A Communication and Optimization Method and System for an Energy Storage System

Abstract

The present invention discloses a communication and optimization method and system for an energy storage system, relating to the field of communication technologies. The present invention obtains the real-time state parameters of a battery pack through wireless communication, and uses time series analysis and data compensation mechanisms to predict the state parameters, solving the problems of a large number of battery packs, communication delay, and packet loss, ensuring the real-time nature and accuracy of data. Then, in combination with a digital twin model, the system can optimize the control strategy based on battery aging assessment and simulation analysis, realizing dynamic regulation and refined energy management. The differential privacy technology is adopted to encrypt the transmitted data, ensuring the privacy protection of the control strategy. Through real-time monitoring and an adaptive adjustment mechanism, the difference between the digital twin model and the actual system is minimized, further enhancing the stability, efficiency, and security of the energy storage system, and guaranteeing the long-term reliable operation of the system.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and specifically to a communication and optimization method and system for an energy storage system. Background Art

[0002] During the communication process of an energy storage system, it is necessary to monitor the state parameters of the battery pack in real time, such as voltage, current, temperature, etc., and adjust the operation strategy of the system according to these parameters to ensure the safe and stable operation of the energy storage system. However, due to the large number and wide distribution of battery packs in the energy storage system, there are problems of communication delay and packet loss between each battery pack, resulting in difficulty in ensuring the real-time and accuracy of monitoring data; at the same time, the aging and faults of the battery pack will also affect the communication quality, and further affect the control decision-making of the energy storage system. Summary of the Invention

[0003] The purpose of the present invention is to provide a communication and optimization method and system for an energy storage system, which realizes intelligent scheduling, data compensation and privacy protection of the energy storage system in a complex environment, and improves the stability, efficiency and security of the system.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] The present application provides a communication and optimization method for an energy storage system, including the following steps:

[0006] Obtain the state parameters of the battery pack in the energy storage system through wireless communication, including voltage, current and temperature data, and then use the time series analysis method to extract the characteristic of the parameter change trend;

[0007] According to the extracted characteristics, judge whether there is a communication delay or packet loss phenomenon. When there is a communication anomaly, start the data compensation mechanism and predict the current state parameters using historical data;

[0008] According to the obtained state parameters and predicted data, combine the geometric, physical, behavior and rule models of digital twin to construct a digital twin model of the energy storage system;

[0009] Adopt the support vector machine algorithm, input the historical operation data of the battery pack, output the evaluation result of the aging degree, adjust the parameter settings of the digital twin model according to the evaluation result of the aging degree, and obtain the optimized control strategy parameters through model simulation;

[0010] Transmit the optimized control strategy parameters to the actual energy storage system, and encrypt the transmitted data using differential privacy technology.

[0011] Further, obtain the state parameters of the battery pack in the energy storage system through wireless communication, including voltage, current, and temperature data, and then use time series analysis methods to extract the characteristics of parameter change trends, specifically including:

[0012] Establish a communication link with the energy storage system through a wireless communication module, periodically obtain the voltage, current, and temperature state parameters of the battery pack, and store the obtained state parameter data in chronological order to form a time series data set;

[0013] Preprocess the time series data to improve data quality, and use time series analysis algorithms to model the preprocessed data to extract the change trend characteristics of voltage, current, and temperature parameters;

[0014] Based on the extracted change trend characteristics, judge the health status and remaining life of the battery pack. When an abnormal trend is found, trigger an early warning mechanism, and then dynamically adjust the charge and discharge strategy of the energy storage system in combination with the state parameters and change trends of the battery pack.

[0015] Further, according to the extracted features, judge whether there is communication delay or data packet loss. When there is communication anomaly, start the data compensation mechanism and predict the current state parameters using historical data, specifically including:

[0016] Monitor the network communication status in real time, obtain key indicators, judge whether there is communication anomaly. When communication anomaly is detected, trigger the data compensation mechanism, and estimate the current state parameters using historical data according to the pre-established prediction model;

[0017] When predicting state parameters, use multiple machine learning algorithms, improve the prediction accuracy through ensemble learning, select suitable historical data features as inputs for different types of state parameters, and dynamically adjust the hyperparameters of the prediction model to adapt to different data distributions and change trends;

[0018] Fuse the predicted state parameters with the actually received data to obtain the state estimation value through weighted average or Bayesian estimation methods;

[0019] During the data fusion process, dynamically adjust the weights of the predicted value and the measured value according to the severity and duration of the communication anomaly, smooth data fluctuations and reduce the impact of anomalies; then perform anomaly detection on the compensated state parameters. When the detected abnormal deviation exceeds the preset threshold, trigger the alarm mechanism.

[0020] Further, according to the obtained state parameters and prediction data, combine the geometric, physical, behavioral, and rule models of digital twin to construct a digital twin model of the energy storage system, specifically including:

[0021] Obtain the real-time state parameters and prediction data of the energy storage system. According to the physical structure and spatial layout of the energy storage system, establish a three-dimensional geometric model and depict the appearance and internal structure of the system;

[0022] Use an equivalent circuit model or a thermal model to describe the physical characteristics of the energy storage system, and then model the behavior patterns and state transition laws of the energy storage system under different working conditions;

[0023] Utilize a business rule engine to embed grid dispatching rules and energy management strategies into the digital twin model for global optimization. Integrate multi-source heterogeneous data with various models to build a mapping of the energy storage system from the physical world to the digital space, forming a real-time updated digital twin model.

