Full life cycle management method and system of energy storage power supply based on digital twinning

Abstract

The invention provides a full life cycle management method and system for an energy storage power supply based on digital twinning, and relates to the technical field of life cycle management. A digital twinning model is established for the energy storage power supply, periodic acquisition data is acquired, data synchronization is performed, and a power supply monitoring model is acquired; according to the power supply monitoring model, different working condition state prediction and periodic state prediction are carried out to obtain different working condition state data and periodic state prediction data; and next-period collected data change analysis of the next period is carried out on the energy storage power supply, periodic change analysis data are obtained, periodic change abnormal factor information is determined, abnormal factor early warning and adjustment are carried out until full-life-cycle data are obtained, and linkage intelligent life cycle analysis and management of the energy storage power supply are achieved.

Description

Technical Field

[0001] This invention proposes a method and system for full lifecycle management of energy storage power sources based on digital twins, which relates to the field of lifecycle management technology, specifically to the field of full lifecycle management technology of energy storage power sources based on digital twins. Background Technology

[0002] Currently, energy storage power supplies are playing an increasingly crucial role in new power systems, but traditional management models have significant drawbacks. The impact of operating conditions is difficult to quantify. Energy storage power supply operation is affected by multiple factors, including temperature, humidity, load, and grid interaction. Traditional methods cannot accurately identify the positive or negative effects of single or combined operating conditions on performance, and improper adaptation to operating conditions can accelerate cell degradation. Cyclic degradation is difficult to predict. The post-event maintenance model, relying on periodic inspections, cannot predict changes in the state of the next cycle in advance, often leading to sudden failures. Abnormal factors are difficult to locate. When multiple factors coexist, the root cause is easily confused, and adjustment strategies rely on experience, resulting in high trial-and-error costs and low efficiency. Summary of the Invention

[0003] This invention provides a method and system for full lifecycle management of energy storage power sources based on digital twins, in order to solve the above-mentioned problems:

[0004] The present invention proposes a method and system for full lifecycle management of energy storage power sources based on digital twins, wherein the method includes:

[0005] S1. Establish a digital twin model for the energy storage power supply, acquire periodic acquisition data, synchronize the data, acquire the power supply monitoring model, predict different operating conditions and periodic conditions based on the power supply monitoring model, acquire multiple operating condition prediction data and multiple combinations of operating condition data for different operating conditions, conduct operating condition type impact analysis, obtain operating condition type impact data, acquire periodic acquisition state prediction data and periodic adjustment state prediction data for periodic conditions, and then obtain different operating condition data and periodic state prediction data.

[0006] S2. Analyze the changes in the data collected from the energy storage power source for the next cycle, obtain cycle change analysis data, identify abnormal factors in cycle changes, and carry out early warning and adjustment for abnormal factors until the full life cycle data is obtained.

[0007] Further, S1 includes:

[0008] Obtain the physical attribute information of the energy storage power source, and establish a digital twin model based on the physical attribute information;

[0009] The energy storage power supply is subjected to periodic data collection throughout its entire life cycle based on physical property information using a sensor array, and periodic data collection is obtained.

[0010] The periodically collected data is synchronously transmitted to the digital twin model to obtain the power monitoring model;

[0011] By combining the power monitoring model with physical attribute information, prediction data for different operating conditions is obtained.

[0012] Periodic state prediction data is obtained by combining power monitoring models with periodic data acquisition.

[0013] Furthermore, the step of predicting different operating conditions by combining a power supply monitoring model with physical attribute information to obtain prediction data for different operating conditions includes:

[0014] Acquire multiple preset working condition data, and generate multiple working condition simulation data based on the multiple preset working condition data and the physical attribute information;

[0015] By using a power monitoring model to predict the operation of energy storage power sources based on simulation data of various operating conditions, prediction data for various operating conditions can be obtained.

[0016] Multiple preset working condition data are arranged and combined to obtain multiple combined working condition data;

[0017] Based on the combined working condition data and the physical attribute information, various working condition combination simulation data are generated.

[0018] The operating performance of energy storage power sources is predicted by using power monitoring models to simulate data of various operating conditions, and prediction data of various operating conditions are obtained.

[0019] Based on the prediction data of various working conditions and the prediction data of various working condition combinations, an analysis of the impact of working condition types is conducted to obtain the impact data of working condition types.

[0020] Furthermore, the step of performing an impact analysis on different operating conditions based on the predicted data of multiple operating conditions combined with the predicted data of multiple operating condition combinations to obtain the impact data of different operating conditions includes:

[0021] The predicted data for each working condition combination is compared with the corresponding simulated data for the working condition combination to obtain the working condition combination comparison results.

[0022] When the predicted data of the working condition combination is greater than the simulated data of the corresponding working condition, the negative impact working condition type is determined by the simulated data of the corresponding working condition to obtain the negative impact working condition type.

