Battery capacity checking method based on edge calculation and related equipment

Abstract

The embodiment of the invention provides a battery capacity checking method based on edge calculation and related equipment, and belongs to the technical field of edge calculation. The method comprises the following steps: performing local preprocessing on acquired running state data to obtain preprocessed running state data; inputting the preprocessed operation state data into a lightweight capacity checking model deployed on an edge calculation node, performing local calculation through the lightweight capacity checking model, and outputting an initial health state evaluation result corresponding to the storage battery pack; and according to the environment data, correcting the initial health state evaluation result through an environment correction model to obtain a corrected capacity checking result, determining an early warning level according to the corrected capacity checking result, and obtaining early warning information according to the corrected capacity checking result based on a dual-mode communication strategy. And uploading the corrected capacity checking result and / or early warning information to a far-end operation and maintenance platform. According to the embodiment of the invention, real-time, high-precision, high-adaptability and high-reliability battery capacity checking can be realized.

Description

Technical Field

[0001] This application relates to the field of edge computing technology, and in particular to a battery capacity balancing method and related equipment based on edge computing. Background Technology

[0002] Batteries in remote, high-altitude substations serve as the core emergency power source for the power system, and their capacity adequacy directly impacts the grid's fault tolerance and power supply reliability. Currently, battery capacity monitoring primarily relies on two methods: offline capacity assessment and cloud-based online capacity assessment. The offline method requires on-site disassembly and full-charge / discharge testing, a time-consuming and labor-intensive process that affects battery life. The cloud-based method, however, suffers from data latency and large capacity assessment errors due to unstable communication and insufficient adaptability to environmental parameters in these remote areas. Because current battery capacity assessment technologies are inefficient and inaccurate, and have accounted for a significant proportion of related grid accidents in the past five years, developing a technology capable of adapting to extreme environments and achieving localized, real-time, high-precision capacity assessment has become a core and urgent need for operation and maintenance in this field. Summary of the Invention

[0003] The main objective of this application is to propose a battery capacity balancing method and related equipment based on edge computing, which can achieve real-time, high-precision, and highly reliable battery capacity balancing with strong adaptability.

[0004] To achieve the above objectives, one aspect of this application proposes a battery capacity balancing method based on edge computing, applied to an electronic device deployed at an edge computing node in a substation. The method includes: The operating status data of the battery packs in the substation are collected in real time through the edge sensing layer; the operating status data includes at least the individual cell voltage, charging and discharging current and environmental data. The collected operational status data is preprocessed locally to obtain preprocessed operational status data; The preprocessed operating status data is input into the lightweight core capacity model deployed on the edge computing node. The lightweight core capacity model performs local calculations and outputs the initial health status assessment result corresponding to the battery pack. Based on the environmental data, the initial health status assessment result is corrected using an environmental correction model to obtain a corrected capacity result. The warning level is determined based on the corrected capacity result, and warning information is obtained based on the corrected capacity result using a dual-mode communication strategy. The corrected capacity result and / or the warning information are then uploaded to a remote operation and maintenance platform.

[0005] In some embodiments, the step of performing local preprocessing on the collected operating status data to obtain preprocessed operating status data includes: The voltage and current data in the operating status data are denoised using the Kalman filter algorithm to obtain the denoised voltage and current data. Based on a unified timestamp, the noise-reduced voltage data and the noise-reduced current data are time-aligned and fused with the environmental data.

[0006] In some embodiments, the lightweight capacity model is a fusion model of a convolutional neural network and a long short-term memory network; the step of inputting the preprocessed operating status data into the lightweight capacity model deployed on the edge computing node, performing local calculations through the lightweight capacity model, and outputting the initial health status assessment result corresponding to the battery pack includes: The local features of the voltage curve in the preprocessed running data are extracted using the convolutional neural network in the lightweight core capacity model. The remaining charge status of the battery pack is calculated by using the long short-term memory network in the lightweight core capacity model and combining it with the ampere-hour integral algorithm for time series analysis. By integrating the local features of the voltage curve with the remaining charge status of the battery pack, the initial health status assessment result of the battery pack is obtained.

[0007] In some embodiments, the environmental data includes ambient temperature and ambient air pressure; the step of correcting the initial health status assessment result based on the environmental data using an environmental correction model to obtain the corrected nuclear capacity result includes: Based on the ambient temperature, the initial health status assessment result is corrected using a low-temperature correction factor to obtain the corrected initial health status assessment result. Based on the ambient air pressure, the modified initial health status assessment result is corrected using a low air pressure correction factor to obtain the modified nuclear capacity result.

[0008] In some embodiments, the formula for calculating the low-temperature correction factor is as follows: , wherein The low temperature correction factor is the... The ambient temperature is given; the formula for calculating the low-pressure correction factor is as follows: , wherein The low-pressure correction factor, the The ambient air pressure is [value missing].

