Method and system for evaluating electrical performance of energy storage system

Abstract

The invention relates to the field of energy storage system management and battery performance evaluation, in particular to an energy storage system electrical performance evaluation method and system.The method comprises the steps that battery cluster data and environment data are collected in real time, the charging and discharging process of a battery is accurately simulated in combination with simulation data and a digital twin model, and the future running state of the battery is predicted. The reinforcement learning algorithm adaptively optimizes the charging and discharging strategy according to simulation data and feedback results, and adjusts the charging power, the discharging rate and the temperature control strategy in real time through model prediction control. The method effectively evaluates the charging and discharging efficiency, the internal resistance change, the temperature fluctuation and the remaining service life of the battery, and provides decision support for operation optimization and maintenance of the system. According to the invention, the energy storage system can significantly improve the charging and discharging efficiency, prolong the service life of the battery, and ensure the stable and efficient operation of the system under different load and environmental conditions.

Description

Technical Field

[0001] The present invention relates to the field of energy storage system management and battery performance evaluation, and in particular to an energy storage system electrical performance evaluation method and system. Background Art

[0002] Energy storage systems are becoming increasingly important in modern energy management, especially in the energy dispatch between renewable energy (such as solar energy and wind energy) and the power grid. Energy storage systems can effectively balance the difference between supply and demand and optimize energy efficiency. However, with the expansion of energy storage systems and the diversification of application scenarios, how to accurately evaluate battery performance, ensure battery charging and discharging efficiency and safety, and extend battery life have become urgent issues to be solved. Battery performance evaluation does not only rely on traditional charging and discharging data, but also needs to comprehensively consider the health of the battery, environmental changes, and the impact of load fluctuations on battery performance.

[0003] In the prior art, the electrical performance evaluation of energy storage systems often relies on the monitoring of simple battery electrical parameters (such as voltage and current) and data analysis through certain algorithms. However, this evaluation method has major deficiencies when dealing with issues such as changes in battery internal resistance, thermal management, and life attenuation effects. For example, existing technologies are usually unable to adapt to battery health status and environmental changes in real time, and often rely on fixed charging and discharging parameters and temperature control strategies, resulting in low battery charging and discharging efficiency, accelerated life attenuation, and even the risk of overheating and overloading. In addition, traditional evaluation methods lack precise environmental adaptability and cannot effectively predict the long-term performance of batteries under different load and environmental conditions. Summary of the invention

[0004] In view of the many problems existing in the above-mentioned prior art, the present invention provides an energy storage system electrical performance evaluation method and system. The present invention uses a digital twin model to virtually simulate the battery, combines real-time battery data and environmental data, uses a reinforcement learning algorithm to optimize the charging and discharging strategy, and adjusts the battery operating status in real time through model predictive control. By comprehensively evaluating the battery's charging and discharging efficiency, internal resistance changes, temperature fluctuations, and remaining service life, accurate prediction and intelligent adjustment of battery performance are achieved. The present invention effectively improves battery efficiency, extends battery life, and ensures efficient and stable operation of the energy storage system in a changing environment.

[0005] A method for evaluating the electrical performance of an energy storage system comprises the following steps:

[0006] Collecting electrical data and environmental data of the battery cluster, and cleaning and removing noise from the collected data to obtain battery cluster data and environmental data; extracting battery characteristic data and environmental characteristic data based on the battery cluster data and environmental data;

[0007] Based on the battery characteristic data and environmental characteristic data, a digital twin model is established to perform virtual simulation to simulate the battery charging and discharging process under different load and environmental conditions, and obtain simulated charging and discharging data, thermal management data, and battery life attenuation data;

[0008] Based on the simulated charging and discharging data, thermal management data and battery life attenuation data, the reinforcement learning algorithm is applied to generate optimization strategy data, the battery charging and discharging is adjusted in real time using model predictive control, and the optimization strategy is adjusted based on real-time feedback to obtain the final optimization control data; based on the final optimization control data, the electrical performance of the energy storage system battery is evaluated to obtain the evaluation results.

[0009] Preferably, the data cleaning and noise removal steps include removing noise from the voltage and current signals in the battery cluster data by using Kalman filtering, smoothing the temperature and humidity signals in the environmental data by using sliding mean filtering, and detecting outliers in the gas leakage data by setting thresholds.

[0010] Preferably, the battery characteristic data includes the battery's state of charge, temperature change rate, and internal resistance, wherein:

[0011] The battery status charge is calculated by collecting the voltage and current data of the battery cluster in combination with a preset battery capacity model; the battery temperature change rate is obtained by real-time monitoring of the battery temperature and calculating the change per unit time; the battery internal resistance is calculated by the ratio of the voltage change to the current change measured during the battery charging and discharging process.

[0012] Preferably, the environmental characteristic data includes the impact of ambient temperature and humidity on battery performance, and the environmental characteristic data is obtained by combining real-time monitored ambient temperature and humidity data with a battery adaptability analysis model, and is used to evaluate the impact of environmental changes on battery charging and discharging efficiency and life.

[0013] Preferably, the virtual simulation step includes:

[0014] By building a digital twin model based on the battery's electrical characteristics, thermal management characteristics, and life decay characteristics, combined with the real-time data and environmental data of the battery cluster, a virtual simulation of the charging and discharging process is performed;

[0015] The simulated charge and discharge data generated by the simulation process includes information reflecting the changes in battery voltage and current;

[0016] Simulation thermal management data includes the temperature distribution and heat dissipation characteristics of the battery during operation;

[0017] The simulated battery life decay data includes information reflecting the battery's internal resistance growth and capacity decay.

[0018] Preferably, the reinforcement learning algorithm generates optimization strategy data by the following steps:

[0019] In the simulation environment, a reward function is defined based on the simulation data and the set evaluation objectives;

[0020] Adjust the battery's charge and discharge timing, charge and discharge current, and charge and discharge voltage through multiple rounds of trial and error;

[0021] Optimization strategy data for optimizing battery charging and discharging efficiency and extending battery life is generated based on trial and error learning results.

[0022] Preferably, the model predictive control comprises:

[0023] Based on real-time collected battery cluster data, simulated charge and discharge data, simulated thermal management data, and simulated battery life attenuation data, the future battery charge and discharge status is predicted; the battery charging power and discharge rate are dynamically adjusted to reduce battery energy loss and extend battery life; and the battery cluster is ensured to operate in the optimal state under different load and environmental conditions.

[0024] Preferably, the evaluation step generates evaluation results based on the final optimized control data, and the evaluation results include the battery's charge and discharge efficiency, internal resistance change, temperature fluctuation, and remaining service life, and provides decision support for operation optimization and maintenance of the energy storage system through a comprehensive evaluation report.

[0025] Preferably, the evaluation results are used to adjust the operation strategy of the energy storage system, including optimizing the charge and discharge time of the battery, adjusting the charge and discharge parameters, and setting a temperature control strategy that adapts to different environmental conditions.

[0026] A system for implementing the energy storage system electrical performance evaluation method, the system comprising:

[0027] A data acquisition module, used for collecting electrical data and environmental data of the battery cluster and transmitting the collected data to the data processing module;

[0028] A data processing module, used for cleaning and removing noise from the collected battery cluster data and environmental data, generating processed battery cluster data and environmental data, and transmitting the processed data to the feature extraction module;

[0029] A feature extraction module, used for extracting battery feature data and environmental feature data based on the processed battery cluster data and environmental data, and transmitting the extracted feature data to the digital twin model module;

[0030] A digital twin model module is used to establish a digital twin model based on the extracted battery characteristic data and environmental characteristic data, perform virtual simulation, simulate the charging and discharging process of the battery under different load and environmental conditions, generate simulated charging and discharging data, simulated thermal management data and simulated battery life attenuation data, and transmit the simulation data to the optimization control module;

[0031] An optimization control module is used to generate optimization strategy data based on simulated charge and discharge data, simulated thermal management data, and simulated battery life decay data by applying a reinforcement learning algorithm, and to adjust battery charge and discharge in real time by using model predictive control, adjust the optimization strategy based on real-time feedback, generate final optimization control data, and transmit the results to an evaluation module;

[0032] The evaluation module is used to evaluate the electrical performance of the battery of the energy storage system based on the final optimized control data and generate an evaluation result.