[0023] When the predicted data of the working condition combination is less than or equal to the corresponding working condition simulation data, the working condition type that is positively affected by the corresponding working condition simulation data is determined, and the working condition type that is positively affected is obtained.

[0024] The negative and positive operating condition types mentioned above are the operating condition type impact data.

[0025] Furthermore, the step of obtaining periodic state prediction data by combining the power supply monitoring model with periodically collected data includes:

[0026] Periodic status data is obtained by combining the periodic collected data with physical attribute information;

[0027] Based on the periodic acquisition data and the periodic status data, the periodic status data of the next cycle in the whole life cycle is predicted to obtain the periodic acquisition status prediction data.

[0028] The periodic acquisition data is adjusted based on the impact data of the aforementioned operating conditions to obtain periodic adjustment data;

[0029] Based on the cycle adjustment data and cycle state data, the cycle state data of the next cycle in the whole life cycle is predicted to obtain the cycle adjustment state prediction data.

[0030] The periodic acquisition state prediction data and the periodic adjustment state prediction data are the periodic state prediction data.

[0031] Further, S2 includes:

[0032] The energy storage power supply will be subjected to periodic data collection for the next cycle of its entire life cycle to obtain the data collected in the next cycle;

[0033] The cycle state prediction data is linked with the data collected in the next cycle for analysis to obtain cycle change analysis data.

[0034] Based on the periodic change analysis data, information on abnormal factors in periodic changes is determined;

[0035] Based on the information on abnormal factors in the cycle change, early warning and adjustment of the negative impact of factors are carried out until the full life cycle data is obtained.

[0036] Furthermore, the predicted periodic state data is linked with the data collected in the next period for analysis to obtain periodic change analysis data, including:

[0037] The ratio of the periodic acquisition state prediction data to the next period acquisition data is used to obtain the prediction period change coefficient.

[0038] The predicted cycle change coefficient is compared with the predicted cycle change threshold to obtain the predicted cycle comparison result;

[0039] Based on the comparison results of the prediction cycle, the development direction of the data collected in the next cycle is determined, and positive development judgment and negative development judgment are obtained.

[0040] When the development direction of the data collected in the next cycle is determined to be negative, the ratio of the predicted periodic adjustment state data to the data collected in the next cycle is used to obtain the adjustment cycle change coefficient.

[0041] The variation data of the adjustment cycle variation coefficient is analyzed to obtain cycle variation analysis data.

[0042] Furthermore, based on the aforementioned periodic variation analysis data, information on anomalous factors in periodic variations is determined, including:

[0043] Obtain a preset adjustment cycle change threshold, calculate the difference between the adjustment cycle change coefficient and the preset adjustment cycle change threshold, and obtain adjustment change difference data;

[0044] A preset adjustment change difference threshold is obtained, and the adjustment change difference data is compared with the preset adjustment change difference threshold to obtain the adjustment change difference comparison result.

[0045] Based on the comparison results of the adjustment change difference, the information on abnormal factors of periodic change is determined according to the operating condition type influence data.

[0046] Furthermore, based on the information on abnormal periodic changes, early warning and adjustment of negative impacts are performed until full life-cycle data is obtained, including:

[0047] Obtain information on the types of negatively impacted working conditions corresponding to abnormal factors in periodic changes;

[0048] Early warning and corresponding adjustments are made for the types of negative impact operating conditions of energy storage power sources, and negative impact adjustment data is obtained;

[0049] Based on the negative impact adjustment data, the next cycle data of the energy storage power supply is obtained, the cycle state prediction data is obtained, and so on until the full life cycle data is obtained.

[0050] Furthermore, the system includes:

[0051] The predictive analysis module is used to establish a digital twin model of the energy storage power supply, acquire periodic acquisition data, synchronize data, acquire the power supply monitoring model, and perform predictions of different operating conditions and periodic states based on the power supply monitoring model. It acquires various operating condition prediction data and various combinations of operating condition data for different operating condition predictions, performs operating condition type impact analysis, obtains operating condition type impact data, acquires periodic acquisition state prediction data and periodic adjustment state prediction data for periodic state prediction, and then obtains different operating condition state data and periodic state prediction data.

[0052] The cycle variation analysis module is used to analyze the changes in the data collected from the energy storage power source for the next cycle, obtain cycle variation analysis data, identify abnormal factors in cycle variation, and perform early warning and adjustment for abnormal factors until the full life cycle data is obtained.

[0053] The beneficial effects of this invention are as follows: This invention breaks away from the traditional post-event maintenance model, using dual-dimensional prediction to proactively understand the impact of operating conditions and cyclical change trends, thus achieving a transformation from passive response to proactive management. From initial model construction to final full lifecycle data generation, the process runs through the entire operation of the energy storage power source, avoiding performance loss or safety risks caused by management gaps. The linkage between periodically collected data and predicted data not only supports current cycle optimization but also provides data references for the management of similar energy storage power sources, forming a reusable management experience library. Attached Figure Description

[0054] Figure 1 A schematic diagram illustrating a digital twin-based full lifecycle management method for energy storage power supplies;

[0055] Figure 2 This is a schematic diagram of the predicted operating conditions.