[0009] In some embodiments, the method of obtaining early warning information based on the modified capacity result using a dual-mode communication strategy includes: If the remaining power status in the corrected capacity result is greater than or equal to the first threshold, it is determined to be a normal state and no warning is triggered. If the remaining battery level is less than the first threshold and greater than or equal to the second threshold, it is determined to be a warning state, triggering a platform reminder; If the remaining battery level is less than the second threshold, it is determined to be an emergency state, and both platform alarms and SMS alarms are triggered simultaneously; wherein, the first threshold is greater than the second threshold.

[0010] In some embodiments, uploading the corrected capacity result and / or the warning information to the remote operation and maintenance platform includes: Monitor the current network communication status of the electronic device; If the network communication link is normal, the complete corrected capacity result and / or the early warning information will be uploaded to the remote operation and maintenance platform through the network communication link; If the network communication link is interrupted, it will automatically switch to BeiDou short message communication and only upload compressed data packets containing key early warning information to the remote operation and maintenance platform.

[0011] To achieve the above objectives, another aspect of this application proposes a battery capacity balancing device based on edge computing, applied to an electronic device deployed at an edge computing node in a substation. The device includes: The data acquisition module is used to collect real-time operating status data of the battery packs in the substation through the edge sensing layer; the operating status data includes at least individual cell voltage, charging and discharging current and environmental data; The data processing module is used to perform local preprocessing on the collected operating status data to obtain preprocessed operating status data. The status calculation module is used to input the preprocessed operating status data into the lightweight core capacity model deployed on the edge computing node, perform local calculations through the lightweight core capacity model, and output the initial health status assessment result corresponding to the battery pack. The capacity correction module is used to correct the initial health status assessment result based on the environmental data using an environmental correction model to obtain a corrected capacity result, determine the warning level based on the corrected capacity result, obtain warning information based on the corrected capacity result using a dual-mode communication strategy, and upload the corrected capacity result and / or the warning information to a remote operation and maintenance platform.

[0012] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described above.

[0013] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described above.

[0014] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer program product, including a computer program that, when executed by a processor, implements the aforementioned method.

[0015] The embodiments of this application include at least the following beneficial effects: This application provides a battery capacity assessment method, device, electronic device, storage medium, and program product based on edge computing. The solution includes: real-time acquisition of operating status data of battery packs in substations through an edge sensing layer; the operating status data includes at least single-cell voltage, charging and discharging current, and environmental data; local preprocessing of the acquired operating status data to obtain preprocessed operating status data; inputting the preprocessed operating status data into a lightweight capacity assessment model deployed on an edge computing node, performing local calculations through the lightweight capacity assessment model, and outputting the initial health status assessment result corresponding to the battery pack; correcting the initial health status assessment result based on environmental data using an environmental correction model to obtain a corrected capacity assessment result, determining the warning level based on the corrected capacity assessment result, obtaining warning information based on the corrected capacity assessment result using a dual-mode communication strategy, and uploading the corrected capacity assessment result and / or warning information to a remote operation and maintenance platform. Implementing the embodiments of this application, the preprocessed operating status data is input into a lightweight capacity model deployed on an edge computing node. This lightweight capacity model can run efficiently on limited edge computing resources and directly output the initial health status assessment results corresponding to the battery pack. An environmental correction model is introduced to compensate and correct the initial health status assessment results for extreme environmental factors unique to remote and impoverished areas, such as low temperature and low air pressure, thereby obtaining more accurate capacity assessment results. The warning level is automatically determined based on the corrected capacity assessment results, realizing the localization of the battery capacity assessment process in remote and harsh areas, completely eliminating the dependence on stable cloud communication, and significantly improving the assessment accuracy under extreme conditions through closed-loop correction of environmental factors, thereby achieving real-time, high-precision, and highly reliable battery capacity assessment. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of an implementation environment provided in an embodiment of this application; Figure 2 This is a flowchart of the battery capacity grading method based on edge computing provided in the embodiments of this application; Figure 3 This is a system schematic diagram of an edge computing node in one embodiment; Figure 4This is a flowchart of a battery capacity calculation based on edge computing in one embodiment; Figure 5 This is a schematic diagram of the structure of the battery capacity device based on edge computing provided in the embodiments of this application; Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0018] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”

[0019] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0021] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0022] 1) SOC (State of Charge) refers to the state of charge of a battery, which represents the ratio of the battery's current remaining capacity to its rated capacity, usually expressed as a percentage; 2) Convolutional Neural Networks (CNNs) are a type of deep learning model that has achieved remarkable success in many fields, including image and video recognition, image classification, and medical image analysis. 3) Long Short-Term Memory (LSTM) networks are a special type of recurrent neural network that can learn long-term dependent information. The key to LSTM lies in its internal structure, which includes three gates (input gate, forget gate, and output gate) and a cell state. These structures enable LSTM to effectively retain long-term information and ignore irrelevant information when processing sequential data.

