Energy storage power station operation and maintenance method and device

Abstract

This disclosure relates to the field of power system energy storage technology, and discloses an operation and maintenance method and apparatus for an energy storage power station. The method includes acquiring operating status data of the energy storage power station; preprocessing the operating status data to obtain preprocessed data, wherein the preprocessing includes local preprocessing of the operating status data, which includes data fusion processing and data storage processing of the operating status data; and performing data analysis on the preprocessed data based on the preprocessing model to obtain the operating status assessment results of the energy storage power station. This disclosure integrates multi-source heterogeneous data through data fusion processing and combines blockchain storage to ensure data credibility, solving the problems of poor data coordination and insufficient anti-tampering capabilities in traditional methods, and improving operation and maintenance efficiency and security.

Description

Technical Field

[0001] This disclosure relates to the field of power system energy storage technology, specifically to operation and maintenance methods and devices for energy storage power stations. Background Technology

[0002] As a key facility for the flexible regulation of power systems, energy storage power stations are becoming increasingly larger and more complex. Traditional operation and maintenance of energy storage power stations present several challenges.

[0003] In related technologies, single-point monitoring is often used. However, differences in sampling frequency, accuracy, and response time among different sensors lead to poor data coordination, with individual parameter monitoring lagging behind actual fault occurrence. Furthermore, the operational data of these technologies is highly susceptible to tampering. These issues negatively impact the operational efficiency and reliability of energy storage power stations. Summary of the Invention

[0004] This disclosure provides an operation and maintenance method for an energy storage power station, including: acquiring operating status data of the energy storage power station; preprocessing the operating status data to acquire preprocessed data, wherein the preprocessing includes local preprocessing of the operating status data, and the local preprocessing includes data fusion processing and data storage processing of the operating status data; and performing data analysis on the preprocessed data based on the preprocessing model to obtain the operating status assessment results of the energy storage power station.

[0005] Secondly, this disclosure provides an operation and maintenance device for an energy storage power station. The device includes: a data acquisition module for acquiring operating status data of the energy storage power station; a data preprocessing module for preprocessing the operating status data to acquire preprocessed data, wherein the preprocessing includes local preprocessing of the operating status data, and the local preprocessing includes data fusion processing and data storage processing of the operating status data; and an acquisition module for performing data analysis on the preprocessed data based on the preprocessing model to acquire the operating status assessment results of the energy storage power station.

[0006] Thirdly, this disclosure provides an electronic device, including: a memory and a processor, which are interconnected and communicate with each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the above-mentioned operation and maintenance method for the energy storage power station.

[0007] As can be seen from the above, the operation and maintenance method and device for an energy storage power station provided in this disclosure integrates multi-source heterogeneous data through data fusion processing and combines blockchain notarization to ensure data credibility. This solves the problems of poor data coordination and insufficient anti-tampering capabilities in traditional methods, and has the advantages of improving operation and maintenance efficiency and security. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the specific embodiments or related technologies of this disclosure, the accompanying drawings used in the description of the specific embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic diagram of the first process of the operation and maintenance method of the energy storage power station according to an embodiment of this application; Figure 2 A second flowchart illustrating the operation and maintenance method of an energy storage power station according to an embodiment of this application; Figure 3 A schematic diagram of the third process of the operation and maintenance method of the energy storage power station according to the embodiments of this application; Figure 4 A schematic diagram of the fourth process of the operation and maintenance method of the energy storage power station according to the embodiments of this application; Figure 5 This is a structural block diagram of the operation and maintenance device of the energy storage power station according to an embodiment of this application; Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0011] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0012] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this disclosure, "a plurality of" means two or more, unless otherwise expressly specified.

[0013] In related technologies, the operation and maintenance of energy storage power stations has long relied on single-point monitoring and manual inspection, resulting in data silos and reliability issues. In a typical scenario, a 100-megawatt energy storage power station deployed multiple types of sensors for temperature, voltage, and pressure. Due to differences in the sampling frequencies of these sensors, data timestamps were misaligned, making it difficult for maintenance personnel to accurately determine whether abnormal battery pack heating was caused by internal faults or environmental factors. Furthermore, the centralized storage of critical maintenance records has led to instances of manual tampering with charge / discharge logs to cover up operational errors, severely impacting the accuracy of fault tracing.

[0014] To address the aforementioned issues, the applicant found that traditional data acquisition methods could not meet the needs of multi-parameter collaborative analysis, while centralized data processing architectures suffered from response latency and data security risks. By researching distributed edge computing frameworks, the applicant proposed implementing data cleaning and feature alignment at local nodes to ensure the timeliness of subsequent analysis. Regarding data credibility, the applicant explored introducing blockchain technology into the device status recording process to form an immutable chain of operational evidence. Ultimately, the applicant determined to construct a layered processing system comprising a local preprocessing layer and an intelligent analysis layer, achieving a dual improvement in data quality and credibility.

[0015] Therefore, this application proposes an operation and maintenance method for energy storage power stations, such as... Figure 1 As shown, the method includes the following steps: Step S101: Obtain the operating status data of the energy storage power station.

[0016] Step S102: Preprocess the running status data to obtain preprocessed data. The preprocessing includes local preprocessing of the running status data, which includes data fusion processing and data storage processing of the running status data. Data fusion processing refers to the spatiotemporal alignment and feature association of multi-source heterogeneous data. Specifically, this can be achieved using a Kalman filter algorithm combined with time series interpolation methods to eliminate data bias caused by sensor sampling delays. Data evidence storage processing refers to the tamper-proof storage of key process data. This can be implemented using a lightweight blockchain module deployed on edge computing nodes, performing hash operations on the structured data after feature extraction and generating evidence records. The preprocessing model refers to the computational model used for state assessment. Specifically, this can be implemented using a hybrid model of convolutional neural networks and long short-term memory networks trained on historical data to extract nonlinear relationships between multidimensional features.