[0056] Figure 3 This is a schematic diagram for analyzing periodic changes. Detailed Implementation

[0057] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0058] In one embodiment of the present invention, the present invention proposes a method and system for full lifecycle management of energy storage power sources based on digital twins, the method comprising:

[0059] S1. Establish a digital twin model for the energy storage power supply, acquire periodic acquisition data, synchronize the data, acquire the power supply monitoring model, predict different operating conditions and periodic conditions based on the power supply monitoring model, acquire multiple operating condition prediction data and multiple combinations of operating condition data for different operating conditions, conduct operating condition type impact analysis, obtain operating condition type impact data, acquire periodic acquisition state prediction data and periodic adjustment state prediction data for periodic conditions, and then obtain different operating condition data and periodic state prediction data.

[0060] S2. Analyze the data changes of the energy storage power supply for the next cycle, obtain cycle change analysis data, identify abnormal factors in cycle changes, and implement early warning and adjustment for abnormal factors until full life cycle data is obtained. Figure 1 As shown.

[0061] The working principle and technical effects of the above solution are as follows: Using a digital twin model as the core carrier, the model accurately maps the physical state of the energy storage power supply through real-time synchronization of periodically collected data (such as voltage, temperature, and load), forming a power supply monitoring model with simulation and prediction capabilities. For different single-factor operating conditions (such as temperature and load) and combined operating conditions, the model simulates the operating performance of the energy storage power supply and analyzes the impact of various operating conditions on power supply performance. Based on historical periodic data, the model predicts the natural operating state of the next period, and simultaneously adjusts parameters based on the operating condition impact data to predict the adjusted periodic state. The predicted data is compared with the actual collected data of the next period to identify and adjust abnormal factors. Through a cycle of prediction, verification, and optimization, the entire lifecycle of the energy storage power supply, from commissioning to decommissioning, is covered.

[0062] This invention breaks away from the traditional reactive maintenance model, using dual-dimensional prediction to proactively understand the impact of operating conditions and cyclical change trends, thus transforming from passive response to proactive management. From initial model building to final full lifecycle data generation, the process spans the entire operation of the energy storage power source, avoiding performance losses or safety risks caused by management gaps. The linkage between periodically collected and predicted data not only supports current cycle optimization but also provides data references for the management of similar energy storage power sources, forming a reusable management experience library.

[0063] In one embodiment of the present invention, S1 includes:

[0064] Obtain the physical attribute information of the energy storage power source, and establish a digital twin model based on the physical attribute information;

[0065] The energy storage power supply is subjected to periodic data collection throughout its entire life cycle based on physical property information through a sensor array, which includes temperature and humidity sensors, load sensors, and interactive sensors.

[0066] The periodically collected data is synchronously transmitted to the digital twin model to obtain the power monitoring model; the power monitoring model is used to perform simulation and predictive data analysis on the energy storage power source.

[0067] By combining a power supply monitoring model with physical attribute information, predictions of different operating conditions are made, resulting in prediction data for various operating conditions, such as... Figure 2 As shown; different operating conditions include operating conditions with multiple individual factors.

[0068] Cyclic state prediction data is obtained by combining power supply monitoring models with periodically collected data. The cycle state prediction provides information about the next cycle from the current cycle.

[0069] The working principle and technical effects of the above solution are as follows: Physical attribute information forms the basic framework of the digital twin model, encompassing the inherent attributes of the power supply, such as its hardware structure and electrochemical characteristics, ensuring consistency between the model and the physical entity in terms of structure and performance benchmarks. The sensor array covers three dimensions: environment, load, and grid interaction. Data collected includes temperature and humidity (environmental factors), load power (operating load), and grid voltage or frequency (interaction parameters), ensuring that periodically collected data fully reflects the power supply's operating environment and status. Periodically collected data updates the parameters of the digital twin model in real time (such as the current temperature of the battery cells and remaining capacity), ensuring that the power supply monitoring model always remains consistent with the actual state of the physical power supply, avoiding prediction deviations caused by model lag.

[0070] Models built upon physical property information inherently possess high fidelity. Combined with real-time, periodic data updates, this significantly reduces the error rate of prediction results (compared to static models). Clearly defining sensor types and data acquisition dimensions avoids management loopholes caused by missing data or inconsistent dimensions, providing a unified data foundation for prediction and analysis. Building models using physical properties and real-time data eliminates the need for additional complex hardware, achieving high model availability while controlling costs.

[0071] In one embodiment of the present invention, the step of predicting different operating conditions by combining a power supply monitoring model with physical attribute information to obtain prediction data for different operating conditions includes:

[0072] Multiple preset operating condition data are acquired, and multiple operating condition simulation data are generated based on the multiple preset operating condition data and the physical attribute information. The multiple preset operating condition data include temperature operating condition data, load operating condition data, humidity operating condition data and power grid interaction operating condition data, etc. The physical attribute information mainly includes fixed information carried at the time of manufacture.