[0023] In related technologies, batteries in substations in remote, high-altitude, and borderless areas serve as the "core emergency power source" of the power system. They are mainly used to supply power to critical equipment such as protection devices, circuit breaker operation and closing, and emergency lighting during power outages. Their adequacy directly determines the grid's fault tolerance and power supply reliability. Currently, the industry mainly relies on two types of related technologies for monitoring (capacity verification) of battery capacity.

[0024] The first related technology is offline capacity assessment, which requires manual on-site removal of the battery from the system and measurement of the actual capacity through a "full charge-full discharge" method. This solution requires interrupting the normal operation of the battery, and in remote and inaccessible areas with poor transportation (such as plateau mountains where a one-way trip takes ≥6 hours), the average capacity assessment time per station is 24-48 hours, resulting in high labor costs (requiring 2-3 people per trip), and frequent charging and discharging will shorten battery life. The second related technology is cloud-based online capacity assessment. This involves collecting battery voltage, current, and temperature data using sensors and uploading it to a remote cloud server, where cloud algorithms calculate the capacity. However, this technology faces two major bottlenecks in remote, high-altitude, and remote areas: First, communication is unstable (4G signal coverage is less than 30%, with some areas having no signal), resulting in data upload delays of ≥30 seconds, or even disconnections that render the capacity assessment results invalid. Second, environmental adaptability is poor—low air pressure at high altitudes reduces electrolyte ion mobility, and low temperatures (≤-20℃) increase battery internal resistance. Traditional cloud models do not incorporate these environmental parameters, resulting in capacity assessment errors as high as 15%-20%, making it impossible to accurately determine the battery's health status. According to industry statistics, in the past five years, 22% of power grid accidents in remote, high-altitude, and remote substation areas were caused by untimely or inaccurate battery capacity assessment. Among these, 80% were directly related to "low efficiency of offline capacity assessment" and "large delay / error of cloud-based capacity assessment." Therefore, developing a battery capacity assessment technology that is adaptable to extreme environments, localized, real-time, and highly accurate has become a core requirement for solving the operation and maintenance challenges of substations in remote, high-altitude, and remote substation areas.

[0025] In view of this, this application provides a battery capacity calculation method based on edge computing, which can achieve real-time, high-precision and highly reliable battery capacity calculation.

[0026] The battery capacity balancing method based on edge computing provided in this application relates to the field of edge computing technology. This method can be applied to terminals, servers, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the battery capacity balancing method based on edge computing, but is not limited to the above forms.

[0027] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0028] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0029] like Figure 1The diagram shown is a schematic representation of an implementation environment provided in an embodiment of this application. (Refer to...) Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected via a network, either wirelessly or via a wired connection, to complete data transmission and exchange.

[0030] Server 101 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0031] Additionally, server 101 can also be a node server in a blockchain network. Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms.

[0032] Terminal 102 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. Terminal 102 and server 101 can be directly or indirectly connected via wired or wireless communication, and this embodiment of the application does not impose any limitations.

[0033] For example, based on Figure 1 The implementation environment shown in this application embodiment provides a battery capacity integration method based on edge computing. The following description uses the application of this edge computing-based battery capacity integration method in server 101 as an example. It can be understood that this edge computing-based battery capacity integration method can also be applied in terminal 102.

[0034] Figure 2 This is an optional flowchart of the battery capacity balancing method based on edge computing provided in the embodiments of this application. The execution subject of the battery capacity balancing method based on edge computing can be any of the aforementioned electronic devices (including servers or terminals). Figure 2 The method may include, but is not limited to, steps S201 to S204.

[0035] Step S201: Real-time acquisition of the operating status data of the substation's battery pack through the edge sensing layer; the operating status data includes at least individual cell voltage, charging and discharging current, and environmental data.

[0036] In some embodiments, electronic devices deployed at edge computing nodes in the substation site can continuously collect operational status data of the battery bank. This data specifically covers key electrical parameters reflecting the internal chemical state of the battery, such as the individual cell voltage of each battery cell, the real-time charging and discharging current in the circuit, and environmental data that significantly affects battery performance, such as ambient temperature and air pressure. The edge sensing layer deployed at the substation site forms the physical basis for real-time data acquisition. This edge sensing layer can be various sensors communicating with the electronic devices, as well as tools used to record the battery's operational status data. Figure 3 A system schematic diagram of an edge computing node in one embodiment, such as Figure 3 As shown, the system includes an edge perception layer, an edge computing layer, a dual-mode communication layer, and an operation and maintenance interaction layer. It does not require modification of existing battery equipment and achieves real-time capacity verification through "localized data acquisition - localized computing - on-demand transmission".

[0037] The main function of the edge sensing layer is to monitor the status of the battery pack, including the voltage, temperature, and internal resistance of each cell, and to perform charging and discharging of the battery pack. It consists of a charging module, a battery monitoring host, a battery monitoring module, and an inverter module. The charging module is used to charge the battery after it has been discharged to prevent the battery from becoming unusable due to long-term undercharging. The battery host and battery monitoring modules are mainly used to monitor the status of the battery pack. Each cell is equipped with a battery monitoring module to collect the temperature, voltage, and current data of a single cell. The modules communicate with each other using RS-485 and transmit the data of the entire battery pack to the battery monitoring host in a daisy-chain manner. The inverter module is mainly used for discharging, converting the discharged DC power from the battery pack into AC power and feeding it back into the AC power grid.