[0017] Step S103: Perform data analysis on the preprocessed data based on the preprocessing model to obtain the operation status assessment results of the energy storage power station.

[0018] Specifically, sensors (arrays) deployed within the battery compartment collect multi-dimensional data streams, including temperature, voltage, and deformation. Edge computing devices process this multi-dimensional data. Specifically, a sliding time window mechanism is used to timestamp and calibrate asynchronously arriving sensor data, and spatial interpolation algorithms are employed to compensate for monitoring blind spots caused by sparse sensor deployment. The spatiotemporally aligned data packets are encrypted and submitted to a local blockchain node, generating a hash value containing timestamps and device identifiers for storage. The fused standard dataset is transmitted to a cloud-based analysis platform via a dedicated power grid (5G private network). A pre-trained deep learning model identifies early degradation characteristics of the battery pack and outputs a status assessment report including health indices and risk levels.

[0019] Compared to related technologies, traditional methods employ independent data acquisition channels and a centralized post-processing model, making real-time correlation analysis of multiple parameters impossible. This solution effectively addresses the spatiotemporal inconsistency issue of sensor data by implementing fusion processing at the data acquisition end. While existing blockchain applications largely focus on transaction record storage, this solution innovatively introduces it into device-level data management, constructing a trusted assurance mechanism covering the entire data lifecycle.

[0020] Through the above technical solutions, this application realizes real-time collaborative processing of multi-source heterogeneous data, reducing the risk of misjudgment caused by data asynchrony; establishes a verifiable data integrity guarantee mechanism to prevent critical operation and maintenance data from being maliciously tampered with; and through the collaboration of edge computing and cloud intelligence, forms a complete analysis link from data collection to status assessment, providing a reliable basis for dynamically adjusting operation and maintenance strategies.

[0021] In some embodiments, this disclosure further proposes operation and maintenance methods for energy storage power stations, such as... Figure 2 As shown, the operation and maintenance methods include: Step S201: Obtain the operating status data of the energy storage power station.

[0022] Step S202 involves preprocessing the operational status data to obtain preprocessed data. This preprocessing includes local preprocessing of the operational status data, which includes data fusion processing and data storage processing. For details, please refer to the detailed description of this technical content in step 102.

[0023] Step S2021: Spatiotemporal alignment and feature extraction of the running status data are performed using a multi-parameter fusion algorithm model, and the sampling frequency of the running status data is aligned using an algorithm combining wavelet transform and Kalman filtering to obtain the data fusion processing result; The multi-parameter fusion algorithm model refers to a computational framework used to integrate heterogeneous data from multiple sources. Specifically, it can be implemented using spatiotemporal interpolation algorithms and feature vector construction methods, eliminating data dimensionality differences by unifying the timestamps and spatial coordinates of different sensors. Feature extraction refers to screening key state indicators from the fused data, which can be achieved using principal component analysis and convolutional neural networks. Dimensionality reduction is used to extract the battery pack's temperature gradient distribution and voltage fluctuation characteristics.

[0024] The wavelet transform-Kalman filter fusion algorithm addresses the issue of inconsistent sampling frequencies for different parameters. In one example, during battery cluster monitoring in an energy storage power station, the temperature sensor samples at 100Hz, while the pressure sensor samples at 50Hz. Traditional methods of directly fusing these data lead to timestamp mismatches and information loss. By employing the wavelet transform and Kalman filter fusion algorithm, the system can effectively align and fuse these data with different sampling frequencies. For instance, pressure data can be synchronized with temperature data through interpolation or prediction, and noise can be filtered out. This precise fusion enables the fault warning model to more accurately identify early signs of battery thermal runaway, increasing the warning accuracy from 80% using traditional methods to over 99.99%, thereby significantly reducing false alarms and missed alarms.

[0025] Step S2022: Perform data storage processing on the preset data in the data fusion processing result based on the blockchain evidence storage model.

[0026] Among them, the blockchain evidence storage model refers to a distributed data storage and verification mechanism, which can be implemented using a consortium blockchain architecture and a hash value on-chain method. A consensus mechanism ensures the immutability of critical operational data. Spatiotemporal alignment refers to the standardization of the time series and spatial location of multi-source data, which can be achieved using a sliding window algorithm and a coordinate transformation matrix to solve the data misalignment problem caused by differences in sensor sampling frequencies.

[0027] Step S203 involves performing data analysis on the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station. See step S103 for a detailed description of this technical content.

[0028] Specifically, during the operation of an energy storage power station, a temperature sensor array collects individual battery cell temperature data once per second, while a pressure sensor array outputs expansion and deformation data once per minute. The multi-parameter fusion algorithm first interpolates the timestamps of both types of data to align all data points to a unified time baseline. Then, it converts the spatial coordinate system of the expansion and deformation data to a layout coordinate system consistent with that of the temperature sensors. After completing the spatiotemporal alignment, a sliding window mechanism is used to extract the temperature change rate and deformation increment of each battery cell within a fixed time interval, constructing a data matrix containing spatiotemporal correlation features.

[0029] After data fusion is completed, the blockchain-based evidence storage model uses hash operations to generate digital fingerprints from key feature data related to device security, such as sudden increases in temperature and excessive deformation. This fingerprint is then written to the consortium blockchain node via a smart contract, with access rules set to allow only authorized devices to retrieve the original data using digital certificates.

[0030] Compared to related technologies, traditional methods use independent databases to store data from different sensors, leading to multiple cross-database queries and the risk of data version conflicts during data analysis. This solution establishes a unified spatiotemporal benchmark framework, enabling real-time matching and correlation of multi-dimensional data such as temperature and pressure. For example, it can synchronously analyze a sudden temperature rise event of a battery cell with its casing expansion data. In related technologies, operational data is typically stored on centralized servers, posing single points of failure and the risk of data tampering. This solution, however, employs distributed ledger technology, ensuring that key evidence chains are synchronously backed up on multiple nodes, and modifications require verification from a majority of nodes.