[0024] Furthermore, through model simulation, obtain the optimized control strategy parameters, specifically including:

[0025] Obtain the historical operation data of the battery pack, preprocess the data to remove outliers and noise data, obtain a standardized data set, and use the support vector machine algorithm to train the standardized historical operation data of the battery pack to establish a battery aging degree evaluation model. Determine the optimal model parameters through cross-validation and grid search;

[0026] Use the trained battery aging degree evaluation model to predict the operation data of the new battery pack, obtain the evaluation result of the battery pack's aging degree, judge the health status and remaining life of the battery pack, and dynamically adjust the parameter settings of the digital twin model according to the evaluation result of the battery pack's aging degree;

[0027] Use the adjusted digital twin model to conduct multi-scenario and multi-condition simulation analysis on the battery pack to obtain the performance data and degradation trend of the battery pack under different conditions;

[0028] Optimize the charge and discharge control strategy of the battery pack through the simulation results and machine learning algorithms;

[0029] Apply the optimized control strategy parameters to the actual battery pack management system to perform adaptive control and energy management on the battery pack, and conduct online monitoring and real-time optimization of the battery pack performance.

[0030] Furthermore, transmit the optimized control strategy parameters to the actual energy storage system and use differential privacy technology to encrypt the transmitted data, specifically including:

[0031] According to the optimized control strategy parameters, use differential privacy technology for data encryption processing, protect the parameter privacy by adding random noise, and transmit the encrypted control strategy parameters to the control module of the actual energy storage system through a secure communication protocol;

[0032] After the control module of the energy storage system receives the encrypted control strategy parameters, it uses the corresponding decryption algorithm to decrypt and obtain the optimized control strategy parameters;

[0033] The control module monitors and analyzes the operating state of the energy storage system in real time according to the decrypted control strategy parameters to determine whether the control strategy needs to be adjusted;

[0034] When the control strategy needs to be adjusted, the control module generates corresponding control instructions according to the optimized control strategy parameters and sends them to the actuator of the energy storage system. After receiving the control instructions, the parameters of the energy storage device are dynamically adjusted to optimize the control of the energy storage system.

[0035] Further, after encrypting the transmitted data, it also includes: monitoring the operating state of the system where the control strategy is deployed in the actual system in real time, judging the difference degree between the digital twin model and the actual system. When the difference exceeds the preset threshold, trigger the model parameter adaptive adjustment mechanism to update the digital twin model parameters.

[0036] Further, monitoring the operating state of the system where the control strategy is deployed in the actual system in real time, judging the difference degree between the digital twin model and the actual system. When the difference exceeds the preset threshold, trigger the model parameter adaptive adjustment mechanism to update the digital twin model parameters, specifically including:

[0037] Obtain the operating data of the actual system, including the execution situation of the control strategy, system state parameters, etc., as the input of real-time monitoring. Input the real-time monitoring data into the digital twin model, run the digital twin model, and obtain the model output result;

[0038] Calculate the difference degree between the output of the digital twin model and the operating data of the actual system, use indicators such as mean square error and relative error to measure the difference degree, judge whether the difference degree exceeds the preset threshold. When it exceeds the threshold, trigger the model parameter adaptive adjustment mechanism. When it does not exceed the threshold, continue real-time monitoring;

[0039] After triggering the model parameter adaptive adjustment mechanism, use optimization algorithms such as gradient descent algorithm and evolutionary algorithm to adjust the parameters of the digital twin model to minimize the difference between the model output and the operating data of the actual system;

[0040] Update the adjusted model parameters into the digital twin model to obtain the updated digital twin model and keep it synchronized with the actual system.

[0041] Further, after updating the parameters of the digital twin model, it further includes: continuously collecting the state parameters of the battery pack and the communication quality index, using the random forest algorithm to establish a fault prediction model, and dynamically adjusting the simulation accuracy and computing resource allocation of the digital twin model according to the fault prediction result; through cyclic iterative optimization, continuously synchronize the digital twin model with the actual system.

[0042] The present invention provides a communication and optimization system for an energy storage system, which is used to implement a communication and optimization method for an energy storage system, including:

[0043] A data acquisition and preprocessing module establishes a communication link with the energy storage system through a wireless communication module, periodically obtains the state parameters of the battery pack, and stores them in chronological order to form a time series data set;

[0044] A communication anomaly detection module uses a time series analysis algorithm to model the preprocessed data, extracts the change trend characteristics of voltage, current, and temperature parameters, then judges the health status and remaining life of the battery pack according to the characteristics, and at the same time monitors key indicators to judge whether there is a communication anomaly; when an anomaly is judged, trigger a data compensation mechanism, and use historical data and machine learning algorithms to predict and compensate the current state parameters;

[0045] A model construction and optimization module combines the geometric, physical, behavioral, and rule models of the digital twin, constructs a digital twin model of the energy storage system according to the obtained state parameters and prediction data, adds spatial constraint conditions by clarifying the hierarchical relationship and assembly order of the model, performs model assembly from parts to components to equipment, and fuses multi-source heterogeneous data with various models, uses a random finite set to model the simulation state and measurement data, and fuses real-time measurement data during the simulation operation through the Bayesian inference method to perform dynamic correction of the model;

[0046] An evaluation and control strategy optimization module uses the support vector machine algorithm, inputs the historical operation data of the battery pack, outputs the evaluation result of the aging degree, and dynamically adjusts the parameter settings of the digital twin model according to the evaluation result of the aging degree; then uses the adjusted digital twin model to perform simulation analysis in multiple scenarios and working conditions, obtains the performance data and degradation trend of the battery pack under different conditions, optimizes the charge and discharge control strategy of the battery pack through the simulation results and machine learning algorithms, extends the life of the battery pack, and applies the optimized control strategy parameters to the actual battery pack management system to perform online monitoring and real-time optimization of the battery pack performance;