[0038] The edge computing layer employs low-power embedded edge nodes (such as the Rockchip RK3588 processor, with a computing power of 6 TOPS and a power consumption of ≤15W), and integrates a data preprocessing module, a lightweight core capacity model, and an environment correction module to achieve localized data processing. The dual-mode communication layer enables stable data transmission using "4G + BeiDou short message" dual-mode communication. When the 4G signal is normal, the edge computing layer uploads core capacity results (such as SOC=75%, SOH=80%) and device status (such as whether sensors are online) to the remote operation and maintenance platform via 4G, with a transmission latency of ≤1 second.

[0039] The operations and maintenance interaction layer can perform tiered early warnings. Based on the corrected capacity control results, it is divided into 3 levels of early warning: Normal (SOC≥80%): No warning, only data is recorded; Warning (50%≤SOC<80%): "Observe carefully" reminder is pushed through the operations and maintenance platform; Emergency (SOC<50%): Platform alarm and SMS alarm are triggered at the same time to notify operations and maintenance personnel to handle on-site.

[0040] Step S202: Perform local preprocessing on the collected operating status data to obtain preprocessed operating status data.

[0041] In some embodiments, the Kalman filter algorithm is used to denoise the voltage and current data in the operating status data, resulting in denoised voltage and current data. Based on a unified timestamp, the denoised voltage and current data are time-aligned and fused with the environmental data. Kalman filtering is an optimal estimation algorithm based on a state-space model. It can dynamically predict, correct, and fuse measured values ​​recursively, effectively filtering out sensor noise and random interference, thus obtaining smooth and reliable denoised voltage and current data. Based on a unified high-precision timestamp, these three types of heterogeneous data—denoised voltage and current data—can be strictly time-aligned and deeply fused with the environmental data to construct a standardized dataset that is time-consistent and multi-dimensionally correlated.

[0042] Furthermore, in terms of noise suppression, a Kalman filter algorithm is used to construct a data correction model. Its core principle is to predict the theoretical values ​​of the data through state equations, and then calculate the deviation between the actual collected voltage and current data and the predicted values ​​using observation equations. The output results are continuously corrected through recursive optimal estimation. This process effectively filters out random interference such as instantaneous current spikes and temperature drift caused by power grid impacts, improving the data signal-to-noise ratio to over 40dB and ensuring the stability of the input data. In the data alignment stage, voltage, current, and ambient temperature and pressure data are synchronously calibrated based on a unified timestamp benchmark. A timestamp matching algorithm controls the acquisition time difference between different sensors to within ≤10ms, avoiding calculation deviations caused by inconsistent data acquisition timing and laying the foundation for multi-dimensional data fusion analysis.

[0043] Taking a 2V / 300Ah battery as an example, a high-altitude substation collected a set of current data in a low-temperature environment: [8A, 35A (grid impulse interference), 7.5A, 8.2A]. Using a Kalman filter algorithm, the theoretical value was predicted based on historical current trends (average 7.8A). After comparison with the actual collected values, the theoretical value was corrected, filtering out the instantaneous interference of 35A, and outputting smooth data [7.9A, 8.1A, 7.6A, 8.0A], improving the signal-to-noise ratio from 28dB to 43dB. Simultaneously, the voltage (collected at 09:15:00.002), current (collected at 09:15:00.007), and ambient temperature and pressure (collected at 09:15:00.009) data were aligned by timestamp and uniformly corrected to synchronous data at 09:15:00.000, eliminating the 2-9ms acquisition time difference and ensuring that multi-dimensional data participates in calculations under the same time reference.

[0044] By employing the Kalman filter algorithm for noise reduction and performing time alignment and fusion based on a unified timestamp, the signal-to-noise ratio and time consistency of the operational status data are significantly improved. This lays a clean and synchronized data foundation for the accurate calculation of subsequent models, thereby ensuring the reliability and robustness of the preprocessing stage of operational status data in complex field environments.

[0045] Step S203: Input the preprocessed operating status data into the lightweight core capacity model deployed on the edge computing node, perform local calculations through the lightweight core capacity model, and output the initial health status assessment results corresponding to the battery pack.

[0046] In some embodiments, the lightweight kernel capacity model is a fusion model of convolutional neural networks and long short-term memory networks, which organically combines and synergistically computes the advantages of convolutional neural networks and long short-term memory networks. The model can simultaneously extract key spatial features and temporal context information from complex and multidimensional input data, so that the lightweight kernel capacity model not only has powerful nonlinear fitting and sequence modeling capabilities, but also significantly reduces computational complexity and resource consumption through lightweight techniques such as layer simplification and parameter quantization.