[0073] By using a power monitoring model to predict the operation of energy storage power sources based on simulation data of various operating conditions, prediction data for various operating conditions can be obtained.

[0074] Multiple preset working condition data are arranged and combined to obtain multiple combined working condition data;

[0075] Based on the combined working condition data and the physical attribute information, various working condition combination simulation data are generated.

[0076] The operating performance of energy storage power sources is predicted by using power monitoring models to simulate data of various operating conditions, and prediction data of various operating conditions are obtained.

[0077] Based on the prediction data of various working conditions and the prediction data of various working condition combinations, an analysis of the impact of working condition types is conducted to obtain the impact data of working condition types.

[0078] The working principle and technical effects of the above-mentioned technical solution are as follows: The preset operating conditions cover three key influencing factors: environment, load, and power grid. This includes both single factors (such as temperature changes alone) and multiple factors combined (such as coordinated changes in temperature and load), covering most scenarios that energy storage power supplies may encounter in actual operation. Based on physical property information (such as the temperature tolerance range of a certain lithium iron phosphate battery), the preset operating condition data is transformed into parameters recognizable by the model (such as the ambient temperature parameter from 25℃ to the model's temperature range). By simulating the electrochemical reactions and thermal management efficiency of the power supply under these operating conditions, predicted data (such as charge / discharge efficiency and capacity decay rate) are output. By permuting and combining these factors, the potential impact of multiple factors combined is exposed (such as small capacity decay at high temperatures alone, and large decay at high temperatures and high loads), avoiding the limitations of single-condition analysis. The specific impact of different operating conditions (especially combined operating conditions) on power supply performance is clearly defined; for example, high temperature and high humidity are negatively impactful operating conditions, while normal temperature and medium load are positively impactful operating conditions. By simulating extreme operating conditions (such as 20°C low temperature and 120% overload), potential performance bottlenecks of the power supply (such as sudden capacity drop and protection mechanism triggering) can be predicted in advance, avoiding failures caused by sudden operating conditions in actual operation. Based on the impact data of different operating conditions, the operating scenarios of the power supply can be adjusted accordingly. For example, a certain energy storage power supply can be prioritized for deployment in grid peak-shaving scenarios with normal temperature and medium load, rather than in industrial and commercial energy storage scenarios with high temperature and high load, thereby improving the overall service life of the power supply.

[0079] In one embodiment of the present invention, the step of performing an influence analysis on different operating conditions based on the multiple operating condition prediction data and the multiple operating condition combination prediction data to obtain operating condition influence data includes:

[0080] The predicted data for each working condition combination is compared with the corresponding simulated data for the working condition combination to obtain the working condition combination comparison results.

[0081] When the predicted data of the working condition combination is greater than the simulated data of the corresponding working condition, the negative impact working condition type is determined by the simulated data of the corresponding working condition to obtain the negative impact working condition type.

[0082] When the predicted data of the working condition combination is less than or equal to the corresponding working condition simulation data, the working condition type that is positively affected by the corresponding working condition simulation data is determined, and the working condition type that is positively affected is obtained.

[0083] The negative and positive operating condition types mentioned above are the operating condition type impact data.

[0084] For example, for combined temperature and humidity conditions, combined prediction data of temperature and humidity conditions are obtained and compared with temperature conditions and humidity conditions respectively to determine whether temperature is a positive influence condition and humidity is a negative influence condition.

[0085] The working principle and technical effects of the above-mentioned technical solution are as follows: Using single-condition simulation data as a benchmark, since single-condition data can reflect the impact of the factor acting independently, the difference between the combined data and the benchmark can be attributed to the synergistic effect of multiple factors, thereby accurately identifying the types of conditions that exacerbate negative impacts or weaken positive impacts. By comparing numerical values, subjective judgment is replaced; for example, a higher attenuation rate and lower efficiency directly correspond to a negative impact, avoiding judgment errors caused by experience bias.

[0086] By breaking down the complex effects of combined operating conditions into individual factors, we can avoid attributing them solely to multiple influences and thus preventing targeted adjustments. For example, once humidity is identified as a major negative factor, we can focus on strengthening moisture-proofing measures for the power supply. Based on quantitative comparison logic, the impact assessment of each type of operating condition has clear numerical basis, facilitating verification and correction and improving the reliability of the analysis results. After identifying the types of operating conditions with negative impacts, we can prioritize avoiding or mitigating their effects in subsequent operations, while appropriately strengthening operating conditions with positive impacts to improve management efficiency.