[0047] The data encryption and transmission module performs data encryption processing using differential privacy technology according to the optimized control strategy parameters, protects parameter privacy by adding random noise, and transmits the encrypted control strategy parameters to the control module of the actual energy storage system through a secure communication protocol. After receiving the encrypted control strategy parameters, the control module decrypts them and monitors and analyzes the operating state of the energy storage system in real time according to the decrypted parameters to determine whether the control strategy needs to be adjusted. When the control strategy needs to be adjusted, the optimized control of the energy storage system is performed;

[0048] The adaptive adjustment and fault prediction module monitors the operating state of the system where the control strategy is deployed in the actual system in real time, judges the degree of difference between the digital twin model and the actual system. When the difference exceeds the preset threshold, it triggers the model parameter adaptive adjustment mechanism, uses an optimization algorithm to adjust the parameters of the digital twin model to minimize the difference between the model output and the actual system operation data, and updates the model parameters. At the same time, it continuously collects the state parameters of the battery pack and communication quality indicators, uses the random forest algorithm to establish a fault prediction model to obtain the fault prediction result, and dynamically adjusts the simulation accuracy and computing resource allocation of the digital twin model according to the fault prediction result. Through cyclic iterative optimization, the continuous synchronization between the digital twin model and the actual system is carried out.

[0049] The beneficial effects of the present invention are:

[0050] By introducing wireless communication, time series analysis, data compensation mechanism and digital twin technology, the problems of a large number of battery packs, communication delay and packet loss in the energy storage system are solved, and the timeliness and accuracy of monitoring data are improved. By monitoring the state parameters such as voltage, current and temperature of the battery pack in real time, using time series analysis to extract the change trend characteristics, identifying communication anomalies and starting the data compensation mechanism, and using historical data for state prediction, the stable operation and data accuracy of the system in an unstable communication environment are ensured;

[0051] Combined with the digital twin model and machine learning algorithms, by establishing a digital twin system including geometric, physical, behavioral and rule models, the control strategy can be adjusted according to the health state and remaining life of the battery pack to achieve more refined energy management and dynamic regulation. The degree of battery aging is evaluated by the support vector machine algorithm, and the charge and discharge strategy is optimized in combination with the simulation results, thereby improving the overall efficiency of the energy storage system and extending the battery life;

[0052] The application of differential privacy technology solves the problem of privacy protection of transmitted data in the process of optimizing control strategies. By encrypting the optimized control strategy parameters and transmitting them to the actual energy storage system through a secure communication protocol, sensitive data leakage is effectively prevented. On this basis, through real-time monitoring and model adaptive adjustment mechanisms, the digital twin model can be synchronized with the actual system, further optimizing the operating state of the energy storage system and ensuring its stability and security. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] For a better understanding and implementation, the technical solutions of the present application will be described in detail below with reference to the accompanying drawings.

[0054] Figure 1 FIG. is a schematic flowchart of a communication and optimization method for an energy storage system provided in Embodiment 1 of the present application;

[0055] Figure 2 FIG. is a schematic flowchart of extracting parameter change trend features of a communication and optimization method for an energy storage system provided in Embodiment 1 of the present application;

[0056] Figure 3 FIG. is a schematic structural diagram of a communication and optimization system for an energy storage system provided in Embodiment 2 of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] To further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, exemplary embodiments will be described in detail herein, and the examples are shown in the accompanying drawings. When the following description refers 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 the present application. On the contrary, they are merely examples of methods and systems consistent with some aspects of the present application as detailed in the appended claims.

[0058] The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0059] The following will describe in detail the specific embodiments, features, and their effects of the present invention with reference to the accompanying drawings and preferred embodiments.

[0060] Embodiment 1

[0061] Please refer to Figure 1 - Figure 2, this embodiment provides a communication and optimization method for an energy storage system, including the following steps:

[0062] S1. Obtain the state parameters of the battery pack in the energy storage system through wireless communication, including voltage, current, and temperature data, and then use the time series analysis method to extract the characteristic of the parameter change trend.

[0063] Further, obtain the state parameters of the battery pack in the energy storage system through wireless communication, including voltage, current, and temperature data, and then use the time series analysis method to extract the characteristic of the parameter change trend, which specifically includes:

[0064] S11. Establish a communication link with the energy storage system through a wireless communication module, periodically obtain the state parameters such as the voltage, current, and temperature of the battery pack, and store the obtained state parameter data in chronological order to form a time series data set.

[0065] S12. Preprocess the time series data, including removing outliers and smoothing noise, etc., to improve the data quality, and use a time series analysis algorithm, such as the ARIMA model, to model the preprocessed data and extract the change trend characteristics of parameters such as voltage, current, and temperature.

[0066] S13. According to the extracted change trend characteristics, judge the health state and remaining life of the battery pack. When an abnormal trend is found, trigger the warning mechanism, and then dynamically adjust the charge and discharge strategy of the energy storage system in combination with the state parameters and change trend of the battery pack to optimize the system efficiency and extend the battery life.