[0047] As an optional implementation, the local features of the voltage curve in the preprocessed operating data are extracted by using a convolutional neural network in the lightweight capacity model; the remaining charge state of the battery pack is calculated by using a long short-term memory network in the lightweight capacity model combined with the ampere-hour integral algorithm for time series analysis; and the initial health status assessment result of the battery pack is obtained by fusing the local features of the voltage curve with the remaining charge state of the battery pack.

[0048] The convolutional neural network in the lightweight capacity model scans the preprocessed operating data, specifically extracting local features and subtle change patterns from the voltage curve to capture the instantaneous dynamics of the battery's internal reaction. These local features can be used to reflect the battery's external electrical characteristics, specifically the voltage local features. Simultaneously, the long short-term memory network in the lightweight capacity model leverages its powerful time-series modeling capabilities, combined with the classic ampere-hour integral algorithm for cumulative battery charge calculation, to perform continuous time-series analysis on the charge and discharge current data. This allows for the calculation of the battery pack's remaining state of charge, which characterizes the battery's internal energy state. Furthermore, electronic devices can employ feature-level fusion strategies to deeply fuse and jointly analyze the local features of the voltage curve with the battery pack's remaining state of charge, generating an initial health status assessment result for the battery pack.

[0049] The lightweight capacity model employs a hybrid architecture of "1-layer CNN + 2-layer LSTM," maintaining computational accuracy while keeping the model size ≤50MB and inference time ≤500ms, perfectly suited for low-computing-power deployments at edge nodes. The convolutional neural network layer focuses on extracting local features from the voltage curve. By sliding the convolutional kernel through voltage time-series data, it captures key features strongly correlated with capacity decay, such as voltage inflection points during discharge and voltage plateaus during charging, achieving precise local characterization of the battery state.

[0050] The Long Short-Term Memory (LSTM) network layer focuses on trend analysis of time-series data, using its gating mechanism to memorize long-term information such as the rate of change of voltage and the cumulative current over the past hour. It also integrates the ampere-hour integration method to complete the preliminary SOC (State of Charge) calculation. The core formula of the ampere-hour integration method is:

[0051] In the formula, This indicates the remaining battery level. I(t) represents the nominal capacity of the battery, and I(t) represents the charging and discharging current (positive for discharging and negative for charging). The interval from t to t represents the calculation time interval.

[0052] By deeply fusing local features extracted by CNN with time-series trends analyzed by LSTM, a preliminary assessment result of the remaining battery capacity is finally output.

[0053] Taking a 2V 300Ah battery as an example, the CNN layer can extract voltage curve features—in a healthy battery, the inflection point where the voltage drops from 2.1V to 1.8V occurs after 20 hours of discharge, while in this battery, the inflection point occurs earlier, at 16 hours. The CNN layer quantifies this feature as a "degradation coefficient of 0.2" (representing a 20% capacity degradation). The LSTM layer can then use an ampere-hour integral algorithm to calculate the initial state of charge (SOC). Assuming an initial... Average discharge current over the past 5 hours Substitute into the formula:

[0054] The attenuation coefficients of the CNN layers further correct the LSTM results, ultimately yielding the preliminary kernel capacity result as follows:

[0055] By combining deep learning feature extraction with traditional power algorithms, this approach not only fully leverages the respective advantages of convolutional neural networks in spatial feature mining and long short-term memory networks in time-dependent modeling, but also enhances the robustness of the assessment through information complementarity. This enables highly reliable and accurate preliminary assessment of real-time battery health status even on resource-constrained edge devices.

[0056] Step S204: Based on the environmental data, the initial health status assessment results are corrected using an environmental correction model to obtain the corrected capacity results. The warning level is determined based on the corrected capacity results, and the warning information is obtained based on the corrected capacity results using a dual-mode communication strategy. The corrected capacity results and / or warning information are then uploaded to the remote operation and maintenance platform.

[0057] In some embodiments, the initial health status assessment result is corrected by a low temperature correction factor based on the ambient temperature to obtain a corrected initial health status assessment result; the corrected initial health status assessment result is corrected by a low pressure correction factor based on the ambient air pressure to obtain a corrected nuclear capacity result.

[0058] To achieve accurate assessment of battery health status, after obtaining the initial health status assessment results, the electronic device further utilizes a specially constructed environmental correction model to eliminate the interference caused by extreme environments on battery performance characterization. The environmental correction model uses a low-temperature correction factor to initially correct the initial results based on real-time ambient temperature data. This low-temperature correction factor is a compensation function calibrated based on the battery's low-temperature electrochemical characteristics and extensive experimental data. Its core function is to quantify and offset the effects of low temperature on battery effective capacity decay and increased internal resistance, outputting a temperature-compensated intermediate assessment result. The electronic device can also use a low-pressure correction factor to further refine the intermediate result (the corrected initial health status assessment result) based on ambient air pressure. The low-pressure correction factor aims to simulate and compensate for the systematic impact of changes in electrolyte ion mobility under high-altitude conditions.