[0033] Compared with the prior art, the advantages and beneficial effects of the present invention are:

[0034] The present invention achieves accurate evaluation and intelligent optimization of batteries in energy storage systems by introducing digital twin models, reinforcement learning algorithms, and model predictive control (MPC) technology. By combining real-time data with simulation data, the system can predict the future operating status of the battery, automatically adjust the charging and discharging parameters, optimize the charging and discharging time, and dynamically adjust the temperature control strategy according to the battery health status. Compared with traditional technologies, the present invention solves the defects of the prior art such as insufficiently precise battery performance evaluation, insufficiently intelligent temperature control management, and inaccurate life attenuation prediction through an adaptive control strategy, significantly improving the charging and discharging efficiency and extending the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the process of the present invention;

[0036] Figure 2 It is a simulation schematic diagram of the digital twin model in the present invention;

[0037] Figure 3 This is a schematic diagram of the optimization strategy of the reinforcement learning algorithm in the present invention;

[0038] Figure 4 This is a schematic diagram of charging and discharging regulation by model predictive control (MPC) in the present invention;

[0039] Figure 5 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0040] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is obvious that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0041] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0042] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0043] like Figure 1 As shown, a method for evaluating the electrical performance of an energy storage system comprises the following steps:

[0044] Collecting electrical data and environmental data of the battery cluster, and cleaning and removing noise from the collected data to obtain battery cluster data and environmental data; extracting battery characteristic data and environmental characteristic data based on the battery cluster data and environmental data;

[0045] Preferably, the data cleaning and noise removal steps include removing noise from the voltage and current signals in the battery cluster data by using Kalman filtering, smoothing the temperature and humidity signals in the environmental data by using sliding mean filtering, and detecting outliers in the gas leakage data by setting thresholds.

[0046] In the process of evaluating the electrical performance of the energy storage system, the collection and subsequent processing of the electrical data and environmental data of the battery cluster are the basis for the accuracy and reliability of the entire evaluation system. The electrical data of the battery cluster usually includes key parameters such as the voltage, current and internal resistance of the battery, while the environmental data includes temperature, humidity and possible gas leakage information. Since the data collection process may be interfered by various factors, such as sensor accuracy limitations, external environmental changes, etc., the collected data often contains noise and errors. Therefore, in order to ensure the accuracy of the data, it is first necessary to clean and remove noise from the collected battery cluster data and environmental data to ensure that subsequent analysis can be based on high-quality data.

[0047] In order to achieve this goal, the present invention adopts two data processing technologies, Kalman filtering and sliding mean filtering, which are respectively used for noise removal of the electrical data and environmental data of the battery cluster. Kalman filtering is an adaptive filtering technology, which is widely used for noise removal in dynamic systems. It can estimate the optimal state based on the system model and actual observations, thereby reducing the impact of noise on battery electrical data (such as voltage and current). In the present invention, Kalman filtering is used to remove noise from battery voltage and current signals. In particular, during the charging and discharging process, the voltage and current data of the battery may fluctuate due to equipment fluctuations, sensor errors or changes in the external environment. Kalman filtering effectively removes these unstable factors through weighted averaging, retains the valid signals in the data, and improves the accuracy and reliability of the battery data.

[0048] For environmental data, especially temperature and humidity data, the present invention adopts a sliding mean filtering method. By calculating the average value of the data within a certain time window, the sliding mean filter can smooth the rapid fluctuations caused by external disturbances, remove sudden noise, and ensure the stability of the temperature and humidity data. For example, the ambient temperature data of the energy storage system may be affected by factors such as equipment start and stop, ambient temperature fluctuations, etc. The sliding mean filter can smooth these rapid changes to make the temperature and humidity data more stable and reliable. In addition, for gas leakage data, in order to remove the influence of outliers, the solution detects and eliminates obvious abnormal data by setting a threshold, which ensures that only accurate environmental data is relied upon during the evaluation process, avoiding errors from interfering with the battery performance evaluation results.

[0049] Through the comprehensive application of Kalman filtering, sliding mean filtering and outlier detection, the battery cluster data and environmental data can be cleaned and noise removed to effectively remove errors and fluctuations caused by equipment and environmental changes, and finally obtain more accurate and reliable input data, providing solid data support for subsequent battery performance evaluation and optimization control. These technical means ensure the high accuracy and high reliability of the electrical performance evaluation method of the entire energy storage system, and can effectively cope with complex environmental factors in practical applications.

[0050] like Figure 2 As shown, based on the battery characteristic data and environmental characteristic data, a digital twin model is established to perform virtual simulation to simulate the charging and discharging process of the battery under different load and environmental conditions, and obtain simulated charging and discharging data, thermal management data and battery life attenuation data;

[0051] The core principle of the digital twin model is to create a virtual copy in the computer that is highly consistent with the actual battery system operating status by modeling the dynamic characteristics of the battery cluster. To achieve this goal, first of all, a digital twin model of the battery is established based on the collected battery characteristic data and environmental characteristic data, combined with the physical and electrochemical models of the battery. This model not only includes the electrical characteristics of the battery (such as voltage, current, internal resistance, etc.), but also combines the thermal management characteristics and life attenuation characteristics of the battery. The electrical characteristics of the battery cluster are closely related to environmental characteristic data such as temperature and humidity. Therefore, the model needs to be dynamically updated based on the real-time collected battery cluster data and environmental data to reflect the actual performance of the battery under different load and environmental conditions.

[0052] In the virtual simulation stage, the digital twin model generates simulation data by simulating the battery's charging and discharging process. These simulation data include three key categories: simulated charging and discharging data, thermal management data, and battery life decay data. The simulated charging and discharging data can simulate the voltage and current changes of the battery under different load conditions, helping to evaluate the battery's charging and discharging efficiency; the thermal management data describes the heat distribution and temperature changes generated by the battery during operation due to the charging and discharging process, analyzes the battery's heat dissipation performance, and avoids battery damage due to overheating; the battery life decay data simulates the changes in the battery's internal resistance and the decay of the battery capacity during long-term use, and predicts the remaining service life of the battery.

[0053] Through virtual simulation of the digital twin model, the system can accurately simulate the battery's charging and discharging process, thermal management characteristics, and life attenuation characteristics without actual interference. This simulation provides efficient and actionable data support for the performance evaluation of the energy storage system, making it possible to predict the performance of the battery under specific conditions before actual deployment, thus avoiding potential safety risks.

[0054] By establishing a digital twin model and performing virtual simulation, the present invention can comprehensively evaluate the performance of the battery cluster under different load and environmental conditions, and provide detailed information on battery charging and discharging efficiency, thermal management effects, and life decay process. By analyzing the simulated charging and discharging data, the system can optimize the battery's charging and discharging strategy to ensure that the battery's efficiency is maximized during operation and extend the battery's service life. Thermal management data helps monitor battery temperature changes to ensure that the battery operates within a safe temperature range and prevent battery performance degradation or damage due to excessively high or low temperatures. Life decay data provides a basis for predicting the remaining life of the battery, helping to identify battery clusters that need to be replaced or maintained in advance, and improving the reliability and economy of the entire energy storage system.