[0067] Specifically, in the energy storage system, it is crucial to select a suitable wireless communication technology because different technologies have their own characteristics in terms of transmission distance, power consumption, bandwidth, etc. For example, Zigbee is a short-distance, low-power wireless communication technology suitable for battery-powered devices, with a transmission distance between 10 and 100 meters and a rate of 20 kbps to 250 kbps. LoRa is a long-distance communication technology with working frequency bands including 433 MHz, 868 MHz, 915 MHz, etc., and the transmission distance can reach several kilometers to more than a dozen kilometers, which is suitable for long-distance data transmission and low-power devices. For the energy storage system, if data transmission needs to be carried out in a large range, LoRa may be a better choice; while if low power consumption and short-distance communication are concerned, Zigbee is more suitable.

[0068] More specifically, by periodically obtaining the state parameters of the battery pack through wireless communication technology and using time series analysis methods to extract the characteristics of parameter change trends, the real-time monitoring and early warning of the health status and remaining life of the battery pack are realized. At the same time, according to the extracted characteristics, the charge and discharge strategies of the energy storage system are dynamically adjusted to optimize the system efficiency and extend the battery life. This process not only improves the timeliness and accuracy of data, but also enhances the intelligent management level of the energy storage system, ensuring the safe and stable operation of the system.

[0069] S2. According to the extracted characteristics, judge whether there is communication delay or data packet loss. When there is communication anomaly, start the data compensation mechanism and use historical data to predict the current state parameters.

[0070] Furthermore, according to the extracted characteristics, judge whether there is communication delay or data packet loss. When there is communication anomaly, start the data compensation mechanism and use historical data to predict the current state parameters, specifically including:

[0071] By real-time monitoring the network communication status, obtain key indicators such as communication delay and data packet loss rate, judge whether there is communication anomaly. When communication anomaly is detected, trigger the data compensation mechanism and estimate the current state parameters using historical data according to the pre-established prediction model.

[0072] When predicting the state parameters, adopt a variety of machine learning algorithms, such as decision tree, support vector machine and neural network, etc., and improve the prediction accuracy through the way of ensemble learning. For different types of state parameters, select suitable historical data characteristics as input and dynamically adjust the hyperparameters of the prediction model to adapt to different data distributions and change trends.

[0073] Fuse the predicted state parameters with the actually received data to obtain a more accurate and stable state estimation value through methods such as weighted average or Bayesian estimation.

[0074] In the data fusion process, dynamically adjust the weights of the predicted value and the measured value according to the severity and duration of the communication anomaly to smooth data fluctuations and reduce the impact of anomalies; then perform anomaly detection on the compensated state parameters. When it is found that the anomaly deviation exceeds the preset threshold, trigger the alarm mechanism to notify relevant personnel for further processing and analysis.

[0075] Specifically, the predicted state parameters are fused with the actually received data, which emphasizes the importance of data fusion in improving data accuracy and reliability. By integrating information from multiple data sources, data fusion can eliminate noise and incorrect data, thereby enhancing data quality. Specifically in terms of methods, weighted average is a simple and effective way. It assigns different weights according to the credibility and accuracy of each data source and then calculates the average value. This method is simple and easy to implement, but it requires accurate evaluation of the weights of each data source. Bayesian estimation is a method based on probability theory. It updates the understanding and prediction of data by calculating the posterior probability and is applicable to scenarios dealing with uncertain information. Both of these methods can effectively improve the accuracy and reliability of data, providing more solid data support for the monitoring and optimization of energy storage systems.

[0076] More specifically, by real-time monitoring the network communication status, detecting communication anomalies and activating the data compensation mechanism, predicting the current state parameters using historical data and various machine learning algorithms, and then fusing with the actual data, the accuracy and reliability of the data are improved, the data fluctuations are smoothed, the abnormal impacts are reduced, the stable operation of the energy storage system is ensured, and the security and reliability of the system are further enhanced through the anomaly detection and alarm mechanism.

[0077] S3. According to the obtained state parameters and prediction data, combined with the geometric, physical, behavioral, and rule models of digital twin, construct the digital twin model of the energy storage system; among them, constructing the digital twin model of the energy storage system realizes the mapping between the physical entity and the virtual model, enabling the model to reflect the operating state of the energy storage system in real time, providing support for the monitoring, diagnosis, and optimization of the system, which not only improves the accuracy and reliability of data processing but also enhances the intelligent management level of the energy storage system.

[0078] Furthermore, according to the obtained state parameters and prediction data, combined with the geometric, physical, behavioral, and rule models of digital twin, construct the digital twin model of the energy storage system, specifically including:

[0079] Obtain the real-time state parameters and prediction data of the energy storage system, and establish a three-dimensional geometric model according to the physical structure and spatial layout of the energy storage system to accurately depict the appearance and internal structure of the system;

[0080] Adopt equivalent circuit models, thermal models, etc. to describe the physical characteristics of the energy storage system, including processes such as energy conversion and heat transfer; model the behavioral patterns and state transition laws of the energy storage system under different working conditions through methods such as finite state machines and Petri nets.

[0081] Using a business rule engine, grid dispatching rules, energy management strategies, etc. are embedded into the digital twin model to achieve global optimization. Multi-source heterogeneous data is fused with various models to build a mapping of the energy storage system from the physical world to the digital space, forming a real-time updated digital twin model, providing a basis for applications such as monitoring, prediction, optimization, and control.

[0082] Specifically, by combining the obtained state parameters and prediction data, a digital twin model of the energy storage system is constructed using the geometric, physical, behavioral, and rule models of the digital twin, realizing the real-time mapping between the physical entity and the virtual model. This not only enables the model to accurately reflect the operating state of the energy storage system, providing strong support for system monitoring, diagnosis, and optimization, but also significantly improves the accuracy and reliability of data processing, thereby enhancing the intelligent management level of the energy storage system and providing a solid guarantee for the efficient and stable operation of the energy storage system.