[0059] Furthermore, considering the extreme environmental characteristics of low pressure and low temperature in remote, high-altitude areas, a temperature-pressure coupling correction factor can be introduced to perform a secondary calibration of the preliminary capacity test results to eliminate the influence of environmental factors on battery performance. The formula is: in, This is the final corrected state of charge. This is a low-temperature correction factor. This is a low-pressure correction factor.

[0060] Through this compensation correction combining ambient temperature and atmospheric pressure, the environmental correction model ultimately outputs a capacity result that is comparable to a standard environmental benchmark and accurately reflects the battery's true aging state, thereby significantly improving the accuracy and reliability of capacity assessment in harsh environments such as extreme cold and low oxygen.

[0061] Furthermore, the formula for calculating the low-temperature correction factor is as follows: ,in, This is a low-temperature correction factor. The ambient temperature is used; the formula for calculating the low-pressure correction factor is: ,in, This is a low-pressure correction factor. This refers to ambient air pressure.

[0062] Low temperature correction factor The design is based on the nonlinear relationship between battery internal resistance and temperature; low temperatures lead to increased battery internal resistance and a faster rate of voltage drop during discharge. When T = -40℃, This low-temperature correction factor can effectively correct the problem of inflated battery capacity at low temperatures.

[0063] Low pressure correction factor This addresses the inhibitory effect of low-pressure environments on electrolyte ion mobility. Low pressure reduces the actual discharge capacity of the battery. When P = 50 kPa (corresponding to an altitude of approximately 5000 m), It can accurately compensate for capacity assessment deviations caused by low air pressure.

[0064] Preliminary results based on a 2V / 300Ah battery After adjustments based on environmental parameters (T=-40℃, P=50kPa, corresponding to an altitude of approximately 5000m), the final capacity result is as follows:

[0065] Through temperature and pressure coupling correction, the capacity error is reduced from 15%-20% in traditional schemes to ≤5%, achieving high-precision capacity assessment under extreme environments.

[0066] In some embodiments, if the remaining power status in the corrected capacity result is greater than or equal to a first threshold, it is determined to be a normal state and no warning is triggered; if the remaining power status is less than the first threshold but greater than or equal to a second threshold, it is determined to be a warning state and a platform reminder is triggered; if the remaining power status is less than the second threshold, it is determined to be an emergency state and both a platform alarm and an SMS alarm are triggered; wherein, the first threshold is greater than the second threshold.

[0067] Electronic devices can compare the remaining battery status in the corrected capacity result with a first threshold. If the status value is greater than or equal to this threshold, the battery pack is determined to be in normal condition, without triggering any warnings. This ensures silent monitoring of healthy batteries and avoids unnecessary interference. If the remaining battery status is below the first threshold but has not yet fallen to the second threshold, the battery is determined to enter a warning state. At this time, a platform reminder will be triggered to notify remote maintenance personnel to pay attention to its degradation trend, facilitating advance inspection or maintenance, reflecting the concept of preventative maintenance. Once the remaining battery status further decreases and falls below the even lower second threshold, the system will determine an emergency state. At this time, not only will a high-level alarm be generated on the platform, but an SMS alarm will also be initiated simultaneously. Specifically, the first threshold can be set to 80%, and the second threshold can be set to 50%, with no specific limitation.

[0068] By using a dual-link system to ensure that alarm information is delivered promptly and reliably, it drives maintenance personnel to take emergency intervention measures. Based on a progressive judgment logic with dual thresholds, it achieves seamless connection and accurate response from "normal monitoring" to "early warning" and then to "emergency alarm", which significantly improves the precision of risk management and the timeliness of emergency response.

[0069] In some embodiments, the current network communication status of the electronic device is monitored. If the network communication link is normal, the complete and corrected capacity verification results and / or early warning information are uploaded to the remote operation and maintenance platform via the network communication link. If the network communication link is interrupted, it automatically switches to BeiDou short message communication, uploading only compressed data packets containing key early warning information to the remote operation and maintenance platform. Real-time monitoring of the current network communication link status of the electronic device serves as the basis for communication path selection decisions. When a normal network communication link is detected, this high-bandwidth link is prioritized to efficiently upload the complete and corrected capacity verification results and related early warning information to the remote operation and maintenance platform, ensuring the richness and completeness of the uploaded data. Once a normal network link is detected to be interrupted or the signal deteriorates to the point of unavailability, it automatically switches to BeiDou short message communication mode immediately without manual intervention. BeiDou short message service is a two-way message communication service based on my country's BeiDou satellite navigation system. Its unique advantage is that it does not rely on ground base stations and can achieve wide-area communication coverage around the world. In this mode, core warning information such as warning level and key power status will be efficiently encoded and compressed to generate extremely simple data packets and sent through satellite links, ensuring that even in the worst communication environment, the most critical status alarms can be reliably transmitted to the operation and maintenance platform.