[0087] In one embodiment of the present invention, the step of performing periodic state prediction by combining a power supply monitoring model with periodically collected data to obtain periodic state prediction data includes:

[0088] Periodic status data is obtained by combining the periodic collected data with physical attribute information;

[0089] Based on the periodic acquisition data and the periodic status data, the periodic status data of the next cycle in the whole life cycle is predicted to obtain the periodic acquisition status prediction data.

[0090] The periodic data is adjusted based on the operating condition type influence data to obtain periodic adjustment data; the influence data of the adjustment condition on the evolution of the result is obtained by changing one or more operating condition types in the current cycle.

[0091] Based on the cycle adjustment data and cycle state data, the cycle state data of the next cycle in the whole life cycle is predicted to obtain the cycle adjustment state prediction data.

[0092] The periodic acquisition state prediction data and the periodic adjustment state prediction data are the periodic state prediction data.

[0093] The working principle and technical effect of the above solution are as follows: Based on historical trend extrapolation, if the SOH of a certain battery cell in the first three cycles is 95%, 92%, and 90% respectively, the next cycle is predicted to naturally decay to 88% based on this decay trend, reflecting the cycle change pattern under no intervention conditions. By introducing operating condition intervention variables, and adjusting key operating condition parameters according to the impact data of different operating conditions, the cycle state is re-predicted, reflecting the impact of intervention measures on cycle changes. Cycle data collection serves as the historical basis for prediction, physical attribute information is the baseline boundary for prediction, and operating condition impact data is the adjustment tool for prediction; the three work together to ensure the rationality and operability of the prediction.

[0094] By anticipating the natural degradation trend and post-adjustment trend of the next cycle, we can avoid reactive maintenance due to unknown degradation. By comparing the collected data and predicted adjustment data, we can quantify the impact of adjustment measures and highlight the role of adjustment factors. For different energy storage power sources, we can develop differentiated adjustment strategies to improve management accuracy.

[0095] In one embodiment of the present invention, S2 includes:

[0096] The energy storage power supply will be subjected to periodic data collection for the next cycle of its entire life cycle to obtain the data collected in the next cycle;

[0097] The cycle state prediction data is linked with the data collected in the next cycle for analysis to obtain cycle change analysis data.

[0098] Based on the periodic change analysis data, information on abnormal factors in periodic changes is determined;

[0099] Based on the information on abnormal factors in the cycle change, early warning and adjustment of the negative impact of factors are carried out until the full life cycle data is obtained.

[0100] The working principle and technical effect of the above solution are as follows: Using the predicted periodic state data as the expected target and the data collected in the next period as the actual result, the deviation between the two is compared to pinpoint the abnormal factors causing the deviation. Combining the influence data of different operating conditions, if the actual temperature is higher than the predicted temperature, and the actual SOH decay is greater than the predicted decay, it can be attributed to the abnormal factor of abnormal temperature rise, avoiding confusion in attribution. After each adjustment, the data for the next period is re-collected and compared with the new predicted data, continuously correcting abnormal factors and adjustment strategies to ensure that the power supply is always in optimal operating condition throughout its entire life cycle.

[0101] By rapidly identifying abnormal factors from prediction deviations, early warning and adjustment response times are controlled within minutes, preventing malfunctions caused by the escalation of anomalies (such as cell bulging). Through cyclical data collection and recording, a complete archive is created covering prediction data, actual data, abnormal factors, and adjustment measures for each cycle, providing comprehensive data support for decommissioning assessments (such as the value of secondary utilization). As cyclical data accumulates, the accuracy of anomaly identification and the effectiveness of adjustment measures continuously improve, forming a virtuous cycle of data-driven and strategy optimization.

[0102] In one embodiment of the present invention, the step of performing a linkage analysis between the periodic state prediction data and the data collected in the next period to obtain periodic change analysis data includes:

[0103] The ratio of the periodic acquisition state prediction data to the next period acquisition data is used to obtain the prediction period change coefficient.

[0104] The predicted cycle change coefficient is compared with the predicted cycle change threshold to obtain the predicted cycle comparison result;

[0105] Based on the comparison results of the prediction cycle, the development direction of the data collected in the next cycle is determined, and positive development judgment and negative development judgment are obtained.

[0106] When the development direction of the data collected in the next cycle is determined to be negative, the ratio of the predicted periodic adjustment state data to the data collected in the next cycle is used to obtain the adjustment cycle change coefficient.

[0107] The variation data of the adjustment cycle variation coefficient is analyzed to obtain cycle variation analysis data.

[0108] Among them, the development direction of the data collected in the next cycle is determined based on the comparison results of the prediction cycle, resulting in positive development and negative development determinations, including:

[0109] When the predicted state data of a cycle is greater than the data collected in the next cycle, a positive development judgment is made on the data collected in the next cycle.

[0110] When the predicted state data of a cycle is less than or equal to the data collected in the next cycle, a negative development judgment is made on the data collected in the next cycle.