[0083] It should be explained that to construct a digital twin model of the energy storage system, first, the hierarchical relationship and assembly order of the model need to be clarified, and appropriate spatial constraint conditions, such as angles, contacts, offsets, etc., are added to achieve model assembly from parts to components and then to equipment. Then, the geometric model, physical model, behavioral model, and rule model are organically integrated. The geometric model is verified by measuring key geometric features; the physical model is verified based on comparative analysis of anchor point changes; the behavioral model constructs a sequence diagram and a state diagram and uses a formal verification method; the rule model is verified by comparing the drive response and the actual response. Finally, based on a data-driven approach, the random finite set (RFS) is used to model the simulation state and measurement data, and real-time measurement data is fused during the simulation operation through Bayesian inference methods to achieve dynamic correction of the model, thereby improving the accuracy and credibility of the model.

[0084] S4. Adopt the support vector machine algorithm, input the historical operation data of the battery pack, output the evaluation result of the aging degree, and adjust the parameter settings of the digital twin model according to the evaluation result of the aging degree to reduce the model calculation complexity. Through model simulation, obtain the optimized control strategy parameters;

[0085] Furthermore, through model simulation, obtain the optimized control strategy parameters, specifically including:

[0086] Obtain the historical operation data of the battery pack, including parameters such as voltage, current, and temperature. Preprocess the data to remove outliers and noise data to obtain a standardized data set. Use the support vector machine algorithm to train the standardized historical operation data of the battery pack, establish a battery aging degree evaluation model, and determine the optimal model parameters through cross-validation and grid search;

[0087] Using the trained battery aging degree evaluation model, predict the operation data of the new battery pack to obtain the evaluation result of the battery pack's aging degree, judge the health status and remaining life of the battery pack, and dynamically adjust the parameter settings of the digital twin model according to the evaluation result of the battery pack's aging degree, including material properties, physical characteristics, etc., to reduce the model calculation complexity and improve the simulation efficiency and accuracy;

[0088] Using the adjusted digital twin model, conduct multi-scenario and multi-condition simulation analysis on the battery pack to obtain the performance data and degradation trend of the battery pack under different conditions, providing a basis for optimizing the control strategy;

[0089] Through the simulation results and machine learning algorithms, such as reinforcement learning, genetic algorithms, etc., optimize the charge and discharge control strategy of the battery pack, including parameters such as charging current, cut-off voltage, temperature control, etc., to extend the life of the battery pack;

[0090] Apply the optimized control strategy parameters to the actual battery pack management system, conduct adaptive control and energy management on the battery pack, realize online monitoring and real-time optimization of the battery pack performance, and ensure the safe and reliable operation of the battery pack.

[0091] Specifically, by using the support vector machine algorithm to train the historical operation data of the battery pack, an aging degree evaluation model is established, realizing the accurate evaluation of the health status and remaining life of the battery pack. Dynamically adjusting the digital twin model parameters based on the evaluation results reduces the model calculation complexity and improves the simulation efficiency and accuracy. Further, using the adjusted digital twin model for multi-scenario and multi-condition simulation analysis, combined with machine learning algorithms to optimize the charge and discharge control strategy, not only extends the life of the battery pack, but also realizes online monitoring and real-time optimization of the battery pack performance by applying the optimized control strategy parameters to the actual battery pack management system, ensuring the safe and reliable operation of the battery pack.

[0092] S5. Transmit the optimized control strategy parameters to the actual energy storage system, and use differential privacy technology to encrypt the transmitted data.

[0093] Furthermore, transmit the optimized control strategy parameters to the actual energy storage system, and use differential privacy technology to encrypt the transmitted data, specifically including:

[0094] According to the optimized control strategy parameters, use differential privacy technology for data encryption processing, protect the parameter privacy by adding random noise, and transmit the encrypted control strategy parameters to the control module of the actual energy storage system through a secure communication protocol;

[0095] After receiving the encrypted control strategy parameters, the control module of the energy storage system uses the corresponding decryption algorithm to decrypt and obtain the optimized control strategy parameters;

[0096] The control module monitors and analyzes the operating status of the energy storage system in real time according to the decrypted control strategy parameters, and determines whether the control strategy needs to be adjusted.

[0097] When the control strategy needs to be adjusted, the control module generates corresponding control instructions according to the optimized control strategy parameters and sends them to the actuators of the energy storage system. After receiving the control instructions, the charging and discharging power, voltage and other parameters of the energy storage device are dynamically adjusted to achieve the optimized control of the energy storage system.

[0098] Specifically, the optimized control strategy parameters are encrypted by using differential privacy technology and transmitted to the control module of the actual energy storage system through a secure communication protocol, effectively protecting the parameter privacy and preventing data from being stolen or tampered with during transmission. After the control module decrypts, it monitors and analyzes the operating status of the energy storage system in real time according to the optimized control strategy parameters, dynamically adjusts the control strategy, generates and sends control instructions to the actuators, and realizes the precise adjustment of parameters such as the charging and discharging power and voltage of the energy storage device, thus ensuring the safe, reliable and efficient operation of the energy storage system and improving the overall performance and stability of the system.

[0099] Furthermore, after encrypting the transmitted data, it also includes: monitoring the operating status of the system in which the control strategy is deployed in the actual system in real time, judging the difference degree between the digital twin model and the actual system. When the difference exceeds the preset threshold, the model parameter adaptive adjustment mechanism is triggered to update the digital twin model parameters.