[0070] The dual-mode collaborative communication mechanism not only makes full use of the economy and high bandwidth of the terrestrial network, but also provides a blind-spot-free communication guarantee through BeiDou short messages, thereby achieving a seamless connection from full transmission with network access to precise transmission without network access, and thoroughly ensuring the ultimate reliability of the upload of capacity assessment results and early warning information.

[0071] Figure 4 A flowchart of battery capacity calculation based on edge computing is provided in one embodiment, such as... Figure 4 As shown, after acquiring the raw voltage, current, and ambient temperature and pressure data, the electronic device enters the data preprocessing stage. Kalman filtering is used to reduce noise and filter out transient interference, and timestamp alignment is performed to ensure a time difference of less than or equal to 10ms. Next, the model calculation stage begins. A CNN layer extracts local features of the voltage curve, followed by LSTM layer processing for time-series trend analysis and ampere-hour integration. Feature fusion then outputs a preliminary SOCInitial. Afterward, it is determined whether the environment is extreme. If not, the final result equals the preliminary result; otherwise, environmental correction is performed using a low-temperature correction factor Kt and a low-pressure correction factor Kp, through a temperature and pressure coupling correction formula. After obtaining the corrected result, the final capacity result is output.

[0072] Steps S201 to S204 as illustrated in this embodiment include: real-time acquisition of operating status data of the substation's battery bank through an edge sensing layer; the operating status data includes at least individual cell voltage, charging and discharging current, and environmental data; local preprocessing of the acquired operating status data to obtain preprocessed operating status data; inputting the preprocessed operating status data into a lightweight capacity model deployed on an edge computing node, performing local calculations through the lightweight capacity model, and outputting the initial health status assessment result corresponding to the battery bank; correcting the initial health status assessment result based on environmental data using an environmental correction model to obtain a corrected capacity result, determining the warning level based on the corrected capacity result, and obtaining warning information based on the corrected capacity result using a dual-mode communication strategy, and then connecting the corrected capacity result with... Alternatively, early warning information can be uploaded to a remote operation and maintenance platform. The pre-processed operational status data can be input into a lightweight capacity model deployed on edge computing nodes. This lightweight capacity model can run efficiently on limited edge computing resources and directly output the initial health status assessment results corresponding to the battery pack. An environmental correction model is introduced to compensate for and correct the initial health status assessment results for extreme environmental factors unique to remote and remote areas, such as low temperature and low air pressure, thereby obtaining more accurate capacity assessment results. The warning level is automatically determined based on the corrected capacity assessment results. This realizes the localization of the battery capacity assessment process in remote and harsh areas, completely eliminating the dependence on stable cloud communication. Through closed-loop correction of environmental factors, the assessment accuracy under extreme conditions is significantly improved, thereby enabling real-time, high-precision, and highly reliable battery capacity assessment.

[0073] Please see Figure 5 This application also provides a battery capacity integration device based on edge computing, which can implement the above method. The device includes: The data acquisition module 501 is used to collect real-time operating status data of the battery pack in the substation through the edge sensing layer; the operating status data includes at least the individual cell voltage, charging and discharging current and environmental data; The data processing module 502 is used to perform local preprocessing on the collected operating status data to obtain preprocessed operating status data. The state calculation module 503 is used to input the preprocessed operating state data into the lightweight core capacity model deployed on the edge computing node, perform local calculations through the lightweight core capacity model, and output the initial health state assessment results corresponding to the battery pack. The capacity correction module 504 is used to correct the initial health status assessment results based on environmental data and through an environmental correction model to obtain the corrected capacity results, determine the warning level based on the corrected capacity results, obtain warning information based on the corrected capacity results using a dual-mode communication strategy, and upload the corrected capacity results and / or warning information to the remote operation and maintenance platform.

[0074] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0075] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0076] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0077] Please see Figure 6 , Figure 6 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 601 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 602 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 602 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 602 and is called and executed by the processor 601 using the methods described in the embodiments of this application. The input / output interface 603 is used to implement information input and output; The communication interface 604 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 605 transmits information between various components of the device (e.g., processor 601, memory 602, input / output interface 603, and communication interface 604); The processor 601, memory 602, input / output interface 603, and communication interface 604 are connected to each other within the device via bus 605.

[0078] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0079] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0080] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0081] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0082] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0083] The battery capacity assessment method, apparatus, electronic device, storage medium, and program product based on edge computing provided in this application collect real-time operating status data of battery banks in substations through an edge sensing layer. The operating status data includes at least individual cell voltage, charging and discharging current, and environmental data. The collected operating status data is preprocessed locally to obtain preprocessed operating status data. The preprocessed operating status data is input into a lightweight capacity assessment model deployed on an edge computing node. The lightweight capacity assessment model performs local calculations and outputs the initial health status assessment result corresponding to the battery bank. Based on the environmental data, the initial health status assessment result is corrected through an environmental correction model to obtain a corrected capacity assessment result. The warning level is determined based on the corrected capacity assessment result, and the warning information is obtained based on the corrected capacity assessment result using a dual-mode communication strategy. The corrected capacity assessment result and / or warning information are uploaded to a remote operation and maintenance platform, enabling real-time, high-precision, and highly adaptable high-reliability battery capacity assessment.