[0055] Through virtual simulation, the digital twin model can provide high-precision prediction and optimization solutions for battery cluster performance evaluation, reduce risks in actual applications, and provide data support for the battery management system (BMS), thereby realizing intelligent management and optimized scheduling of battery clusters. This not only effectively improves the operating efficiency of the energy storage system, but also reduces battery loss and maintenance costs, and ensures the long-term stability and safety of the system.

[0056] Take the charging and discharging process of a battery cluster as an example. Assuming that in an energy storage system, the battery cluster operates in a high-temperature environment, the system simulates the charging and discharging process of the battery cluster in real time through a digital twin model. The simulation results show that under high temperature conditions, the internal resistance of the battery increases and the charging efficiency decreases. Excessive temperature may also cause the battery's overheating protection mechanism to activate, thereby affecting the overall efficiency of the system. Through the thermal management data provided by the model, the system can adjust the charging current in time, reduce the temperature rise of the battery, avoid the occurrence of overheating problems, and optimize the charging strategy, improve the battery's charging and discharging efficiency, and extend the battery life. This simulation based on the digital twin model not only improves the safety of the battery cluster, but also effectively optimizes the system's operating efficiency.

[0057] Preferably, the battery characteristic data includes the battery's state of charge, temperature change rate, and internal resistance, wherein:

[0058] The battery status charge is calculated by collecting the voltage and current data of the battery cluster in combination with a preset battery capacity model; the battery temperature change rate is obtained by real-time monitoring of the battery temperature and calculating the change per unit time; the battery internal resistance is calculated by the ratio of the voltage change to the current change measured during the battery charging and discharging process.

[0059] The state of charge (SOC) of a battery refers to the percentage of electrical energy currently stored in the battery. It is an important parameter for evaluating the remaining power of the battery and estimating the battery life. The calculation of SOC usually relies on the voltage and current data of the battery cluster, combined with the preset capacity model of the battery. Specifically, the calculation of SOC measures the voltage and current of the battery cluster and uses the preset battery capacity model to infer the battery's state of charge. In energy storage systems, the SOC calculation formula generally uses the integral method:

[0060]

[0061] Among them, SOC(t) is the battery state charge at time t (%); SOC(t0) is the initial state charge at time t0 (%); I(t) is the battery charge and discharge current (A); C is the rated capacity of the battery (Ah).

[0062] This calculation method based on current integration can update the SOC value in real time, thereby reflecting the battery's state of charge. In the present invention, the voltage and current of the battery cluster are collected in real time by sensors, and combined with a preset capacity model (taking into account factors such as battery discharge efficiency and loss), the current state of charge of the battery is calculated. This process can provide high-precision battery power estimation, help determine whether the battery is close to the critical state of overcharge and overdischarge, and ensure that the battery operates within a safe and stable range.

[0063] The battery temperature change rate is another important battery characteristic data, which reflects the thermal management performance of the battery by monitoring the battery temperature change rate. The battery generates heat during the charging and discharging process, especially when charging and discharging at high currents, the battery temperature will rise. If the temperature is not properly controlled, it may cause the battery to overheat, affecting its performance and even causing safety problems. The temperature change rate is usually obtained by monitoring the battery temperature in real time through a temperature sensor and calculating the temperature change per unit time. Specifically, the temperature change rate is calculated as follows:

[0064]

[0065] Wherein, T(t) is the battery temperature at time t (°C); T(t-1) is the battery temperature at time t-1 (°C); Δt is the time interval (seconds).

[0066] This calculation method can be used to obtain the temperature change rate of the battery during the charging and discharging process. If the temperature change rate is too fast, it means that there may be a problem with the battery's thermal management system. The system can adjust the charging and discharging strategy or start the cooling system in time based on this data to prevent the battery from overheating.

[0067] The internal resistance of a battery is one of the important indicators of battery performance. It directly affects the battery's charging and discharging efficiency and life. During the battery's charging and discharging process, the current passing through the battery is limited by the internal resistance. An increase in the battery's internal resistance usually means a decrease in battery life and capacity. The battery's internal resistance is calculated by the changes in voltage and current during the battery's charging and discharging process, usually using Ohm's law. Specifically, the battery's internal resistance R int By measuring the ratio of voltage and current changes during battery charging and discharging, we can calculate:

[0068]

[0069] Among them, R int is the internal resistance of the battery (Ω); ΔV is the voltage change of the battery during the charge and discharge process (V); ΔI is the current change of the battery during the charge and discharge process (A).

[0070] Through this calculation method, the system can monitor the changes in the internal resistance of the battery in real time and promptly detect whether the battery has entered the aging stage. When the internal resistance of the battery increases, the discharge efficiency of the battery decreases and the battery heat increases, which affects the overall efficiency and safety of the energy storage system. Therefore, real-time monitoring and analysis of internal resistance is very important for battery health management.

[0071] Embodiment: In an energy storage system, assuming that the battery cluster is undergoing a charging operation, the system monitors the current, voltage, and temperature data of the battery cluster in real time, and applies the above calculation method to obtain the battery status charge, temperature change rate, and internal resistance. If the battery's temperature change rate is large, it may mean that the battery is experiencing rapid charge and discharge or there is a problem with the thermal management system. The system will adjust the charging power based on this information to avoid excessive battery temperature. Similarly, by monitoring the internal resistance of the battery, if it is found that the internal resistance increases rapidly, the system can determine through analysis whether the battery life is declining, and take corresponding maintenance measures in advance to ensure the stability and safety of the system.

[0072] Preferably, the environmental characteristic data includes the impact of ambient temperature and humidity on battery performance, and the environmental characteristic data is obtained by combining real-time monitored ambient temperature and humidity data with a battery adaptability analysis model, and is used to evaluate the impact of environmental changes on battery charging and discharging efficiency and life.

[0073] Ambient temperature and humidity are key factors that affect battery performance. Battery chemical reactions, ion migration, and charging efficiency are all directly related to temperature. When the temperature is too high or too low, the battery's charging and discharging efficiency will decrease, and may even cause battery damage. At the same time, high humidity may also cause battery short circuits or corrosion. Therefore, monitoring and evaluating the impact of environmental conditions on battery performance is crucial to the health management of energy storage systems.

[0074] In order to quantify the impact of temperature and humidity on battery performance, the present invention combines a battery adaptability analysis model. The model can evaluate the impact of changes in environmental conditions on the battery's charging and discharging efficiency, thermal management capabilities, and life decay process based on real-time monitored ambient temperature and humidity data. The core principle of the battery adaptability analysis model is to establish a mathematical model that can describe the battery's operating performance under different environmental conditions based on the physical and chemical properties of the battery, as well as the impact of temperature and humidity on the battery's internal reactions and thermal state. The model combines the battery's electrical characteristics (such as voltage, current, internal resistance) with environmental data (temperature, humidity) to simulate the real-time impact of environmental changes on battery performance.

[0075] In the implementation process, we first install temperature and humidity sensors to monitor the temperature and humidity data of the environment where the energy storage system is located in real time. After processing, these data are input into the battery adaptability analysis model together with the battery performance data (such as battery voltage, current, internal resistance, etc.). Based on the known battery behavior patterns and environmental impact laws, the model calculates the specific impact of environmental factors on battery performance and evaluates the battery's charge and discharge efficiency, thermal management effect, and life decay rate under current environmental conditions.

[0076] The core function of the battery adaptability analysis model is to evaluate the impact of environmental changes on the battery charging and discharging process in real time. For example, in a high temperature environment, the internal resistance of the battery may increase, resulting in a decrease in charging efficiency. At the same time, the temperature rise of the battery may also accelerate the life decay process. The model will update these effects in real time and output corresponding adjustment strategies, such as adjusting the charging current or starting the temperature control system to cope with adverse environmental changes. In addition, the model can also predict the remaining life of the battery under different temperature and humidity conditions, and provide specific optimization suggestions, such as reducing the charging rate or changing the charge and discharge cycle to delay the battery life decay.

