module 302 and is used to monitor multiple single cells included in the target battery pack to obtain the operating data corresponding to each single cell.

[0066] The spatial feature acquisition module 306 is connected to the single cell monitoring module 304 and is used to determine the spatial features of the battery pack based on the operating condition information and the operating data corresponding to multiple battery cells. The spatial features of the battery pack are used to characterize the position of multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack.

[0067] The battery health status monitoring module 308 is connected to the spatial feature acquisition module 306 and is used to determine the battery health status of the target battery pack based on the spatial features of the battery pack.

[0068] In a battery state determination device provided by this invention, an operating condition monitoring module 302 is set up to determine the operating condition information corresponding to the target load powered by the target battery pack; a single cell monitoring module 304, connected to the operating condition monitoring module 302, is used to monitor multiple battery cells included in the target battery pack and obtain the operating data corresponding to each of the multiple battery cells; a spatial feature acquisition module 306, connected to the single cell monitoring module 304, is used to determine the spatial features of the battery pack based on the operating condition information and the operating data corresponding to the multiple battery cells, wherein the spatial features of the battery pack are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack; a battery health status monitoring module 308, connected to the spatial feature acquisition module 306, is used to determine the battery health status of the target battery pack based on the battery health status. This achieves the goal of determining the battery health status by combining operating condition information and battery pack spatial features, realizing the technical effect of improving the accuracy of battery health status determination by comprehensively considering the impact of external operating condition changes and battery spatial arrangement characteristics on battery health status, thereby solving the technical problem of inaccurate battery health status determination in related technologies.

[0069] It should be noted that the above modules can be implemented by software or hardware. For example, for the latter, it can be implemented in the following ways: the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0070] It should be noted that the aforementioned operating condition monitoring module 302, single-cell monitoring module 304, spatial feature acquisition module 306, and battery health status monitoring module 308 correspond to steps S102 to S108 in the embodiments. The instances and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run on a computer terminal.

[0071] It should be noted that the optional or preferred implementation methods of this embodiment can be found in the relevant descriptions in the embodiments, and will not be repeated here.

[0072] The aforementioned battery status determination device may also include a processor and a memory. The operating condition monitoring module 302, the single cell monitoring module 304, the spatial feature acquisition module 306, and the battery health status monitoring module 308 are all stored as program units in the memory, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0073] The processor contains a core that retrieves the corresponding program unit from memory. One or more cores may be configured. Memory may include non-persistent memory in computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory includes at least one memory chip.

[0074] This invention provides a non-volatile storage medium storing a program that, when executed by a processor, implements a battery state determination method.

[0075] This invention provides an electronic device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs the following steps: determining the operating condition information corresponding to a target load powered by a target battery pack; monitoring multiple individual battery cells included in the target battery pack to obtain operating data corresponding to each individual battery cell; determining the spatial characteristics of the battery pack based on the operating condition information and the operating data corresponding to each individual battery cell, wherein the spatial characteristics of the battery pack characterize the position of the multiple individual battery cells in the target battery pack and their impact on the electrical performance of the target battery pack; and determining the battery health status of the target battery pack based on the spatial characteristics of the battery pack. The device described herein may be a server, PC, etc.

[0076] This invention also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program with the following method steps: determining the operating condition information corresponding to the target load powered by the target battery pack; monitoring multiple battery cells included in the target battery pack to obtain operating data corresponding to each of the multiple battery cells; determining the spatial characteristics of the battery pack based on the operating condition information and the operating data corresponding to each of the multiple battery cells, wherein the spatial characteristics of the battery pack are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the target battery pack; and determining the battery health status of the target battery pack based on the spatial characteristics of the battery pack.

[0077] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0078] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0079] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0080] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0081] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0082] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0083] 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.