[0031] Through the above technical solutions, this disclosure achieves spatiotemporal consistency integration of multi-source sensing data, effectively eliminating feature extraction errors caused by differences in sensor sampling; at the same time, it establishes a blockchain-based verifiable data storage system to ensure the integrity and traceability of key operation and maintenance data, providing a reliable data foundation for subsequent fault diagnosis.

[0032] In some embodiments, this disclosure further proposes to store operation and maintenance data on the blockchain using hash values ​​and to enable authorized access to the data through smart contracts.

[0033] Operation and maintenance data refers to key business data generated during the operation and maintenance of energy storage power stations. Specifically, it can be achieved through battery charging and discharging records, equipment maintenance logs, and abnormal event reports. This data directly affects the accuracy of equipment status assessment and fault diagnosis.

[0034] Hash value on-chain storage refers to the operation of writing data fingerprints into the blockchain. Specifically, the SHA-256 algorithm can be used to generate data hash values ​​and store them in the blockchain distributed ledger. The blockchain's chain storage structure and consensus mechanism ensure that the data is tamper-proof.

[0035] Smart contracts are computer programs that automatically execute access rules. Specifically, they can be implemented using contract code deployed on the Ethereum Virtual Machine, which triggers the granting or revocation of access permissions based on preset conditions.

[0036] Data authorization access refers to the operation of controlling the permissions of data users. Specifically, it can be achieved by using an identity verification mechanism based on digital certificates, combined with the access policies defined in smart contracts to realize dynamic permission management.

[0037] Specifically, after processing the operational data using a multi-parameter fusion algorithm, only key data fields are selected to generate hash values. These hash values ​​are broadcast to the network via blockchain nodes and, after consensus verification, are written into a new block. The original data is stored locally, with the hash values ​​mapped to the storage paths. When an external system requests access to the data, the smart contract automatically verifies the requester's digital certificate and access purpose, generating a temporary access token after matching preset rules. During access, the system compares the data hash values ​​with on-chain records in real time; if inconsistencies are found, an alarm is triggered and access is terminated.

[0038] Compared to related technologies, existing operation and maintenance platforms typically use centralized databases to store raw data, which poses a single point of tampering risk and lacks an effective traceability mechanism. This solution, however, separates the storage of hash values ​​from the raw data, ensuring data integrity while avoiding the storage pressure on the blockchain. Traditional access control relies on manual approval, and response delays lead to inefficient data sharing. This solution, however, achieves automated permission management through smart contracts, striking a balance between data security and availability.

[0039] Through the above technical solutions, this disclosure addresses the problems of easily tampered data and inefficient access control in energy storage power station operation and maintenance. On-chain hash value storage ensures data authenticity and verifiability, while smart contracts enable fine-grained access control to prevent unauthorized operations. The selective on-chain strategy for key data reduces the blockchain storage load while meeting data traceability requirements. The dynamic permission management mechanism adapts to multi-role collaborative operation and maintenance scenarios, improving data sharing efficiency and reducing the risk of human intervention.

[0040] This disclosure further proposes a method for obtaining operational status data of an energy storage power station, including obtaining operation and maintenance data of the energy storage power station, obtaining battery cell temperature data through a temperature sensor array, obtaining battery cell expansion and deformation data through a pressure sensor array, obtaining battery pack charging and discharging current data and voltage data through current sensors and voltage sensors respectively, obtaining indoor and outdoor environmental data of the energy storage power station through temperature and humidity sensors, and obtaining stress and vibration data of the load-bearing structure of the energy storage power station through strain sensors and vibration sensors.

[0041] Among them, the temperature sensor array refers to multiple temperature detection units arranged on the surface or inside of the battery cell. Specifically, it can be implemented using a distributed fiber optic temperature measurement system or a patch thermocouple array, and is used to capture the differences in heat distribution in different areas during the charging and discharging process of the battery.

[0042] Among them, the pressure sensor array refers to the pressure sensing module installed on the battery casing or fixed bracket. Specifically, it can be implemented using piezoelectric thin film sensors or microelectromechanical system pressure sensors to monitor the surface pressure changes caused by battery expansion and deformation.

[0043] Among them, strain sensors refer to strain measurement devices attached to the surface of load-bearing structures. Specifically, they can be implemented using resistance strain gauges or fiber optic grating sensors, and are used to detect micro-strains generated by structural deformation in real time.

[0044] Among them, the vibration sensor refers to the acceleration detection device installed on the energy storage cabinet or bracket. Specifically, it can be implemented using a piezoelectric vibration sensor to collect the mechanical vibration spectrum during equipment operation.

[0045] Specifically, in individual battery cell monitoring, temperature sensor arrays are distributed in a grid pattern on the battery surface. Each sensor node synchronously collects temperature data at a preset sampling frequency, forming a temperature field distribution map. Combined with expansion and deformation data recorded by pressure sensor arrays, abnormal heat accumulation and mechanical deformation caused by internal short circuits or electrolyte leaks can be identified. For battery pack operation status monitoring, current and voltage sensors record the dynamic characteristics of the charging and discharging process in a synchronous acquisition mode. The health status of the battery pack is determined by comparing the slope changes of the charging and discharging curves. Temperature and humidity sensors are deployed at different heights within the energy storage compartment and at external environmental monitoring points to establish a temperature and humidity gradient distribution model and assess the impact of environmental factors on battery performance. In terms of structural safety monitoring, strain sensors are deployed along the key stress points of the load-bearing beams, and vibration sensors are installed at the four corners of the energy storage cabinet. By analyzing the correlation between strain data and vibration spectrum, the degree of structural fatigue damage is determined.