[0077] By combining environmental temperature and humidity data with a battery adaptability analysis model, the present invention can effectively evaluate the real-time impact of the environment on battery performance and provide an optimized control strategy to ensure that the battery operates in the best working state.

[0078] Temperature changes directly affect the electrochemical reaction rate of the battery. Too high or too low a temperature will reduce the battery's charging efficiency. Through real-time monitoring of environmental data, the system can dynamically adjust charging parameters, such as reducing charging current or extending the charging cycle, to ensure charging efficiency and avoid overheating.

[0079] Batteries generate heat during the charging and discharging process. Excessive temperature may cause the battery to overheat and even cause safety problems. By evaluating the impact of ambient temperature on battery temperature in real time, the model can predict the thermal state of the battery and adjust the battery cooling system (such as starting the fan, enhancing heat dissipation, etc.) as needed to ensure that the battery operates within a safe temperature range.

[0080] The battery life decay process is closely related to temperature. Excessive temperature or temperature fluctuations may accelerate the battery life decay. Through the correlation analysis of ambient temperature and humidity with battery internal resistance, capacity and other life decay indicators, the system can predict the remaining service life of the battery and take maintenance measures in advance based on the prediction results to delay the battery decay.

[0081] Example: In an energy storage system, assuming that the system is operating in a high temperature summer environment, the real-time monitored temperature is 40°C and the humidity is 80%. By inputting these data into the battery adaptability analysis model, the system finds that the internal resistance of the battery has changed significantly at this time, and the charging efficiency has decreased. Based on the feedback from the model, the system automatically adjusts the charging strategy, reduces the charging current, and starts the cooling system to reduce the battery temperature rise, while reducing the high temperature duration during the charging cycle. This adjustment strategy based on real-time environmental data can effectively prevent battery overheating, maintain battery charging and discharging efficiency, and extend battery life.

[0082] Preferably, the virtual simulation step includes:

[0083] By building a digital twin model based on the battery's electrical characteristics, thermal management characteristics, and life decay characteristics, combined with the real-time data and environmental data of the battery cluster, a virtual simulation of the charging and discharging process is performed;

[0084] The simulated charge and discharge data generated by the simulation process includes information reflecting the changes in battery voltage and current;

[0085] Simulation thermal management data includes the temperature distribution and heat dissipation characteristics of the battery during operation;

[0086] The simulated battery life decay data includes information reflecting the battery's internal resistance growth and capacity decay.

[0087] The virtual simulation step first simulates the actual behavior of the battery cluster by establishing a digital twin model based on the battery electrical characteristics, thermal management characteristics and life decay characteristics. The digital twin model is a highly accurate virtual model that can truly reproduce the operation of the battery under different working conditions, including the battery's charge and discharge state, thermal state, and life decay effects that occur over time. In the present invention, the establishment of the digital twin model needs to rely on the electrical data, thermal management data and environmental data of the battery cluster. The electrical characteristics of the battery, such as voltage, current, internal resistance, etc., are direct indicators of the current working state and efficiency of the battery; the thermal management characteristics include the heat and temperature distribution generated by the battery during the charge and discharge process, which directly affect the safety and performance of the battery; and the battery's life decay characteristics reflect the performance decay of the battery over time, including increased internal resistance and capacity decay.

[0088] Simulated charge and discharge data is an important part of the virtual simulation process. It can reflect in detail the voltage and current changes of the battery under different load conditions. In actual applications, the charging and discharging process of the battery may be affected by many factors, including ambient temperature, battery health status, and load changes. Through the simulation of the digital twin model, the battery charging and discharging process can be simulated in different scenarios, and the voltage and current changes of the battery can be predicted. For example, when the battery is working in a low temperature environment, a large voltage drop may occur. Simulation can help identify this phenomenon in advance and adjust the charging strategy or optimize the system load. Simulated charge and discharge data provides in-depth insights into battery performance, helping to optimize the battery's charge and discharge strategy and maximize the battery's energy output efficiency.

[0089] Batteries generate heat during the charging and discharging process, and the distribution and heat dissipation of this heat directly affect the working performance and safety of the battery. In virtual simulation, thermal management data simulates the temperature changes and heat distribution of the battery under different charging and discharging states. The increase in battery temperature usually leads to an increase in the internal resistance of the battery, which affects the discharge efficiency of the battery and may even lead to safety risks of the battery, such as overheating and thermal runaway. Therefore, simulated thermal management data provides valuable information for the system, which can help determine the heat dissipation capacity of the battery under different environmental conditions. For example, when the battery is working under high load, the digital twin model can simulate the temperature distribution inside the battery and predict the possible thermal runaway risk of the battery, thereby providing system designers with necessary adjustment suggestions, such as optimizing the cooling system or reducing the load power to ensure that the battery operates within a safe temperature range.

[0090] As batteries are used, the battery life decay effect is inevitable, especially the increase in internal resistance and the decay of capacity will significantly affect the performance of the battery. The increase in internal resistance and the decay of capacity of the battery are usually caused by chemical reactions, temperature changes during the charging and discharging process, and material degradation caused by long-term use. During the virtual simulation process, the digital twin model can simulate the life decay process of the battery under different charging and discharging cycles and predict the remaining life of the battery. Specifically, the increase in internal resistance usually means that the charging and discharging efficiency of the battery decreases, while the decay of capacity means that the maximum energy that the battery can provide is reduced. By simulating the battery life decay data, it is possible to predict the performance of the battery in the next few years, and take maintenance or replacement measures in advance to avoid the overall efficiency of the energy storage system from decreasing due to battery decay. For example, during the simulation process, if the system predicts that the internal resistance of the battery will increase significantly in a certain period of time in the future, the charging strategy can be adjusted in advance or the battery can be regularly maintained to extend the battery life.

[0091] By building a digital twin model and performing virtual simulation, the energy storage system can comprehensively evaluate the performance of the battery cluster before actual deployment. The simulated charge and discharge data provides a basis for optimizing the charge and discharge strategy, the simulated thermal management data ensures that the battery operates within a safe temperature range, and the simulated battery life decay data reveals the decline trend of battery performance in advance. The comprehensive application of all these simulation data helps to improve the overall efficiency and safety of the energy storage system and provides a scientific basis for system optimization.

[0092] Example: In an energy storage system, assume that the simulation results of the digital twin model show that under high temperature and high load conditions, the temperature of the battery rises sharply, and the internal resistance of the battery increases, resulting in reduced charging and discharging efficiency. Based on this simulation result, the system can start the cooling system in advance and adjust the charging current to avoid battery overheating or premature decay. Furthermore, the system can predict the service life of the battery based on the battery life decay data, and provide decision support for later maintenance or battery replacement, thereby avoiding unplanned downtime of the system and increasing maintenance costs.

[0093] Based on the simulated charging and discharging data, thermal management data and battery life attenuation data, the reinforcement learning algorithm is applied to generate optimization strategy data, the battery charging and discharging is adjusted in real time using model predictive control, and the optimization strategy is adjusted based on real-time feedback to obtain the final optimization control data; based on the final optimization control data, the electrical performance of the energy storage system battery is evaluated to obtain the evaluation results.

[0094] Reinforcement learning is a machine learning method that uses simulation and reward mechanisms to enable the system to automatically optimize the decision-making process in actual operation. In the present invention, the reinforcement learning algorithm is used to generate optimization strategy data, which is mainly used to adjust the battery's charging and discharging timing, current and voltage. By combining with simulated charging and discharging data, thermal management data, and battery life decay data, reinforcement learning can continuously learn the battery's performance under different environmental and load conditions, and then generate the optimal charging and discharging strategy. Specifically, Figure 3 As shown, the reinforcement learning algorithm analyzes historical data and real-time data to predict future charging and discharging status, and adjusts the strategy based on actual feedback to achieve optimal charging efficiency and battery protection effect.