[0100] Furthermore, monitoring the operating status of the system in which the control strategy is deployed in the actual system in real time, judging the difference degree between the digital twin model and the actual system. When the difference exceeds the preset threshold, the model parameter adaptive adjustment mechanism is triggered to update the digital twin model parameters, which specifically includes:

[0101] Obtain the operating data of the actual system, including the execution of the control strategy, system status parameters, etc., as the input of real-time monitoring. Input the real-time monitoring data into the digital twin model and run the digital twin model to obtain the model output result.

[0102] Calculate the difference degree between the output of the digital twin model and the operating data of the actual system, and use indicators such as mean square error and relative error to measure the difference degree. Judge whether the difference degree exceeds the preset threshold. When it exceeds the threshold, trigger the model parameter adaptive adjustment mechanism. When it does not exceed the threshold, continue real-time monitoring.

[0103] After triggering the model parameter adaptive adjustment mechanism, use optimization algorithms such as gradient descent algorithm and evolutionary algorithm to adjust the parameters of the digital twin model to minimize the difference between the model output and the operating data of the actual system.

[0104] Update the adjusted model parameters into the digital twin model to obtain an updated digital twin model, making it synchronized with the actual system, ensuring that the digital twin model can accurately reflect the operating state of the actual system, and providing support for optimizing the control strategy.

[0105] Specifically, the dynamic synchronization between the digital twin model and the actual system is realized, improving the accuracy and reliability of the model, providing strong support for optimizing the control strategy, and ensuring the efficient and stable operation of the energy storage system.

[0106] Furthermore, after updating the digital twin model parameters, it also includes: continuously collecting the state parameters of the battery pack and communication quality indicators, using the random forest algorithm to establish a fault prediction model, and dynamically adjusting the simulation accuracy and computing resource allocation of the digital twin model according to the fault prediction results; through cyclic iterative optimization, realizing the continuous synchronization between the digital twin model and the actual system, and ensuring the effectiveness of the control strategy and the safety of system operation.

[0107] Furthermore, continuously collecting the state parameters of the battery pack and communication quality indicators, using the random forest algorithm to establish a fault prediction model, and dynamically adjusting the simulation accuracy and computing resource allocation of the digital twin model according to the fault prediction results, specifically including:

[0108] Continuously collect the state parameters of the battery pack, including voltage, current, temperature, etc., and at the same time obtain quality indicators such as signal strength and bit error rate of the communication link, and transmit the collected data to the cloud server; then use the random forest algorithm to train the fault prediction model, optimize the model parameters through methods such as cross-validation to improve the prediction accuracy, deploy the trained fault prediction model to the cloud, and receive the state parameters of the battery pack and communication quality indicators in real time to predict the fault probability of the battery pack;

[0109] When the predicted fault probability exceeds the preset threshold, dynamically adjust the simulation accuracy of the digital twin model to improve the time resolution and space resolution of the simulation calculation to obtain more detailed battery pack state information.

[0110] According to the simulation results of the digital twin model, optimize the allocation strategy of cloud computing resources, dynamically adjust the number and configuration of virtual machines to ensure the real-time performance and reliability of the simulation calculation;

[0111] Feed back the simulation results of the optimized digital twin model to the fault prediction model, continuously iterate and optimize the prediction algorithm to improve the accuracy and interpretability of the fault prediction, and provide a reliable basis for the operation and maintenance decision-making of the battery pack.

[0112] Embodiment 2

[0113] Please refer to Figure 3, this embodiment provides a communication and optimization system for an energy storage system, which is used to implement a communication and optimization method for an energy storage system, including:

[0114] Data acquisition and preprocessing module: Establish a communication link with the energy storage system through a wireless communication module, periodically obtain state parameters such as the voltage, current, and temperature of the battery pack, and store them in chronological order to form a time series data set;

[0115] Communication anomaly detection module: Adopt a time series analysis algorithm, such as the ARIMA model, to model the preprocessed data, extract the change trend characteristics of parameters such as voltage, current, and temperature, then judge the health status and remaining life of the battery pack according to the characteristics, and at the same time monitor key indicators such as communication delay and data packet loss rate to determine whether there is a communication anomaly; When an anomaly is judged, trigger the data compensation mechanism, and use historical data and machine learning algorithms to predict and compensate the current state parameters;

[0116] Model construction and optimization module: Combine the geometric, physical, behavioral, and rule models of digital twins. According to the obtained state parameters and prediction data, construct a digital twin model of the energy storage system. By clarifying the hierarchical relationship and assembly order of the model, adding spatial constraint conditions, perform model assembly from parts to components to equipment, and fuse multi-source heterogeneous data with various models. Use the random finite set (RFS) to model the simulation state and measurement data, and fuse real-time measurement data during the simulation operation through Bayesian inference methods to achieve dynamic correction of the model and improve the accuracy and credibility of the model;

[0117] Evaluation and control strategy optimization module: Adopt the support vector machine algorithm, input the historical operation data of the battery pack, output the evaluation result of the aging degree, and dynamically adjust the parameter settings of the digital twin model according to the evaluation result of the aging degree to reduce the model calculation complexity; Then use the adjusted digital twin model to perform multi-scenario and multi-condition simulation analysis to obtain the performance data and degradation trend of the battery pack under different conditions. Through the simulation results and machine learning algorithms, optimize the charge and discharge control strategy of the battery pack, extend the life of the battery pack, and apply the optimized control strategy parameters to the actual battery pack management system to realize online monitoring and real-time optimization of the battery pack performance;