[0111] The working principle and technical effect of the above technical solution are as follows: If the adjustment cycle change coefficient is close to 1 (e.g., 1.005), it indicates that the deviation between the prediction and the actual situation after adjustment is small, and the current adjustment strategy is effective; if the coefficient deviates significantly from 1 (e.g., 1.023), it indicates that the adjustment strategy has not achieved the expected effect (e.g., the attenuation is not suppressed after temperature adjustment), and further parameter optimization is required (e.g., increasing the cooling fan speed from 1500 rpm to 2000 rpm). From the predicted cycle change coefficient to the adjustment cycle change coefficient, a calculation chain of deviation identification, direction determination, and adjustment verification is formed, ensuring that each step of the calculation provides data support for subsequent anomaly analysis and strategy optimization, avoiding the blind decision-making caused by isolated calculations.

[0112] The adjustment coefficient calculation is initiated only when the trend is negative, reducing unnecessary computing power consumption. This is especially beneficial in the management of large-scale energy storage power stations (containing hundreds of energy storage power sources), as it reduces cloud computing pressure and improves the overall management system's response speed. Based on the deviation value of the adjustment cycle variation coefficient, optimization targets can be accurately set, avoiding repeated trial and error caused by experience-based adjustments and shortening the strategy optimization cycle.

[0113] In one embodiment of the present invention, determining the information on abnormal factors of periodic changes based on the periodic change analysis data includes:

[0114] Obtain a preset adjustment cycle change threshold, calculate the difference between the adjustment cycle change coefficient and the preset adjustment cycle change threshold, and obtain adjustment change difference data;

[0115] A preset adjustment change difference threshold is obtained, and the adjustment change difference data is compared with the preset adjustment change difference threshold to obtain the adjustment change difference comparison result. When the difference is less than the threshold, the corresponding working condition type influence data is obtained and identified as periodic change abnormal factor information. All thresholds in this application can be determined based on historical data analysis experience.

[0116] Based on the comparison results of the adjustment change difference, the information on abnormal factors of periodic change is determined according to the operating condition type influence data.

[0117] The working principle and technical effects of the above technical solution are as follows: The preset adjustment cycle change threshold and difference threshold are determined based on historical normal operation data and optimal management cases. For example, the average adjustment cycle change coefficient of the past 100 normal cycles is selected as the preset threshold to ensure that the threshold has practical reference significance rather than being subjectively set. The adjustment change difference data reflects the degree of deviation between the expected and actual adjustment strategy. A difference exceeding the threshold indicates that the deviation exceeds the normal range, suggesting the existence of abnormal factors that are not effectively controlled. By identifying the negatively impacting operating conditions through the data on the types of operating conditions, and then combining this with the actual collected operating condition data, a causal chain is established between abnormal difference, negatively impacting operating conditions, and actual data exceeding the standard, avoiding the randomness of locating abnormal factors. Sorting by the size of the difference allows for the priority handling of abnormal factors that have a greater impact on the cycle status, ensuring that resources (such as maintenance personnel and equipment) are prioritized for solving key problems.

[0118] By employing a dual approach of threshold comparison and operational condition correlation, normal fluctuations are avoided from being misjudged as anomalies, while the root cause is accurately located, significantly improving the accuracy of anomaly detection. When multiple anomalies coexist, differential sorting clarifies processing priorities, preventing delays in addressing critical issues due to indiscriminate handling (e.g., prioritizing humidity anomalies can quickly suppress the rate of degradation). As full lifecycle data accumulates, preset thresholds can be updated periodically to ensure they consistently adapt to the aging state of the energy storage power source, enhancing the adaptability of anomaly detection.

[0119] In one embodiment of the present invention, the step of performing early warning and adjustment of negative impact factors based on the information of abnormal periodic changes, until full life-cycle data is obtained, includes:

[0120] Obtain information on the types of negatively impacted working conditions corresponding to abnormal factors in periodic changes;

[0121] Early warning and corresponding adjustments are made for the types of negative impact operating conditions of energy storage power sources, and negative impact adjustment data is obtained;

[0122] Based on the negative impact adjustment data, the next cycle data of the energy storage power supply is obtained, the cycle state prediction data is obtained, and so on until the full life cycle data is obtained.

[0123] The working principle and technical effects of the above-mentioned technical solution are as follows: Minor anomalies are given low-priority warnings to avoid excessive interference with operation and maintenance; severe anomalies are given high-priority warnings to ensure rapid response, conforming to the management principle of matching risk with response intensity. Based on the locked negative impact condition type (such as humidity), adjustment measures directly target the control link of that condition (such as dehumidification device), avoiding resource waste and secondary problems (such as increased energy consumption due to excessive cooling) caused by broad-spectrum adjustments (such as simultaneously adjusting temperature, humidity, and load). Through a cycle of adjustment, data collection, and analysis, it is ensured that abnormal factors are continuously monitored and handled throughout the entire life cycle. Even during the aging stage of the energy storage power supply, its basic performance can be maintained through cyclical optimization, extending its retirement time.