[0095] In the application of the present invention, the reinforcement learning algorithm generates a series of optimization strategies by training the simulation data, and these strategies are constantly adjusted as the battery operating environment changes. During each charge and discharge process, the algorithm adjusts the charging current, charging voltage, charging time and other parameters according to the actual battery performance feedback to ensure that the battery is always in the best working state.

[0096] like Figure 4As shown, model predictive control (MPC) is a control method based on system model and optimization algorithm. In the present invention, MPC is used to adjust the charging and discharging process of the battery in real time. The core idea of ​​MPC is to predict the state of the battery in the future based on the current operating state and environmental conditions of the battery, and optimize the control strategy to minimize the energy loss and battery loss during the charging and discharging process. MPC calculates the optimal charging and discharging current and voltage by solving the optimal control problem, and makes real-time adjustments.

[0097] In energy storage systems, MPC not only relies on the current battery status data, but also takes into account simulated charge and discharge data, thermal management data, and battery life decay data to comprehensively evaluate the battery's operating performance under different working conditions. MPC will adjust the battery's charge and discharge strategy in real time, and adjust and optimize the control data based on real-time feedback to ensure that the battery always maintains the best charge and discharge efficiency under different load and environmental conditions.

[0098] Real-time feedback to adjust optimization strategy The combination of reinforcement learning and MPC enables the system to provide real-time feedback on the battery's operating conditions, and the optimization strategy will be dynamically adjusted as the battery status changes. For example, when the battery temperature is too high, the system can identify this situation through the reinforcement learning algorithm and reduce the charging current to prevent the battery from overheating. For another example, when the battery life decays more rapidly, MPC can predict the battery's decay trend based on the change in the battery's internal resistance and adjust the charge and discharge strategy to extend the battery life.

[0099] In addition, by combining simulated charge and discharge data and environmental data with real-time feedback, the system can dynamically adjust the charging strategy according to the actual state of the battery (such as SOC, internal resistance, temperature, etc.), avoiding the battery being in an unfavorable state for a long time, thereby preventing damage to the battery caused by overcharging or over-discharging.

[0100] Final optimization control data and battery performance evaluation Based on the optimization control data adjusted in real time by reinforcement learning algorithm and model predictive control, the energy storage system can evaluate the electrical performance of the battery and obtain the battery evaluation results. These evaluation results include the battery's charging and discharging efficiency, internal resistance change, temperature fluctuation, and the remaining service life of the battery. After the optimization control data is continuously adjusted through real-time feedback, it ultimately helps evaluate the performance of the battery in long-term operation and provides data support for system maintenance, energy management, and life cycle prediction.

[0101] Through reinforcement learning and model predictive control based on simulated charge and discharge data, thermal management data, and battery life decay data, the present invention can significantly improve the charge and discharge efficiency of the energy storage system, extend the battery life, and minimize energy loss. This method can achieve the following key effects:

[0102] Reinforcement learning algorithms can adjust the battery's charging and discharging strategies in real time by analyzing historical and real-time data to ensure that the battery always maintains the best charging and discharging efficiency under different load conditions. For example, reinforcement learning can identify the problem of reduced battery charging efficiency in certain high-temperature environments and adjust the charging current in time to avoid efficiency loss.

[0103] Through simulation and real-time feedback of battery life decay data, reinforcement learning algorithms and MPC can predict the battery life decay trend and adjust the charging and discharging strategy to slow down the battery's decline. For example, when the system detects that the internal resistance of the battery increases, MPC can appropriately reduce the charging power to avoid overloading the battery and extend its service life.

[0104] By dynamically adjusting the charging strategy to avoid battery overheating or overcharging and discharging, the system can effectively prevent battery damage or thermal runaway events. Real-time monitoring of battery temperature and health status and timely response based on the prediction model help ensure system safety.

[0105] The entire optimization process is adaptive. Through reinforcement learning, the system can autonomously learn from historical and real-time data to further optimize battery performance. This not only reduces the need for human intervention, but also enables the energy storage system to adapt to changing load and environmental conditions, thereby improving the intelligence level of the system.

[0106] For example, in an energy storage system, assuming that the battery is working under high load and the temperature is too high during the charging process, the digital twin model finds that the temperature exceeds the standard through simulation thermal management data. Based on this data, the reinforcement learning algorithm automatically adjusts the charging strategy to reduce the charging current, and optimizes the battery's temperature control strategy through MPC to start the cooling system in time, thereby avoiding battery overheating and reducing the risk of increased internal resistance. Through these real-time feedback adjustments, the battery charging and discharging efficiency has been improved, and the battery life has been effectively extended.

[0107] Preferably, the reinforcement learning algorithm generates optimization strategy data by the following steps:

[0108] In the simulation environment, a reward function is defined based on the simulation data and the set evaluation objectives;

[0109] Adjust the battery's charge and discharge timing, charge and discharge current, and charge and discharge voltage through multiple rounds of trial and error;

[0110] Optimization strategy data for optimizing battery charging and discharging efficiency and extending battery life is generated based on trial and error learning results.

[0111] The core of the reinforcement learning algorithm is to continuously optimize the control strategy through a "trial and error" learning method. In the present invention, the reinforcement learning algorithm first needs to be run in a simulation environment. The simulation environment is a virtual environment built based on the operating data of the battery cluster (such as voltage, current, internal resistance, etc. during the charging and discharging process) and environmental data (such as temperature, humidity, etc.), which reflects the behavior of the battery under actual working conditions. In the simulation environment, the algorithm adjusts the control variables such as charging and discharging timing, charging and discharging current and voltage according to the performance of the battery under different loads, temperatures, humidity and other conditions.

[0112] In reinforcement learning, the reward function is a key element used to evaluate the effectiveness of a strategy. The reward function gives real-time feedback based on the battery's charge and discharge efficiency, health status (such as temperature, internal resistance, capacity, etc.) and other related indicators. For example, if the battery's internal resistance increases, the battery temperature rises, or the charging efficiency decreases during the charge and discharge process, the value of the reward function will decrease; if the charge and discharge process goes smoothly and the battery performance is maintained at its best, the value of the reward function will increase. By defining this feedback mechanism, reinforcement learning can guide strategy optimization and gradually find the optimal charge and discharge strategy.

[0113] The reinforcement learning algorithm adjusts the charging and discharging strategy through multiple rounds of trial and error. During each trial, the algorithm operates according to the current charging and discharging timing, current and voltage settings, and the battery generates feedback data based on these strategies. By combining with the reward function in the simulation environment, the algorithm can gradually adjust the control parameters to improve the battery charging and discharging efficiency and optimize the battery health status. For example, if the battery temperature rises too quickly during charging, the system will avoid overheating by reducing the charging current or extending the charging time to avoid damage to the battery.

[0114] Through multiple iterations, the reinforcement learning algorithm can continuously improve the control strategy to ensure that the charging and discharging process can maintain high efficiency while reducing the risk of battery life degradation and damage. The algorithm not only relies on short-term charging and discharging efficiency optimization, but also takes into account the long-term health of the battery, thereby avoiding over-discharging or overcharging of the battery and maximizing the battery life.

[0115] Through the trial and error process of reinforcement learning, the algorithm generates optimization strategy data including the optimal charging time, charging current, charging voltage and other parameters. The generation of these strategy data is based on the real-time feedback of battery performance during the reinforcement learning process, and after multiple optimization iterations, the charging and discharging scheme that best suits the current battery health status and environmental conditions is obtained. For example, the algorithm may adjust the upper limit of the charging current to reduce the risk of temperature rise, or adjust the charging time interval to avoid the battery from overheating during the charging process. The generated optimization strategy data is applied to the battery management system in real time to guide the battery to always maintain the optimal state during different charging and discharging processes.