[0084] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0085] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0086] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0087] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0088] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application 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 this application 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 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.

[0089] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0090] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0091] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0092] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0093] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A battery capacity assessment method based on edge computing, characterized in that, Applied to electronic devices deployed at edge computing nodes in a substation, the method includes the following steps: The operating status data of the battery packs in the substation are collected in real time through the edge sensing layer; the operating status data includes at least the individual cell voltage, charging and discharging current and environmental data. The collected operational status data is preprocessed locally to obtain preprocessed operational status data; The preprocessed operating status data is input into the lightweight core capacity model deployed on the edge computing node. The lightweight core capacity model performs local calculations and outputs the initial health status assessment result corresponding to the battery pack. Based on the environmental data, the initial health status assessment result is corrected using an environmental correction model to obtain a corrected capacity result. The warning level is determined based on the corrected capacity result, and warning information is obtained based on the corrected capacity result using a dual-mode communication strategy. The corrected capacity result and / or the warning information are then uploaded to a remote operation and maintenance platform.

2. The method according to claim 1, characterized in that, The step of performing local preprocessing on the collected operating status data to obtain preprocessed operating status data includes: The voltage and current data in the operating status data are denoised using the Kalman filter algorithm to obtain the denoised voltage and current data. Based on a unified timestamp, the noise-reduced voltage data and the noise-reduced current data are time-aligned and fused with the environmental data.

3. The method according to claim 1 or 2, characterized in that, The lightweight capacity model is a fusion model of convolutional neural networks and long short-term memory networks; the preprocessed operating status data is input into the lightweight capacity model deployed on the edge computing node, and the lightweight capacity model performs local calculations to output the initial health status assessment results corresponding to the battery pack, including: The local features of the voltage curve in the preprocessed running data are extracted using the convolutional neural network in the lightweight core capacity model. The remaining state of charge of the battery pack is calculated by using the long short-term memory network in the lightweight core capacity model and combining it with the ampere-hour integral algorithm for time series analysis. By integrating the local features of the voltage curve with the remaining charge status of the battery pack, the initial health status assessment result of the battery pack is obtained.

4. The method according to claim 1, characterized in that, The environmental data includes ambient temperature and ambient air pressure; the process of correcting the initial health status assessment results based on the environmental data using an environmental correction model to obtain the corrected core capacity results includes: Based on the ambient temperature, the initial health status assessment result is corrected using a low-temperature correction factor to obtain the corrected initial health status assessment result. Based on the ambient air pressure, the modified initial health status assessment result is corrected using a low air pressure correction factor to obtain the modified nuclear capacity result.

5. The method according to claim 4, characterized in that, The formula for calculating the low-temperature correction factor is as follows: , wherein The low temperature correction factor is the... The ambient temperature is given; the formula for calculating the low-pressure correction factor is as follows: , wherein The low-pressure correction factor, the The ambient air pressure is [value missing].

6. The method according to claim 1, characterized in that, The early warning information obtained based on the dual-mode communication strategy according to the corrected capacity result includes: If the remaining power status in the corrected capacity result is greater than or equal to the first threshold, it is determined to be in a normal state and no warning is triggered; If the remaining battery level is less than the first threshold and greater than or equal to the second threshold, it is determined to be a warning state, triggering a platform reminder; If the remaining battery level is less than the second threshold, it is determined to be an emergency, and both platform alarms and SMS alarms are triggered simultaneously; wherein, the first threshold is greater than the second threshold.

7. The method according to claim 1, characterized in that, Uploading the corrected capacity result and / or the early warning information to the remote operation and maintenance platform includes: Monitor the current network communication status of the electronic device; If the network communication link is normal, the complete corrected capacity result and / or the early warning information will be uploaded to the remote operation and maintenance platform through the network communication link; If the network communication link is interrupted, it will automatically switch to BeiDou short message communication and only upload compressed data packets containing key early warning information to the remote operation and maintenance platform.

8. A battery capacity integration device based on edge computing, characterized in that, The device is applied to electronic equipment deployed at an edge computing node in a substation, and includes: The data acquisition module is used to collect real-time operating status data of the battery packs in the substation through the edge sensing layer; the operating status data includes at least individual cell voltage, charging and discharging current and environmental data; The data processing module is used to perform local preprocessing on the collected operating status data to obtain preprocessed operating status data. The status calculation module is used to input the preprocessed operating status data into the lightweight core capacity model deployed on the edge computing node, perform local calculations through the lightweight core capacity model, and output the initial health status assessment result corresponding to the battery pack. The capacity correction module is used to correct the initial health status assessment result based on the environmental data using an environmental correction model to obtain a corrected capacity result, determine the warning level based on the corrected capacity result, obtain warning information based on the corrected capacity result using a dual-mode communication strategy, and upload the corrected capacity result and / or the warning information to a remote operation and maintenance platform.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

Images