[0046] Compared to related technologies, traditional methods rely on a single type of sensor for independent monitoring, such as using a single temperature sensor to monitor the temperature at a specific point on the battery surface. This fails to capture the localized temperature rise phenomena in the early stages of thermal runaway within the battery. In contrast, this solution utilizes a collaborative deployment of a multi-sensor array to achieve distributed measurement of the battery temperature field in the spatial dimension and synchronous acquisition of electro-thermal-mechanical multi-physics data in the temporal dimension, thus overcoming the problem of insufficient representativeness of single-point monitoring data. While related technologies typically focus only on the average temperature and humidity inside the chamber for environmental monitoring, this solution establishes a three-dimensional environmental parameter model through multi-point deployment, enabling precise assessment of the impact of temperature stratification on battery heat dissipation.

[0047] Through the above technical solutions, this disclosure enables multi-dimensional perception of the equipment status of energy storage power stations. By simultaneously collecting heterogeneous data from multiple sources, including electrochemical parameters, mechanical deformation, environmental conditions, and structural safety, it provides complete input information for subsequent data fusion and analysis. For battery thermal runaway early warning scenarios, combining temperature field distribution characteristics with expansion pressure change trends can identify internal short-circuit faults in advance. In structural safety assessments, joint analysis of vibration spectrum and strain data can accurately determine potential hazards such as loose support bolts or welding cracks. The establishment of this multi-parameter perception system effectively improves the comprehensiveness of equipment status monitoring and the timeliness of anomaly detection.

[0048] In some embodiments, this disclosure also provides an operation and maintenance method for an energy storage power station, such as... Figure 3 As shown, the method includes: Step S301: Obtain the operating status data of the energy storage power station.

[0049] Step S302 involves preprocessing the operational status data to obtain preprocessed data. This preprocessing includes local preprocessing of the operational status data, which includes data fusion processing and data storage processing. For details, please refer to the detailed description of this technical content in step 202.

[0050] Step S3021 involves performing spatiotemporal alignment and feature extraction on the running status data using a multi-parameter fusion algorithm model, and aligning the sampling frequencies of the running status data using an algorithm combining wavelet transform and Kalman filtering to obtain the data fusion processing result. For details, please refer to the detailed description of this technique in step 2021.

[0051] Step S3022 involves performing data storage processing on the preset data in the data fusion processing result based on the blockchain evidence storage model. For details, please refer to the detailed description of this technical content in step 2022.

[0052] Step S303 involves performing data analysis on the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station. See step S203 for a detailed description of this technical content.

[0053] Step S304: Classify the preprocessed data according to the service quality level standard to obtain classified data.

[0054] Among them, the service quality level standard refers to the priority system divided according to the timeliness and completeness requirements of data. Specifically, it can be implemented by matching data tags using a preset rule engine. For example, battery thermal runaway warning signals are classified as the highest level, while environmental temperature and humidity monitoring data are classified as the ordinary level.

[0055] Step S305: Based on the dedicated power grid, the classified data is transmitted to the preprocessing model for data analysis according to the service quality level standard.

[0056] Specifically, after data preprocessing, the metadata carried in the data packets is parsed by a rule engine to match preset priority tags. For example, when abnormal fluctuations in battery pack voltage are detected, it is automatically marked as an emergency and a transmission channel switching command is triggered. The power grid dynamically allocates transmission resources according to data priority. Emergency-level data is transmitted through low-latency slice channels, and its transmission path adopts a multi-node parallel transmission mode, while ordinary-level data is transmitted through standard slice channels, allowing batch transmission during network idle periods. This mechanism enables high-priority data to overcome the queuing limitations of traditional network transmission, such as prioritizing the transmission bandwidth of safety alarm signals when data congestion occurs.

[0057] Compared to related technologies, traditional operation and maintenance systems using a unified transmission strategy share transmission channels with security alarm data and regular monitoring data, which can easily lead to delays in critical information due to network congestion. This solution, however, establishes a tiered transmission mechanism, allowing high-priority data to receive independent transmission resources. For example, under the same network load, the end-to-end transmission latency of security alarm data can be reduced to one-fifth of the traditional method, while also avoiding resource waste caused by low-priority data excessively consuming bandwidth.

[0058] Through the above technical solution, this disclosure achieves differentiated transmission control for multiple types of data, effectively solving the channel contention problem during concurrent transmission of massive monitoring data. By dynamically allocating network slice resources, it ensures deterministic transmission of critical data such as battery thermal runaway early warning and voltage anomalies, while optimizing the transmission efficiency of ordinary monitoring data, providing stable and reliable data input for subsequent data analysis modules.

[0059] In some embodiments, this disclosure also provides an operation and maintenance method for an energy storage power station, such as... Figure 4 As shown, the method includes: Step S401: Obtain the operating status data of the energy storage power station.

[0060] Step S402 involves preprocessing the operational status data to obtain preprocessed data. This preprocessing includes local preprocessing of the operational status data, which includes data fusion processing and data storage processing. For details, please refer to the detailed description of this technical content in step 302.

[0061] Step S4021 involves performing spatiotemporal alignment and feature extraction on the running status data using a multi-parameter fusion algorithm model, and aligning the sampling frequencies of the running status data using an algorithm combining wavelet transform and Kalman filtering to obtain the data fusion processing result. For details, please refer to the detailed description of this technique in step 3021.

[0062] Step S4022 involves performing data storage processing on the preset data in the data fusion processing result based on the blockchain evidence storage model. For details, please refer to the detailed description of this technical content in step 3022.

[0063] Step S403 involves performing data analysis on the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station. See step S303 for a detailed description of this technical content.