[0116] An important feature of reinforcement learning is its adaptive ability, that is, it can adjust the optimization strategy based on real-time feedback. Whenever the battery's operating state changes, the system will dynamically adjust the charge and discharge strategy based on the battery's current voltage, internal resistance, temperature and other data. For example, when the battery temperature is too high, the optimization strategy data can automatically adjust the charging current to prevent the battery from overheating. When the battery's internal resistance increases, the system may adjust the charging voltage or discharge rate to prevent the battery from further deteriorating due to excessive load. This real-time feedback adjustment mechanism ensures the stability and efficiency of the battery throughout its life cycle.

[0117] Through reinforcement learning optimization, the battery can use energy more efficiently during the charging and discharging process, avoid energy waste, and ensure maximum efficiency of each charging and discharging process. For example, in a low temperature environment, the battery charging efficiency is usually low. The reinforcement learning algorithm can recognize this and automatically adjust the charging current or voltage to improve the charging efficiency.

[0118] Reinforcement learning not only focuses on short-term efficiency, but also optimizes based on battery life decay data (such as internal resistance increase, capacity decay, etc.). During the battery life decay process, the charge and discharge strategy will automatically adjust to reduce damage to the battery, thereby extending the battery life. For example, as the battery usage time increases, the system will reduce the charging current to slow down the increase in internal resistance.

[0119] By monitoring the battery status in real time, the reinforcement learning algorithm can adaptively adjust the charging strategy to avoid problems such as overcharging, over-discharging or overheating, thereby effectively protecting the health of the battery. Especially under high temperature or extreme load conditions, the reinforcement learning algorithm can dynamically adjust the timing of charging and discharging to ensure that the battery is always in the best working condition.

[0120] For example, suppose the battery cluster in the energy storage system is in a high temperature environment and the current load is heavy. In this case, the temperature of the battery may rise rapidly, affecting the charging efficiency. Through the reinforcement learning algorithm, the system first identifies the impact of high temperature on the charging process based on historical data and real-time temperature feedback. Based on this information, the algorithm adjusts the charging current, reduces the battery load, and adjusts the charging time to avoid overheating of the battery. After a round of optimization, the system can automatically adjust the charging and discharging strategy in real time according to the changes in battery temperature and load to ensure that the battery always maintains optimal performance in different environments.

[0121] Preferably, the model predictive control comprises:

[0122] Based on real-time collected battery cluster data, simulated charge and discharge data, simulated thermal management data, and simulated battery life attenuation data, the future battery charge and discharge status is predicted; the battery charging power and discharge rate are dynamically adjusted to reduce battery energy loss and extend battery life; and the battery cluster is ensured to operate in the optimal state under different load and environmental conditions.

[0123] In the present invention, model predictive control relies on real-time collected battery cluster data (such as voltage, current, internal resistance, etc.) and simulation data (including charge and discharge data, thermal management data and battery life decay data). Real-time data provides the current working status of the battery cluster, while simulation data predicts the possible state of the battery during future charge and discharge based on models under different environmental and load conditions. By combining these two types of data, MPC can predict the electrical state, temperature changes, internal resistance growth and other factors that may affect battery performance of the battery cluster in the future.

[0124] During the prediction process, MPC simulates the future battery status through the model based on the input real-time data and simulation data of the battery cluster. This process is usually based on the battery's electrical model, thermal management model, and life decay model to predict the battery's state in the next time step. For example, the system predicts the battery's voltage change, temperature fluctuation, and internal resistance increase over a period of time in the future based on the current battery voltage, current, and temperature data combined with historical simulation data.

[0125] By predicting the future state of the battery cluster, MPC can dynamically adjust the battery's charging power and discharge rate. The adjustment of charging power and discharge rate is based on the system's prediction of the battery state and the response to anomalies or risks that occur during the current charging and discharging process. Temperature changes, internal resistance changes, and loss of charging efficiency during battery charging may affect the battery's health, especially when the battery life is attenuated or under high load conditions, the charging process may cause overheating or reduced efficiency.

[0126] For example, if it is predicted that the battery temperature will rise during the next charging cycle, MPC can adjust the charging power in advance, slow down the charging rate, and reduce the risk of temperature rise. Similarly, if the internal resistance of the battery increases, MPC will reduce the internal energy loss of the battery by reducing the charging current or discharging current, reducing the risk of further increase in internal resistance.

[0127] The goal of MPC is to ensure that the battery cluster can operate in an optimal state under various load and environmental conditions, which means that the battery should always maintain the best charge and discharge efficiency and avoid battery degradation or damage caused by factors such as overcharging, over-discharging or excessive temperature. Under different load conditions, the charge and discharge efficiency of the battery may be significantly affected. For example, under high load conditions, the battery is prone to overheating, and the increase in temperature will cause the internal resistance to increase, affecting the battery performance. Therefore, MPC dynamically adjusts the charging current, charging voltage and charging time according to the predicted battery status and thermal management data to ensure that the battery is always in a safe, stable and efficient working state.

[0128] When environmental conditions change, such as temperature rise or humidity changes, MPC also adjusts the charging and discharging strategies based on the predictions of the battery cluster and environmental data. For example, when the temperature rises, MPC can reduce the charging power or reduce the discharge rate to avoid overheating of the battery. In this way, MPC can ensure that the battery can achieve the best energy efficiency output under different environmental and load conditions and extend the life of the battery.

[0129] Preferably, the evaluation step generates evaluation results based on the final optimized control data, and the evaluation results include the battery's charge and discharge efficiency, internal resistance change, temperature fluctuation, and remaining service life, and provides decision support for operation optimization and maintenance of the energy storage system through a comprehensive evaluation report.

[0130] In the present invention, the final optimized control data is generated by model predictive control and reinforcement learning algorithms, which reflects the optimal strategy adopted by the system during the charging and discharging process. The optimized control data covers multiple control variables such as the battery's charging current, voltage, and charging timing, which have a direct impact on the battery performance during the charging and discharging process. These data provide a basis for the evaluation of battery performance. By combining with real-time collected data, the performance of the battery under different working conditions can be more accurately evaluated.

[0131] The final optimized control data not only reflects the adjustments made by the system based on the current state of the battery and the predicted results, but also reflects the system's long-term strategy for battery health management. Through these data, we can fully understand the performance of the battery under different loads, environmental conditions and working hours, ensuring that the battery always operates in the best state.

[0132] The charge and discharge efficiency is one of the core indicators for evaluating battery performance. It directly reflects the energy utilization efficiency of the battery during the charging and discharging process. Based on the final optimized control data, the system can calculate the battery's charge and discharge efficiency, and evaluate the battery's performance under different conditions by comparing the input and output energy during the actual charge and discharge process.

[0133] For example, if the battery's charging and discharging efficiency is low, this may be due to energy loss caused by increased internal resistance, battery life decay, or excessive temperature. The system will evaluate the efficiency of the battery in this working state based on the charging current, voltage and other parameters in the optimization control data, combined with real-time data, to promptly discover potential causes of performance degradation and provide a basis for subsequent optimization.

[0134] The change in the internal resistance of the battery is an important indicator of the battery's health. The increase in internal resistance will lead to a decrease in battery efficiency, increased heat generation, and may even affect the safety of the battery. Based on the optimized control data, the system can monitor the change in the battery's internal resistance in real time and evaluate it in combination with the battery's operating status.

[0135] The internal resistance of the battery usually increases with the increase of usage time, especially when there are many charge and discharge cycles, the trend of internal resistance increase is more obvious. By optimizing the control data, the system can track the changes in the internal resistance of the battery and predict the future change trend of the internal resistance through simulation data. For example, in the simulated charge and discharge data, if the system detects a trend of increasing internal resistance, it may recommend adjusting the charging strategy or changing the charge and discharge rate to slow down the increase of internal resistance and extend the battery life.