[0118] Data encryption and transmission module: According to the optimized control strategy parameters, use differential privacy technology for data encryption processing, protect parameter privacy by adding random noise, and transmit the encrypted control strategy parameters to the control module of the actual energy storage system through a secure communication protocol. After receiving the encrypted control strategy parameters, the control module decrypts them and performs real-time monitoring and analysis on the operating state of the energy storage system according to the decrypted parameters to determine whether the control strategy needs to be adjusted, so as to realize the optimized control of the energy storage system;

[0119] The adaptive adjustment and fault prediction module monitors the operating state of the system where the control strategy is deployed in the actual system in real time, judges the degree of difference between the digital twin model and the actual system. When the difference exceeds the preset threshold, it triggers the adaptive adjustment mechanism of the model parameters, and uses an optimization algorithm to adjust the parameters of the digital twin model to minimize the difference between the model output and the actual system operation data, and updates the model parameters to ensure that the model can accurately reflect the operating state of the actual system. At the same time, it continuously collects the state parameters of the battery pack and the communication quality index, uses the random forest algorithm to establish a fault prediction model to obtain the fault prediction result, and dynamically adjusts the simulation accuracy and computing resource allocation of the digital twin model according to the fault prediction result. Through cyclic iterative optimization, the continuous synchronization of the digital twin model and the actual system is realized, ensuring the effectiveness of the control strategy and the safety of system operation.

[0120] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some changes or modifications to the equivalent embodiments with equivalent changes within the scope of the technical solution of the present invention. However, as long as it does not depart from the content of the technical solution of the present invention, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A communication and optimization method for an energy storage system, characterized in that: The steps include: The state parameters of the battery pack in the energy storage system are obtained through wireless communication, including voltage, current and temperature data. Then, the time series analysis method is used to extract the parameter change trend characteristics. Based on the extracted change trend characteristics, the health status and remaining life of the battery pack are judged. When an abnormal trend is found, the early warning mechanism is triggered. Then, combined with the state parameters and change trends of the battery pack, the charging and discharging strategy of the energy storage system is dynamically adjusted. Based on the extracted features, determine whether there is communication delay or data packet loss. If there is communication anomaly, start the data compensation mechanism and use historical data to predict the current state parameters. Based on the acquired state parameters and prediction data, the digital twin model of the energy storage system is constructed by combining the geometric, physical, behavioral and rule models of the digital twin. Use the support vector machine algorithm to input the historical operation data of the battery pack, output the aging assessment results, and adjust the parameter settings of the digital twin model based on the aging assessment results; Use the adjusted digital twin model to conduct multi-scenario and multi-operating condition simulation analysis on the battery pack to obtain the performance data and degradation trend of the battery pack under different conditions; Optimize the charge and discharge control strategy of the battery pack through simulation results and machine learning algorithms; Apply the optimized control strategy parameters to the actual battery pack management system to perform adaptive control and energy management on the battery pack, and conduct online monitoring and real-time optimization of the battery pack performance; The optimized control strategy parameters are transmitted to the actual energy storage system, and the transmission data is encrypted using differential privacy technology; Among them, according to the acquired state parameters and prediction data, combined with the geometric, physical, behavioral and rule models of the digital twin, a digital twin model of the energy storage system is constructed, which specifically includes: Obtain the real-time status parameters and prediction data of the energy storage system, establish a three-dimensional geometric model based on the physical structure and spatial layout of the energy storage system, and characterize the appearance and internal structure of the system; Use equivalent circuit models or thermal models to describe the physical characteristics of the energy storage system, and then model the behavior mode and state transition law of the energy storage system under different working conditions; By using the business rule engine, the grid dispatching rules and energy management strategies are embedded into the digital twin model for global optimization. Multi-source heterogeneous data are integrated with various models to build a mapping of the energy storage system from the physical world to the digital space, forming a real-time updated digital twin model.

2. The communication and optimization method of an energy storage system according to claim 1, characterized in that: The state parameters of the battery pack in the energy storage system are obtained through wireless communication, including voltage, current and temperature data, and then the time series analysis method is used to extract the trend characteristics of parameter changes, including: A communication link is established with the energy storage system through a wireless communication module, and voltage, current and temperature state parameters of the battery pack are periodically acquired. The acquired state parameter data are stored in chronological order to form a time series data set; The time series data is preprocessed to improve the data quality, and the time series analysis algorithm is used to model the preprocessed data to extract the change trend characteristics of voltage, current and temperature parameters.

3. The communication and optimization method of an energy storage system according to claim 1, characterized in that: Based on the extracted features, determine whether there is communication delay or data packet loss. If there is communication anomaly, start the data compensation mechanism and use historical data to predict the current state parameters, including: By monitoring the network communication status in real time, key indicators are obtained to determine whether there is a communication anomaly. When a communication anomaly is detected, the data compensation mechanism is triggered to estimate the current state parameters based on the pre-established prediction model using historical data; When predicting state parameters, we use a variety of machine learning algorithms to improve prediction accuracy through ensemble learning. For different types of state parameters, we select appropriate historical data features as input and dynamically adjust the hyperparameters of the prediction model to adapt to different data distributions and change trends. The predicted state parameters are integrated with the actual received data, and the state estimation value is obtained through weighted average or Bayesian estimation method; During the data fusion process, the weights of the predicted and measured values ​​are dynamically adjusted according to the severity and duration of the communication anomaly to smooth data fluctuations and reduce the impact of the anomaly. The compensated state parameters are then tested for anomalies, and when the abnormal deviation is found to exceed the preset threshold, the alarm mechanism is triggered.