[0084] 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 process, method, article, or apparatus. Unless otherwise specified, 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 that element.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The above are merely embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for determining battery state, characterized in that, include: Determine the operating condition information corresponding to the target load powered by the target battery pack; The process involves monitoring multiple battery cells within the target battery pack to obtain operational data for each individual cell; determining the battery pack spatial characteristics based on the operating condition information and the operational data of each individual cell, whereby the battery pack spatial characteristics characterize the position of the multiple battery cells within the target battery pack and its impact on the electrical performance of the target battery pack; and determining the battery health status of the target battery pack based on the battery pack spatial characteristics. The determination of the battery pack spatial characteristics based on the operating condition information and the operational data of each individual cell includes: processing the operating condition information and the operational data of each individual cell using a convolutional neural network to generate a multidimensional feature map of the target battery pack. The feature dimension of the multidimensional feature map is determined based on the data dimension of the operating data; dimensionality reduction processing is performed on the multidimensional feature map to determine the spatial features of the battery pack; wherein, the step of generating the multidimensional feature map of the target battery pack by processing the operating condition information and the operating data corresponding to the multiple battery cells using a convolutional neural network includes: processing the operating condition information and the operating data corresponding to the multiple battery cells using the convolutional neural network to determine the individual spatial features corresponding to the multiple battery cells, wherein the individual spatial features are used to characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the corresponding individual battery cells; generating the multidimensional feature map based on the individual spatial features corresponding to the multiple battery cells.

2. The method according to claim 1, characterized in that, The step of determining the operating condition information corresponding to the target load powered by the target battery pack includes: determining the load circuit model of the target load; performing equivalence based on the load circuit model to determine the equivalent circuit parameters of the target load; and determining the operating condition information based on the equivalent circuit parameters.

3. The method according to claim 2, characterized in that, The equivalent circuit parameters include equivalent voltage, equivalent current, and equivalent resistance. Determining the operating condition information based on the equivalent circuit parameters includes: determining the current power supply requirement of the target load based on the equivalent voltage and the equivalent current; determining the rotational speed information of the target load based on the equivalent resistance; and determining the operating condition information based on the current power supply requirement and the rotational speed information.

4. The method according to claim 2, characterized in that, The equivalent circuit parameters include equivalent capacitance and equivalent inductance. Determining the operating condition information based on the equivalent circuit parameters includes: determining the target load as a load type based on the equivalent capacitance and equivalent inductance; and determining the operating condition information based on the load type.

5. The method according to claim 1, characterized in that, The step of determining the battery health status of the target battery pack based on the spatial characteristics of the battery pack includes: processing the input into a gated loop unit based on the spatial characteristics of the battery pack to determine the temporal characteristics of the battery pack, wherein the temporal characteristics of the battery pack are used to represent the dependency relationship between the electrical performance of the multiple battery cells and the monitoring time within a predetermined time period; and determining the battery health status based on the temporal characteristics of the battery pack.

6. A battery state determination device, characterized in that, include: The operating condition monitoring module is used to determine the operating condition information corresponding to the target load powered by the target battery pack; The single-cell monitoring module is used to monitor multiple single cells included in the target battery pack and obtain the operating data corresponding to each of the multiple single cells; A spatial feature acquisition module is used to determine the spatial features of the battery pack based on the operating condition information and the operating data corresponding to each of the multiple battery cells. The spatial features characterize the position of the multiple battery cells within the target battery pack and their impact on the electrical performance of the target battery pack. A battery health status monitoring module is used to determine the battery health status of the target battery pack based on the spatial features. The spatial feature acquisition module is further used to process the operating condition information and the operating data corresponding to each of the multiple battery cells using a convolutional neural network to generate a multidimensional feature map of the target battery pack. The feature dimension of the multidimensional feature map is determined based on the data dimension of the operating data. The multidimensional feature map is subjected to dimensionality reduction processing to determine the spatial features of the battery pack. The spatial feature acquisition module is further configured to generate a multidimensional feature map of the target battery pack by processing the operating condition information and the operating data corresponding to each of the multiple battery cells using a convolutional neural network. This includes: processing the operating condition information and the operating data corresponding to each of the multiple battery cells using the convolutional neural network to determine the individual spatial features corresponding to each of the multiple battery cells, wherein the individual spatial features characterize the position of the multiple battery cells in the target battery pack and their impact on the electrical performance of the corresponding individual cells; and generating the multidimensional feature map based on the individual spatial features corresponding to each of the multiple battery cells.

7. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores multiple instructions adapted for loading and execution by a processor of the battery state determination method according to any one of claims 1 to 5.

8. An electronic device, characterized in that, include: One or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors cause the one or more processors to implement the battery state determination method according to any one of claims 1 to 5.

Images