[0136] Battery temperature fluctuations directly affect battery performance and safety, especially under high load or high temperature environments, where batteries are prone to overheating and performance degradation. Based on optimized control data and real-time monitored temperature data, the system can evaluate battery temperature fluctuations and determine whether there is an overheating risk.

[0137] Temperature fluctuations not only affect the battery's charging and discharging efficiency, but may also accelerate the battery's life decay process. By analyzing and optimizing control data, the system can adjust the battery's temperature control strategy in real time during the charging process (such as adjusting the charging current, extending the charging time, etc.) to ensure that the battery is always within a safe temperature range. For example, when the battery temperature exceeds the set threshold, the optimized control data will adjust the charging strategy, slow down the charging rate, and reduce the temperature rise, thereby avoiding the risk of overheating.

[0138] The remaining service life of a battery is an important evaluation indicator in the battery management of an energy storage system. It is directly related to the battery replacement cycle and the operating cost of the system. Based on data such as the battery's internal resistance change, charge and discharge efficiency, and temperature fluctuations, the system can predict the remaining service life of the battery.

[0139] In the present invention, based on the final optimized control data and historical simulation data, the system simulates the performance changes of the battery in the future by establishing a battery life decay model. For example, if the system predicts that the battery internal resistance will increase in the future and cause a significant decrease in charging and discharging efficiency, the system can prompt in advance that the battery needs to be replaced or maintained, thereby avoiding battery performance decay or failure.

[0140] In addition, the system can also predict the impact of environmental changes (such as temperature fluctuations) on battery life, combine with optimization control strategies, adjust charging and discharging behaviors, slow down the battery degradation process, and extend the actual service life of the battery.

[0141] Ultimately, all evaluation results (charge and discharge efficiency, internal resistance change, temperature fluctuation and remaining service life) will be output in the form of a comprehensive evaluation report. This report not only provides comprehensive feedback on the health of the battery, but also provides an important basis for the operation optimization and maintenance decision of the energy storage system. For example, the report can recommend measures such as optimizing the charge and discharge strategy, adjusting the operating parameters of the battery management system, and regularly checking the battery health, thereby improving the efficiency of the battery and extending its service life.

[0142] Through the evaluation report, the operation and maintenance team of the energy storage system can obtain specific operational suggestions, such as whether the battery needs to be replaced, whether the charging and discharging strategies need to be adjusted, etc., so as to achieve more intelligent and refined battery management and reduce unnecessary maintenance and operating costs.

[0143] For example, suppose that during one operation, the system finds that the battery temperature is slightly high and the internal resistance increases rapidly through optimizing control data and real-time feedback. Based on these evaluation results, the system generates a comprehensive evaluation report, recommending immediate adjustment of the charging current and activation of the cooling system. The report also predicts the growth trend of the battery's internal resistance in the next few months, and recommends arranging regular inspections and optimizing the charging and discharging strategy. Through these measures, the system can effectively delay the battery life decay process, extend the battery life, and ensure the safe operation of the system.

[0144] Preferably, the evaluation results are used to adjust the operation strategy of the energy storage system, including optimizing the charge and discharge time of the battery, adjusting the charge and discharge parameters, and setting a temperature control strategy that adapts to different environmental conditions.

[0145] Optimizing the charge and discharge time is an important aspect of battery management, especially in energy storage systems, where the charge and discharge time directly affects the energy utilization of the system. By evaluating the battery's charge and discharge efficiency, internal resistance changes, and temperature fluctuations, the system can determine whether the battery's charge and discharge time is optimal under different operating conditions. For example, during the charging process, if the evaluation results show that the battery temperature is too high, the system can slow down the charging rate by adjusting the charging time to avoid overheating the battery, while ensuring that the charging process does not affect the overall efficiency of the battery.

[0146] In addition, the optimization of charging and discharging time can also be adjusted by predicting the remaining life of the battery and adjusting the cycle and frequency of charging and discharging. The system can extend or shorten the charging and discharging time according to the health of the battery to avoid the battery being in a state of over-discharge or over-charge, reduce the battery life decay rate, and thus extend the battery life.

[0147] Adjusting the charging and discharging parameters includes key control variables such as charging current, voltage, and discharge rate. The battery internal resistance and charging and discharging efficiency in the evaluation results provide a clear basis for adjusting the charging and discharging parameters. Internal resistance is an important indicator of battery health, which directly affects the battery's charging and discharging efficiency and thermal management requirements. When the battery's internal resistance increases, the increase in charging current will lead to higher energy loss and temperature rise, which in turn affects the health of the battery. Therefore, the system will adjust the charging current and voltage in a timely manner according to the evaluation results to optimize the charging and discharging parameters.

[0148] For example, when the system evaluates that the internal resistance of the battery increases or the battery temperature rises, the system will reduce the charging current in a timely manner to slow down the battery charging process to avoid excessive current from damaging the battery. Similarly, if the battery's charging and discharging efficiency decreases, the system will also adjust the charging voltage to ensure that the battery can operate within a reasonable voltage range, maximize charging efficiency and avoid overcharging or over-discharging the battery.

[0149] Environmental conditions, especially changes in temperature and humidity, directly affect the charging and discharging efficiency and service life of the battery. The temperature fluctuation data and thermal management data in the evaluation results provide a basis for adjusting the temperature control strategy. If the battery is operated at too high or too low a temperature, the battery performance will be significantly reduced, and even cause safety problems. Therefore, the system needs to adjust the temperature control strategy in real time according to the evaluation results to ensure that the battery always operates within the appropriate temperature range.

[0150] For example, when the ambient temperature is high, the battery's thermal management system needs to enhance its heat dissipation capabilities. The system will reduce the battery temperature and prevent overheating by optimizing the charging time and reducing the charging power. In low-temperature environments, the battery's charging and discharging efficiency is low. The system may compensate for the efficiency loss by increasing the charging current or extending the charging time, while using a heating system to increase the battery temperature to ensure that the battery can be charged effectively. Through the dynamic adjustment of these temperature control strategies, the system can ensure that the battery maintains optimal performance under various environmental conditions.

[0151] For example, suppose an energy storage system is in a high temperature environment in summer, and the system detects that the battery temperature is gradually rising and the internal resistance is increasing. Based on the evaluation results, the system decides to reduce temperature fluctuations by optimizing the charging time and reducing the charging current, and starts the temperature control system to enhance heat dissipation. At the same time, the system also adjusts the charging voltage to ensure that the battery charging efficiency is maximized. In this process, the system adjusts the temperature control strategy to avoid overheating of the battery due to high temperature, and effectively slows down the rate of increase of the battery internal resistance, thereby extending the battery life.

[0152] like Figure 5 As shown, a system for implementing the energy storage system electrical performance evaluation method, the system comprising:

[0153] The data acquisition module is used to collect the electrical data and environmental data of the battery cluster and transmit the collected data to the data processing module; the main function of the data acquisition module is to collect the electrical data and environmental data of the battery cluster in the energy storage system. These data include the electrical characteristics of the battery such as voltage, current, temperature, internal resistance, etc., as well as the impact of environmental factors such as temperature, humidity, gas leakage, etc. on battery performance. The collected data is transmitted to the data processing module in real time to provide raw data support for subsequent processing and analysis.

[0154] The data processing module is used to clean and remove noise from the collected battery cluster data and environmental data, generate processed battery cluster data and environmental data, and transmit the processed data to the feature extraction module; the data processing module is responsible for cleaning, denoising and standardizing the collected battery cluster electrical data and environmental data. Specifically, through noise removal algorithms such as Kalman filtering and sliding average filtering, the current and voltage signals in the battery data are smoothed to avoid abnormal data from affecting subsequent steps. In addition, the environmental data is time-aligned to ensure data consistency and accuracy.