[0064] Step S4031: Process the data fusion results based on the deep learning model to obtain fault warning data.

[0065] Deep learning models refer to computational models that use multi-layer neural network structures to perform pattern recognition on time-series data. Specifically, they can be implemented using long short-term memory networks or convolutional neural networks, and are used to learn the mapping relationship between abnormal features of equipment and failure modes from historical operating data.

[0066] Step S4032: Process the data fusion results based on the digital twin model to obtain the equipment status data of the energy storage power station.

[0067] Among them, the digital twin model refers to the simulation system that establishes real-time interaction between physical devices and virtual models. Specifically, it can be implemented using multi-physics coupling modeling technology to simulate the combined effects of electrochemical, thermodynamic and mechanical stresses during battery charging and discharging.

[0068] Specifically, the multi-dimensional operational data after data fusion processing is simultaneously input into the deep learning model and the digital twin model. The deep learning model analyzes the dynamic trends of temperature, voltage, and deformation parameters to identify abnormal combinations of features that deviate from normal operating modes, such as triggering an early warning when the battery pack temperature gradient is abnormal but the voltage has not yet dropped. The digital twin model constructs a three-dimensional model of each battery cell, combining real-time collected expansion and deformation data with the number of charge-discharge cycles to simulate the aging process of electrode materials and calculate the battery capacity decay curve. The output data of the two models are cross-validated during the operational status assessment phase. When the deep learning model detects abnormal charge-discharge behavior, the digital twin model can simultaneously retrieve the corresponding device's lifespan prediction data to help determine whether the fault type is caused by irreversible aging.

[0069] Compared to related technologies, traditional methods typically employ a single analytical model to process operational data, such as using only statistical models for threshold alarms, failing to capture potential faults under the coupling of multiple parameters. This solution constructs a dual-model collaborative analysis architecture, achieving early fault detection in the time-series dimension while simultaneously quantifying equipment health status in the spatial dimension. While digital twin applications in related technologies are mostly limited to equipment visualization and fail to form a data loop with deep learning-based early warning mechanisms, this solution establishes a data interaction channel between models, enabling virtual simulation results to inversely optimize the judgment logic of the early warning algorithm.

[0070] Through the above technical solution, this disclosure solves the problem of delayed early warning caused by single-dimensional data analysis in traditional operation and maintenance, and realizes dynamic monitoring of the equipment's status throughout its entire life cycle. The anomaly detection module captures multi-parameter correlation features through a deep learning model to identify potential risks such as battery thermal runaway in advance; the life assessment module uses a digital twin model to simulate the material aging process and accurately predict the battery replacement cycle. The collaborative operation of the two types of models avoids redundant maintenance caused by misjudgment from a single sensor and prevents the missed detection of hidden defects in the equipment, ultimately forming a complete analysis system covering fault early warning and status assessment.

[0071] In some embodiments, this disclosure further proposes to perform data analysis on preprocessed data based on a preprocessing model to obtain the operational status assessment results of the energy storage power station, and then further includes: configuring operation and maintenance resources through a genetic algorithm based on multi-objective global constraints to obtain operation and maintenance resource configuration results; and generating operation and maintenance strategies based on the operation and maintenance resource configuration results using a reinforcement learning algorithm, wherein the reinforcement learning algorithm takes operational status data as input parameters and economic benefits, battery life, and safety as constraints.

[0072] The operation and maintenance resource configuration is carried out through a genetic algorithm based on multi-objective global constraints, which specifically includes: Step 1: Initialize the genetic algorithm population; Step 2: Calculate the multi-objective fitness function; Step 3: Improve the genetic algorithm operator operation; Step 4: Convergence judgment.

[0073] In one example, initializing the genetic algorithm population generates multiple random resource allocation schemes. A multi-objective fitness function calculation evaluates each scheme's performance across various objectives, including economic efficiency, equipment lifespan, and safety. Improving genetic algorithm operators (such as crossover and mutation) generates new, potentially better schemes based on the best-performing ones. This process iterates until convergence criteria are met, such as when performance improvements across multiple generations become insignificant, ultimately resulting in a globally optimal resource allocation scheme.

[0074] It is understandable that the genetic algorithm based on multi-objective global constraints is not a simple application of genetic algorithms, but rather an optimized and customized approach that can more effectively handle complex resource allocation problems under multi-objective constraints, ensuring the scientific and efficient allocation of operation and maintenance resources, thereby solving the problems of wasted or insufficient operation and maintenance resources.

[0075] The reinforcement learning algorithm includes: Step 1: Definition of the reinforcement learning MDP (Markov Decision Process) framework; Step 2: Calculation of the reward function; Step 3: Training of the DQN (Deep Q Network) algorithm.

[0076] In one example, when the health of a battery module begins to decline, the reinforcement learning module intervenes. First, the MDP framework defines the battery module's state (e.g., temperature, voltage, internal resistance), possible actions (e.g., adjusting charging / discharging current, turning on the cooling fan), and environmental feedback. Next, the reward function calculation assigns positive or negative rewards based on the actions' impact on battery life, energy consumption, and safety. Finally, the DQN algorithm is trained through simulation or real-world operation, allowing the DQN agent to continuously try different actions and learn from the reward signals. Ultimately, it learns how to dynamically adjust charging / discharging thresholds or cooling strategies under different health states to maximize battery life and minimize maintenance costs.

[0077] It is understood that this application uses the DQN (Deep Q-Network) reinforcement learning algorithm, which can make optimal decisions in complex and dynamic environments through interactive learning with the environment, dynamically adjust maintenance thresholds, enhance the system's ability to adaptively adjust operation and maintenance strategies in real time, and further optimize operation and maintenance costs and equipment lifespan.