4. The communication and optimization method of an energy storage system according to claim 1, characterized in that: Through model simulation, the optimized control strategy parameters are obtained, including: Obtain historical operating data of the battery pack, pre-process the data, remove outliers and noise data, obtain a standardized data set, use the support vector machine algorithm to train the standardized historical operating data of the battery pack, establish a battery aging assessment model, and determine the optimal model parameters through cross-validation and grid search; Use the trained battery aging assessment model to predict the new battery pack operating data, obtain the battery pack aging assessment results, judge the health status and remaining life of the battery pack, and dynamically adjust the parameter settings of the digital twin model based on the battery pack aging assessment results.

5. The communication and optimization method of an energy storage system according to claim 1, characterized in that: The optimized control strategy parameters are transmitted to the actual energy storage system, and the transmission data is encrypted using differential privacy technology, including: According to the optimized control strategy parameters, differential privacy technology is used to encrypt data, and random noise is added to protect parameter privacy. The encrypted control strategy parameters are transmitted to the control module of the actual energy storage system through a secure communication protocol. After receiving the encrypted control strategy parameters, the control module of the energy storage system uses the corresponding decryption algorithm to decrypt them and obtain the optimized control strategy parameters; The control module monitors and analyzes the operating status of the energy storage system in real time based on the decrypted control strategy parameters to determine whether the control strategy needs to be adjusted; When the control strategy needs to be adjusted, the control module generates corresponding control instructions based on the optimized control strategy parameters and sends them to the actuators of the energy storage system. The actuators then dynamically adjust the parameters of the energy storage equipment based on the received control instructions to optimize the control of the energy storage system.

6. The communication and optimization method of an energy storage system according to claim 5, characterized in that: After encrypting the transmitted data, it also includes: real-time monitoring of the system operation status of the control strategy deployed in the actual system, judging the degree of difference between the digital twin model and the actual system, and when the difference exceeds the preset threshold, triggering the model parameter adaptive adjustment mechanism to update the digital twin model parameters.

7. The communication and optimization method of an energy storage system according to claim 6, characterized in that: Real-time monitoring of the system operation status of the control strategy deployed in the actual system, judging the degree of difference between the digital twin model and the actual system. When the difference exceeds the preset threshold, the model parameter adaptive adjustment mechanism is triggered to update the digital twin model parameters, including: Obtain the actual system operation data, including the execution of the control strategy and system status parameters, as the input for real-time monitoring, input the real-time monitoring data into the digital twin model, run the digital twin model, and obtain the model output results; Calculate the difference between the digital twin model output and the actual system operation data, use mean square error and relative error indicators to measure the difference, and determine whether the difference exceeds the preset threshold. If it exceeds the threshold, the model parameter adaptive adjustment mechanism is triggered. If it does not exceed the threshold, real-time monitoring continues; After the model parameter adaptive adjustment mechanism is triggered, an optimization algorithm is used to adjust the parameters of the digital twin model to minimize the difference between the model output and the actual system operation data; The adjusted model parameters are updated to the digital twin model to obtain an updated digital twin model so that it is synchronized with the actual system.

8. The communication and optimization method of an energy storage system according to claim 7, characterized in that: After updating the digital twin model parameters, it also includes: continuously collecting battery pack status parameters and communication quality indicators, using the random forest algorithm to establish a fault prediction model, and dynamically adjusting the simulation accuracy and computing resource allocation of the digital twin model based on the fault prediction results; through cyclic iterative optimization, the digital twin model is continuously synchronized with the actual system.

9. A communication and optimization system for an energy storage system, used to implement a communication and optimization method for an energy storage system as claimed in any one of claims 1 to 8, characterized in that: include: The data acquisition and preprocessing module establishes a communication link with the energy storage system through the wireless communication module, periodically obtains the state parameters of the battery pack, and stores them in chronological order to form a time series data set; The communication anomaly detection module uses a time series analysis algorithm to model the preprocessed data, extract the change trend characteristics, and then judge the health status and remaining life of the battery pack based on the characteristics. When an anomaly is detected, the data compensation mechanism is triggered to use historical data and machine learning algorithms to predict and compensate for the current state parameters; The model building and optimization module builds a digital twin model of the energy storage system. By clarifying the hierarchical relationship and assembly order of the model and adding spatial constraints, the model is assembled from parts to components and then to equipment. Multi-source heterogeneous data is integrated with various models, and simulation states and measurement data are modeled using random finite sets. The real-time measurement data is integrated through the Bayesian reasoning method to dynamically correct the model. The evaluation and control strategy optimization module uses the support vector machine algorithm to dynamically adjust the parameter settings of the digital twin model according to the aging assessment results. The adjusted digital twin model is then used to perform simulation analysis of multiple scenarios and multiple working conditions to obtain the performance data and degradation trends of the battery pack under different conditions. The charging and discharging control strategy of the battery pack is optimized through simulation results and machine learning algorithms. The data encryption and transmission module uses differential privacy technology to encrypt data based on the optimized control strategy parameters, transmits the encrypted control strategy parameters to the control module of the actual energy storage system through a secure communication protocol, and decrypts them. Based on the decrypted parameters, the operating status of the energy storage system is monitored and analyzed in real time to determine whether the control strategy needs to be adjusted; The adaptive adjustment and fault prediction module monitors the system operation status of the control strategy deployed in the actual system in real time, determines the degree of difference between the digital twin model and the actual system, and then uses the optimization algorithm to adjust the parameters of the digital twin model. At the same time, the random forest algorithm is used to establish a fault prediction model to obtain fault prediction results, and the simulation accuracy and computing resource allocation of the digital twin model are dynamically adjusted according to the fault prediction results.

Images