[0155] The feature extraction module is used to extract battery feature data and environmental feature data based on the processed battery cluster data and environmental data, and transmit the extracted feature data to the digital twin model module; the feature extraction module extracts the battery feature data and environmental feature data by analyzing the processed battery cluster data and environmental data. The battery feature data includes the battery's state of charge, internal resistance, temperature change rate, etc., while the environmental feature data considers the impact of environmental temperature and humidity, gas leakage, etc. on battery performance. The extracted feature data is standardized and transmitted to the digital twin model module for further modeling.

[0156] The digital twin model module is used to establish a digital twin model based on the extracted battery characteristic data and environmental characteristic data, perform virtual simulation, simulate the charging and discharging process of the battery under different load and environmental conditions, generate simulated charging and discharging data, simulated thermal management data and simulated battery life attenuation data, and transmit the simulation data to the optimization control module; the digital twin model module establishes a virtual model of the battery by inputting the extracted battery characteristic data and environmental characteristic data, and simulates the charging and discharging process of the battery under different load and environmental conditions. The model generates charging and discharging data, thermal management data and battery life attenuation data by simulating the charging and discharging process of the battery. These data are obtained through multi-physics field simulation (such as electrothermal coupling model) and are used to characterize the health status, efficiency changes and life attenuation trends of the battery.

[0157] The optimization control module is used to generate optimization strategy data based on the simulated charge and discharge data, simulated thermal management data and simulated battery life decay data by applying the reinforcement learning algorithm, and to adjust the battery charge and discharge in real time by using the model predictive control, adjust the optimization strategy based on the real-time feedback, generate the final optimization control data and transmit the results to the evaluation module; the optimization control module generates the optimization strategy based on the simulated charge and discharge data, thermal management data and battery life decay data by applying the reinforcement learning algorithm. The reinforcement learning algorithm continuously adjusts the learning strategy through interaction with the environment (such as feedback on the charge and discharge effect) to achieve the optimal battery charge and discharge plan. In addition, the model predictive control (MPC) technology is used to adjust the battery charge and discharge power and discharge rate in real time, and adjust the strategy based on the real-time feedback to maintain the optimal state of battery operation.

[0158] The evaluation module is used to evaluate the electrical performance of the energy storage system battery based on the final optimized control data and generate evaluation results. The evaluation module evaluates the electrical performance of the energy storage system battery based on the final optimized control data, including charging and discharging efficiency, internal resistance change, temperature fluctuation, and remaining service life. This module combines all simulation data, real-time data, and optimized control results to generate a comprehensive evaluation report on the battery health status, providing decision support for system operation optimization and maintenance.

[0159] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware.

[0160] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for evaluating the electrical performance of an energy storage system, characterized in that: The following steps are involved: Collecting electrical data and environmental data of the battery cluster, and cleaning and removing noise from the collected data to obtain battery cluster data and environmental data; extracting battery characteristic data and environmental characteristic data based on the battery cluster data and environmental data; Based on the battery characteristic data and environmental characteristic data, a digital twin model is established to perform virtual simulation to simulate the battery charging and discharging process under different load and environmental conditions, and obtain simulated charging and discharging data, thermal management data, and battery life attenuation data; Based on the simulated charging and discharging data, thermal management data and battery life attenuation data, the reinforcement learning algorithm is applied to generate optimization strategy data, the battery charging and discharging is adjusted in real time using model predictive control, and the optimization strategy is adjusted based on real-time feedback to obtain the final optimization control data; based on the final optimization control data, the electrical performance of the energy storage system battery is evaluated to obtain the evaluation results.

2. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The data cleaning and noise removal steps include removing noise from the voltage and current signals in the battery cluster data through Kalman filtering, smoothing the temperature and humidity signals in the environmental data using a sliding mean filter, and detecting outliers in the gas leakage data by setting a threshold.

3. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: Battery characteristic data includes the battery's state of charge, temperature change rate, and internal resistance, among which: The battery status charge is calculated by collecting the voltage and current data of the battery cluster in combination with a preset battery capacity model; the battery temperature change rate is obtained by real-time monitoring of the battery temperature and calculating the change per unit time; the battery internal resistance is calculated by the ratio of the voltage change to the current change measured during the battery charging and discharging process.

4. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The environmental characteristic data includes the impact of environmental temperature and humidity on battery performance. The environmental characteristic data is obtained by combining real-time monitored environmental temperature and humidity data with a battery adaptability analysis model to evaluate the impact of environmental changes on battery charging and discharging efficiency and life.

5. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The virtual simulation steps include: By building a digital twin model based on the battery's electrical characteristics, thermal management characteristics, and life decay characteristics, combined with the real-time data and environmental data of the battery cluster, a virtual simulation of the charging and discharging process is performed; The simulated charge and discharge data generated by the simulation process includes information reflecting the changes in battery voltage and current; Simulation thermal management data includes the temperature distribution and heat dissipation characteristics of the battery during operation; The simulated battery life decay data includes information reflecting the battery's internal resistance growth and capacity decay.

6. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The reinforcement learning algorithm generates optimized policy data through the following steps: In the simulation environment, a reward function is defined based on the simulation data and the set evaluation objectives; Adjust the battery's charge and discharge timing, charge and discharge current, and charge and discharge voltage through multiple rounds of trial and error; Optimization strategy data for optimizing battery charging and discharging efficiency and extending battery life is generated based on trial and error learning results.

7. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: Model predictive control includes: Based on real-time collected battery cluster data, simulated charge and discharge data, simulated thermal management data, and simulated battery life attenuation data, the future battery charge and discharge status is predicted; the battery charging power and discharge rate are dynamically adjusted to reduce battery energy loss and extend battery life; and the battery cluster is ensured to operate in the optimal state under different load and environmental conditions.

8. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The evaluation step generates evaluation results based on the final optimized control data. The evaluation results include the battery's charge and discharge efficiency, internal resistance change, temperature fluctuation, and remaining service life. A comprehensive evaluation report is used to provide decision support for the operation optimization and maintenance of the energy storage system.

9. The method for evaluating the electrical performance of an energy storage system according to claim 1, characterized in that: The evaluation results are used to adjust the operating strategy of the energy storage system, including optimizing the battery’s charge and discharge time, adjusting the charge and discharge parameters, and setting a temperature control strategy to adapt to different environmental conditions.

10. A system for implementing the method for evaluating the electrical performance of an energy storage system according to any one of claims 1 to 9, characterized in that: The system comprises: A data acquisition module, used for collecting electrical data and environmental data of the battery cluster and transmitting the collected data to the data processing module; A data processing module, used for cleaning and removing noise from the collected battery cluster data and environmental data, generating processed battery cluster data and environmental data, and transmitting the processed data to the feature extraction module; A feature extraction module, used for extracting battery feature data and environmental feature data based on the processed battery cluster data and environmental data, and transmitting the extracted feature data to the digital twin model module; A digital twin model module is used to establish a digital twin model based on the extracted battery characteristic data and environmental characteristic data, perform virtual simulation, simulate the charging and discharging process of the battery under different load and environmental conditions, generate simulated charging and discharging data, simulated thermal management data and simulated battery life attenuation data, and transmit the simulation data to the optimization control module; An optimization control module is used to generate optimization strategy data based on simulated charge and discharge data, simulated thermal management data, and simulated battery life decay data by applying a reinforcement learning algorithm, and to adjust battery charge and discharge in real time by using model predictive control, adjust the optimization strategy based on real-time feedback, generate final optimization control data, and transmit the results to an evaluation module; The evaluation module is used to evaluate the electrical performance of the battery of the energy storage system based on the final optimized control data and generate an evaluation result.

Images