[0078] In some embodiments, this disclosure further proposes a technical solution for generating inspection strategies based on particle swarm optimization. This solution integrates the operational status assessment results of the energy storage power station with spatial layout parameters, and combines signal coverage and power information of the inspection equipment to form a dynamically optimized inspection path planning mechanism.

[0079] Among them, the particle swarm optimization algorithm is an optimization calculation method based on swarm intelligence. Specifically, it can be implemented by initializing the position and velocity of the particle swarm, defining the fitness function, and iteratively updating the particle position. Its fitness function can be set as a weighted combination of parameters such as path length, device state priority, and signal strength, and is used to find the optimal inspection path under complex constraints.

[0080] Among them, signal coverage information refers to the signal strength distribution data of the wireless communication network within the energy storage power station. Specifically, it can be obtained using a base station signal strength detector or network coverage heat map generation technology, and is used to predict communication blind spots and optimize data transmission reliability during the path planning stage.

[0081] Among them, the battery power information of the inspection equipment refers to the remaining battery power data of mobile inspection robots or drones, which can be collected in real time through the battery management system to dynamically adjust the inspection task allocation strategy to avoid power outages in the middle of the process.

[0082] Specifically, after obtaining the operational status assessment results of the energy storage power station, the equipment health grading results are mapped to inspection priority weights, and a three-dimensional spatial model is constructed by combining the physical distribution coordinates of the power station equipment. The particle swarm optimization algorithm takes minimizing the total length of the inspection path as the objective function, while using the access order of high-priority equipment, signal strength threshold, and equipment range as constraints for multi-objective optimization. During each iteration, when the particle position is updated, it is simultaneously checked whether the path node is in a communication blind zone or exceeds the current power range of the equipment, and path schemes that do not meet the conditions are dynamically eliminated. The final output inspection strategy includes the equipment access sequence sorted by priority and the corresponding optimal movement trajectory, and also generates task allocation schemes and charging scheduling suggestions for multiple inspection devices.

[0083] Compared to related technologies, traditional inspection path planning methods typically employ fixed routes or simple polling mechanisms, failing to adjust inspection focus areas based on real-time equipment status and prone to data transmission interruptions when communication conditions change. This solution, however, quantifies status assessment results as algorithm input parameters, enabling the inspection path to adapt to changes in equipment health. Furthermore, it embeds communication quality and equipment battery power as hard constraints into the optimization process, ensuring the continuity of inspection operations and the reliability of data feedback.

[0084] Through the above technical solutions, this disclosure realizes dynamic optimization of inspection strategies and multi-dimensional resource collaboration, effectively solving the problem of low inspection efficiency caused by rigid path planning in traditional methods, avoiding response delays in monitoring the status of important equipment, and significantly reducing the risk of interruption during inspection by predicting communication blind spots and equipment battery life limitations, thereby improving the overall reliability and resource utilization of the energy storage power station operation and maintenance system.

[0085] In some embodiments, this disclosure further proposes sending the inspection feedback information of the inspection equipment to a preprocessing model to process the inspection strategy and obtain optimized processing results.

[0086] The inspection feedback information refers to data collected during the inspection process, including changes in equipment status, signal coverage fluctuations, and power consumption. This can be achieved using real-time sensor monitoring and wireless transmission modules, reflecting the dynamic changes in the actual inspection environment and equipment operating status. The preprocessing model is a computational model with data fusion and analysis capabilities. It can be implemented using a combination of machine learning and optimization algorithms to collaboratively process multi-dimensional data and generate optimization instructions. The optimization result refers to the adjusted inspection path planning parameters, which can be implemented using a dynamic weight adjustment algorithm to improve the adaptability of subsequent inspection strategies.

[0087] Specifically, during the execution of a predetermined inspection strategy by the inspection equipment, real-time data on abnormal battery temperatures, excessive structural vibrations, or network signal attenuation are transmitted to a preprocessing model via a wireless communication module. This model performs spatiotemporal correlation analysis on the feedback data and the original operational status assessment results, identifying fault areas or path obstacles requiring priority handling. By dynamically adjusting the inertia weight parameters in the particle swarm optimization algorithm, the coordinate sequence and priority ranking of the optimal inspection path are recalculated. The optimized path planning parameters are synchronously updated to the inspection equipment control system, forming a closed-loop optimization process from data acquisition and analysis to execution. For example, when an abnormal temperature gradient occurs in a battery cluster in a certain area, the model will automatically increase the inspection frequency for that area and dynamically shorten the single inspection radius based on the remaining battery power.

[0088] Compared to related technologies, traditional operation and maintenance strategies rely on fixed inspection cycles and path templates, making them unable to respond to sudden equipment failures or sudden changes in environmental parameters. Existing methods typically depend on manual experience to adjust strategies, resulting in response lags and subjective biases. This solution establishes an automated feedback optimization mechanism, enabling the inspection strategy to be dynamically corrected based on real-time operational data, eliminating deviations between preset paths and actual conditions.

[0089] Through the above technical solution, this disclosure achieves online adaptive adjustment of the inspection strategy, effectively solving the problem of path planning failure caused by environmental changes in traditional methods. In scenarios of localized overheating of the battery pack, it can quickly adjust the inspection route to prioritize the investigation of faulty areas; in areas with unstable communication signals, it can automatically optimize the equipment inspection trajectory to avoid data loss. This solution also balances equipment power consumption and inspection coverage, avoiding inspection interruptions due to insufficient power.

[0090] In one example, in remote and harsh environments, the performance degradation of inspection equipment (such as inspection robots) can lead to insufficient execution of operation and maintenance strategies. Based on this, this application adds a "health and task feasibility assessment unit for inspection equipment" to the existing operation and maintenance platform to ensure the dynamic adaptability and reliable execution of operation and maintenance strategies.