[0124] A tiered early warning mechanism ensures that the response time for minor anomalies is controlled within 30 minutes and for severe anomalies within 10 minutes, preventing escalation of faults due to delayed warnings. Targeted adjustment measures reduce unnecessary operations (such as adjusting normal temperature parameters without need), significantly reducing the number of warnings required for individual energy storage power plants during operation and maintenance, resulting in substantial annual cost savings when applied on a large scale. Completely recorded lifecycle data can not only be used for the decommissioning assessment of current energy storage power plants but also serve as a design reference for new energy storage power plants of the same type (such as optimizing dehumidification device power), forming a positive feedback loop in management and design.

[0125] According to one embodiment of the present invention, the system includes:

[0126] The predictive analysis module is used to establish a digital twin model of the energy storage power supply, acquire periodic acquisition data, synchronize data, acquire the power supply monitoring model, and perform predictions of different operating conditions and periodic states based on the power supply monitoring model. It acquires various operating condition prediction data and various combinations of operating condition data for different operating condition predictions, performs operating condition type impact analysis, obtains operating condition type impact data, acquires periodic acquisition state prediction data and periodic adjustment state prediction data for periodic state prediction, and then obtains different operating condition state data and periodic state prediction data.

[0127] The cycle variation analysis module is used to analyze the changes in the data collected from the energy storage power source for the next cycle, obtain cycle variation analysis data, identify abnormal factors in cycle variation, and perform early warning and adjustment for abnormal factors until the full life cycle data is obtained.

[0128] The working principle and technical effects of the above solution are as follows: Using a digital twin model as the core carrier, the model accurately maps the physical state of the energy storage power supply through real-time synchronization of periodically collected data (such as voltage, temperature, and load), forming a power supply monitoring model with simulation and prediction capabilities. For different single-factor operating conditions (such as temperature and load) and combined operating conditions, the model simulates the operating performance of the energy storage power supply and analyzes the impact of various operating conditions on power supply performance. Based on historical periodic data, the model predicts the natural operating state of the next period, and simultaneously adjusts parameters based on the operating condition impact data to predict the adjusted periodic state. The predicted data is compared with the actual collected data of the next period to identify and adjust abnormal factors. Through a cycle of prediction, verification, and optimization, the entire lifecycle of the energy storage power supply, from commissioning to decommissioning, is covered.

[0129] This invention breaks away from the traditional reactive maintenance model, using dual-dimensional prediction to proactively understand the impact of operating conditions and cyclical change trends, thus transforming from passive response to proactive management. From initial model building to final full lifecycle data generation, the process spans the entire operation of the energy storage power source, avoiding performance losses or safety risks caused by management gaps. The linkage between periodically collected and predicted data not only supports current cycle optimization but also provides data references for the management of similar energy storage power sources, forming a reusable management experience library.

[0130] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for full lifecycle management of energy storage power sources based on digital twins, characterized in that, The method includes: S1. Establish a digital twin model for the energy storage power supply, acquire periodic acquisition data, synchronize the data, acquire the power supply monitoring model, predict different operating conditions and periodic conditions based on the power supply monitoring model, acquire multiple operating condition prediction data and multiple combinations of operating condition data for different operating conditions, conduct operating condition type impact analysis, obtain operating condition type impact data, acquire periodic acquisition state prediction data and periodic adjustment state prediction data for periodic conditions, and then obtain different operating condition data and periodic state prediction data. S2. Analyze the changes in the data collected from the energy storage power source for the next cycle, obtain cycle change analysis data, identify abnormal factors in cycle changes, and carry out early warning and adjustment for abnormal factors until the full life cycle data is obtained.

2. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 1, characterized in that, S1 includes: Obtain the physical attribute information of the energy storage power source, and establish a digital twin model based on the physical attribute information; The energy storage power supply is subjected to periodic data collection throughout its entire life cycle based on physical property information using a sensor array, and periodic data collection is obtained. The periodically collected data is synchronously transmitted to the digital twin model to obtain the power monitoring model; By combining the power monitoring model with physical attribute information, prediction data for different operating conditions is obtained. Periodic state prediction data is obtained by combining power monitoring models with periodic data acquisition.

3. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 2, characterized in that, The process of predicting different operating conditions by combining a power supply monitoring model with physical attribute information to obtain prediction data for different operating conditions includes: Acquire multiple preset working condition data, and generate multiple working condition simulation data based on the multiple preset working condition data and the physical attribute information; By using a power monitoring model to predict the operation of energy storage power sources based on simulation data of various operating conditions, prediction data for various operating conditions can be obtained. Multiple preset working condition data are arranged and combined to obtain multiple combined working condition data; Based on the combined working condition data and the physical attribute information, various working condition combination simulation data are generated. The operating performance of energy storage power sources is predicted by using power monitoring models to simulate data of various operating conditions, and prediction data of various operating conditions are obtained. Based on the prediction data of various working conditions and the prediction data of various working condition combinations, an analysis of the impact of working condition types is conducted to obtain the impact data of working condition types.

4. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 3, characterized in that, Based on the predicted data of various operating conditions and the predicted data of combinations of various operating conditions, an analysis of the impact of different operating conditions is conducted to obtain the data on the impact of different operating conditions, including: The predicted data for each working condition combination is compared with the corresponding simulated data for the working condition combination to obtain the working condition combination comparison results. When the predicted data of the working condition combination is greater than the simulated data of the corresponding working condition, the negative impact working condition type is determined by the simulated data of the corresponding working condition to obtain the negative impact working condition type. When the predicted data of the working condition combination is less than or equal to the corresponding working condition simulation data, the working condition type that is positively affected by the corresponding working condition simulation data is determined, and the working condition type that is positively affected is obtained. The negative and positive operating condition types mentioned above are the operating condition type impact data.

5. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 2, characterized in that, The process of predicting the periodic state by combining a power supply monitoring model with periodically collected data to obtain periodic state prediction data includes: Periodic status data is obtained by combining the periodic collected data with physical attribute information; Based on the periodic acquisition data and the periodic status data, the periodic status data of the next cycle in the whole life cycle is predicted to obtain the periodic acquisition status prediction data. The periodic acquisition data is adjusted based on the impact data of the aforementioned operating conditions to obtain periodic adjustment data; Based on the cycle adjustment data and cycle state data, the cycle state data of the next cycle in the whole life cycle is predicted to obtain the cycle adjustment state prediction data. The periodic acquisition state prediction data and the periodic adjustment state prediction data are the periodic state prediction data.

6. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 1, characterized in that, S2 includes: The energy storage power supply will be subjected to periodic data collection for the next cycle of its entire life cycle to obtain the data collected in the next cycle; The cycle state prediction data is linked with the data collected in the next cycle for analysis to obtain cycle change analysis data. Based on the periodic change analysis data, information on abnormal factors in periodic changes is determined; Based on the information on abnormal factors in the cycle change, early warning and adjustment of the negative impact of factors are carried out until the full life cycle data is obtained.

7. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 6, characterized in that, The predicted periodic state data is linked with the data collected in the next period for analysis to obtain periodic change analysis data, including: The ratio of the periodic acquisition state prediction data to the next period acquisition data is used to obtain the prediction period change coefficient. The predicted cycle change coefficient is compared with the predicted cycle change threshold to obtain the predicted cycle comparison result; Based on the comparison results of the prediction cycle, the development direction of the data collected in the next cycle is determined, and positive development judgment and negative development judgment are obtained. When the development direction of the data collected in the next cycle is determined to be negative, the ratio of the predicted periodic adjustment state data to the data collected in the next cycle is used to obtain the adjustment cycle change coefficient. The variation data of the adjustment cycle variation coefficient is analyzed to obtain cycle variation analysis data.

8. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 6, characterized in that, The step of determining the information on abnormal factors of periodic changes based on the periodic change analysis data includes: Obtain a preset adjustment cycle change threshold, calculate the difference between the adjustment cycle change coefficient and the preset adjustment cycle change threshold, and obtain adjustment change difference data; A preset adjustment change difference threshold is obtained, and the adjustment change difference data is compared with the preset adjustment change difference threshold to obtain the adjustment change difference comparison result. Based on the comparison results of the adjustment change difference, the information on abnormal factors of periodic change is determined according to the operating condition type influence data.

9. The method for full lifecycle management of energy storage power sources based on digital twins according to claim 6, characterized in that, Based on the information on abnormal factors in the cycle changes, early warning and adjustment of the negative impact of factors are carried out until the full life cycle data is obtained, including: Obtain information on the types of negatively impacted working conditions corresponding to abnormal factors in periodic changes; Early warning and corresponding adjustments are made for the types of negative impact operating conditions of energy storage power sources, and negative impact adjustment data is obtained; Based on the negative impact adjustment data, the next cycle data of the energy storage power supply is obtained, the cycle state prediction data is obtained, and so on until the full life cycle data is obtained.

10. A full lifecycle management system for energy storage power sources based on digital twins, characterized in that, The system includes: The predictive analysis module is used to establish a digital twin model of the energy storage power supply, acquire periodic acquisition data, synchronize data, acquire the power supply monitoring model, and perform predictions of different operating conditions and periodic states based on the power supply monitoring model. It acquires various operating condition prediction data and various combinations of operating condition data for different operating condition predictions, performs operating condition type impact analysis, obtains operating condition type impact data, acquires periodic acquisition state prediction data and periodic adjustment state prediction data for periodic state prediction, and then obtains different operating condition state data and periodic state prediction data. The cycle variation analysis module is used to analyze the changes in the data collected from the energy storage power source for the next cycle, obtain cycle variation analysis data, identify abnormal factors in cycle variation, and perform early warning and adjustment for abnormal factors until the full life cycle data is obtained.

Images