[0091] This optimization unit will serve as an extension of the operations and maintenance platform's intelligent analysis engine, working closely with existing fault warning, health assessment, and operations and maintenance strategy optimization units. Specifically, it includes: The inspection equipment collects key operating parameters in real time, including but not limited to battery health status, motor temperature, sensor operating status, wear and tear of mechanical parts, and environmental parameters of the robot's environment (such as local temperature, humidity, and wind speed). This data will be aggregated to a data platform and stored via a 5G transmission layer.

[0092] The inspection equipment health and task feasibility assessment unit will utilize the aforementioned real-time data, combined with historical operational data and environmental impact models, to construct health assessment models for different types of operation and maintenance units. This model employs machine learning algorithms (such as support vector machines or random forests) to learn the performance degradation patterns and failure modes of various components within the operation and maintenance unit under different environmental conditions (such as low temperature, high humidity, and sandstorms). For example, it predicts that in the current environment of -10 degrees Celsius, the battery life of a certain model of robot will decrease by 30%, or the recognition accuracy of its vision sensor will decrease by 20% in strong winds and sandstorms.

[0093] Based on the real-time health assessment results of the operation and maintenance units, and the specific requirements of the operation and maintenance tasks to be performed (such as task duration, required accuracy, environmental conditions of the target area, path complexity, etc.), the "feasibility score" and "expected success rate" of each operation and maintenance unit in performing a specific task will be dynamically evaluated. For example, a task requiring high-precision visual recognition, in the current environment of heavy sandstorms, will have its feasibility score significantly reduced if assigned to a robot whose visual sensors are affected.

[0094] The aforementioned "task feasibility score" and "expected success rate" are used as constraints or input parameters and fed back to the operation and maintenance strategy optimization unit in real time. This application's two-layer optimization algorithm (upper layer) improves the genetic algorithm and (lower layer) the reinforcement learning algorithm. When generating operation and maintenance strategies, it no longer only considers the number of operation and maintenance units and spare parts inventory, but also comprehensively considers the real-time health status and environmental adaptability of each available operation and maintenance unit.

[0095] When a high-priority task is assigned, the system will prioritize the operation and maintenance unit with the highest "feasibility score".

[0096] If the feasibility score of all available operation and maintenance units for executing a critical task is lower than the preset threshold, the system will trigger an alert and suggest adjusting the task priority, waiting for environmental conditions to improve, or, in extreme cases, manual intervention.

[0097] The system can also dynamically adjust task allocation. For example, tasks that require long-term exposure to harsh environments can be split and performed by multiple healthy robots in turn to reduce the risk of a single robot malfunctioning.

[0098] Through the above technical solutions, the operation and maintenance platform can achieve "self-awareness" and "self-optimization" of the operation and maintenance execution units, ensuring that the generated operation and maintenance strategies are not only theoretically optimal, but also highly reliable and feasible in actual execution. This effectively bridges the gap between operation and maintenance strategies and actual execution capabilities in remote and harsh environments, further improving the overall operation and maintenance efficiency and safety of energy storage power stations.

[0099] This disclosure further proposes an operation and maintenance device for an energy storage power station, such as... Figure 5 As shown, it includes a data acquisition module 501, a data preprocessing module 502, and an acquisition module 503.

[0100] The data acquisition module 501 is used to acquire the operating status data of the energy storage power station.

[0101] The data preprocessing module 502 is used to preprocess the running status data to obtain preprocessed data. The preprocessing includes locally executed data fusion processing and data storage processing.

[0102] Data fusion processing refers to the spatiotemporal alignment and feature extraction of multi-source heterogeneous data using multi-parameter fusion algorithms. Specifically, this can be achieved by combining Kalman filtering with timestamp synchronization technology to eliminate temporal offsets and spatial misalignments in sensor data. Data notarization processing refers to the blockchain notarization of critical operational data. Specifically, this can be achieved by using the SHA-256 hash algorithm to generate data fingerprints and writing them to consortium blockchain nodes. Smart contracts control data access permissions to ensure data integrity and traceability.

[0103] The acquisition module 503 is used to analyze the preprocessed data based on the preprocessing model to obtain the running status evaluation results.

[0104] The preprocessing model includes a deep learning model and a digital twin model. Specifically, a fault prediction model can be constructed using a convolutional neural network, and a digital twin of the battery component can be constructed using 3D modeling technology to achieve a virtual-real mapping of the device status.

[0105] Specifically, the data acquisition module integrates temperature sensor arrays, pressure sensor arrays, and current and voltage sensor groups to collect multi-dimensional data such as battery cell temperature, expansion deformation, and charge / discharge parameters. The data preprocessing module performs data cleaning and format conversion on a local edge computing node, eliminating sampling frequency differences between different sensors through spatiotemporal alignment and extracting key feature vectors using principal component analysis. Simultaneously, the battery health status data is generated into hash values ​​and uploaded to a blockchain network for notarization through a distributed node consensus mechanism. The acquisition module inputs the fused multi-dimensional data into a trained LSTM neural network model to predict the remaining lifespan of the battery pack and simulates equipment performance degradation curves under different operating conditions based on a digital twin model, generating an evaluation report that includes failure probability and equipment health index.

[0106] In some specific implementations, the temperature sensor array can be a distributed DS18B20 digital temperature sensor, the pressure sensor array can be a MEMS piezoresistive sensor, and the current and voltage sensor group can be configured with Hall effect sensors. Blockchain evidence storage processing can utilize the Hyperledger Fabric framework to build a private blockchain and set data access control policies. The digital twin model can be built based on the ANSYS Twin Builder platform, receiving sensor data in real time to update the 3D simulation model.

[0107] Compared to related technologies, traditional operation and maintenance devices typically employ independent data acquisition modules and offline analysis systems, failing to achieve spatiotemporal alignment of multi-source data, resulting in data fusion errors exceeding permissible limits. Existing equipment lacks local preprocessing capabilities; directly uploading raw data to the cloud increases network latency, and a trusted evidence storage mechanism is not established. This solution reduces data processing time by approximately 40% by completing data fusion and evidence storage at the edge, while blockchain evidence storage shortens data tamper detection response time to the minute level.

[0108] Through the above technical solutions, this disclosure achieves accurate alignment and feature extraction of multi-parameter sensing data, solving the problem of poor data coordination in traditional single-point monitoring. The blockchain evidence storage mechanism effectively prevents critical operation and maintenance data from being maliciously tampered with, improving data credibility verification efficiency by more than 50%. The multi-dimensional analysis capabilities of the preprocessing model enable equipment status assessment accuracy to reach over 98%, providing a reliable basis for dynamically adjusting inspection paths and preventative maintenance.

[0109] The operation and maintenance device for the energy storage power station provided in this disclosure can execute the operation and maintenance method for the energy storage power station provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0110] Figure 6This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure.

[0111] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural schematic for implementing an electronic device according to embodiments of the present disclosure. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. The RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0112] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0113] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the operation and maintenance method of the energy storage power station according to embodiments of this disclosure.

[0114] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0115] This disclosure also provides a computer-readable storage medium in which the methods described in this disclosure can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the operation and maintenance method of the energy storage power station shown in the above embodiments is implemented.

[0116] A portion of this disclosure can be applied to computer program products, such as computer program instructions, which, when executed by a computer, can invoke or provide methods and / or technical solutions according to this disclosure through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, and installation package files. Accordingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions; the computer compiling the instructions and then executing the corresponding compiled program; the computer reading and executing the instructions; or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0117] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for the operation and maintenance of an energy storage power station, characterized in that, The method includes: Acquire operational status data of energy storage power stations; The operation status data is preprocessed to obtain preprocessed data, wherein the preprocessing includes local preprocessing of the operation status data, and the local preprocessing includes data fusion processing and data storage processing of the operation status data; The preprocessed data is analyzed based on the preprocessing model to obtain the operational status assessment results of the energy storage power station.

2. The method according to claim 1, characterized in that, The data fusion processing and data storage processing of the operational status data include: The running status data is spatiotemporally aligned and features are extracted using a multi-parameter fusion algorithm model, and the sampling frequency of the running status data is aligned using an algorithm combining wavelet transform and Kalman filtering to obtain the data fusion processing result. Based on the blockchain-based evidence storage model, the preset data in the data fusion processing results are processed for data evidence storage.

3. The method according to claim 2, characterized in that, The preset data includes operation and maintenance data. The data storage processing of preset data in the data fusion processing result based on the blockchain-based evidence storage model includes: The operation and maintenance data is hashed and stored on the blockchain for notarization, and data access authorization is implemented through smart contracts.

4. The method according to claim 1, characterized in that, The acquisition of operational status data of the energy storage power station includes: Obtain the operation and maintenance data of the energy storage power station; and / or Acquire battery cell temperature data via a temperature sensor array; and / or The expansion and deformation data of the battery cells are acquired through a pressure sensor array; and / or The charging and discharging current data and voltage data of the battery pack are acquired through current sensors and voltage sensors, respectively; and / or The indoor and outdoor environmental data of the energy storage power station are acquired through temperature and humidity sensors; and / or The stress and vibration data of the load-bearing structure of the energy storage power station are obtained through strain sensors and vibration sensors.

5. The method according to claim 1, characterized in that, The process of preprocessing the operational status data to obtain preprocessed data further includes: The preprocessed data is classified according to the service quality level standard to obtain classified data; Based on the dedicated power grid, the categorized data is transmitted to the preprocessing model for data analysis according to the service quality level standards.

6. The method according to claim 2, characterized in that, The preprocessing models include deep learning models and digital twin models. The step of performing data analysis on the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station includes: Based on a deep learning model, the data fusion processing results are processed to obtain fault early warning data; wherein, the deep learning model is trained based on historical data of the data fusion processing results; Based on the digital twin model, the data fusion processing results are processed to obtain the equipment status data of the energy storage power station, wherein the equipment status data includes equipment lifespan data.

7. The method according to any one of claims 1-6, characterized in that, The process of analyzing the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station further includes: Operation and maintenance resource allocation is performed using a genetic algorithm based on multi-objective global constraints, and the operation and maintenance resource allocation results are obtained. Based on the reinforcement learning algorithm, an operation and maintenance strategy is generated according to the operation and maintenance resource configuration results. The reinforcement learning algorithm takes the operation status data as input parameters and economic benefits, battery life, and safety as constraints.

8. The method according to any one of claims 1-6, characterized in that, The process of analyzing the preprocessed data based on the preprocessing model to obtain the operational status assessment results of the energy storage power station further includes: Based on the particle swarm optimization algorithm, an inspection strategy is obtained according to the operational status evaluation results and the spatial layout of the energy storage power station, wherein the inspection strategy includes an inspection path. The energy storage power station is inspected based on the inspection path, the signal coverage information of the energy storage power station, and the power information of the inspection equipment.

9. The method according to claim 8, characterized in that, The method further includes: The inspection feedback information of the inspection equipment is sent to the preprocessing model to process the inspection strategy and obtain the optimization result. The inspection feedback information is inspection environment information.

10. An operation and maintenance device for an energy storage power station, characterized in that, The device includes: The data acquisition module is used to acquire the operating status data of the energy storage power station; A data preprocessing module is used to preprocess the running status data to obtain preprocessed data. The preprocessing includes local preprocessing of the running status data, which includes data fusion processing and data storage processing of the running status data. The acquisition module is used to perform data analysis on the preprocessed data based on the preprocessing model to obtain the operation status assessment results of the energy storage power station.

Images