A large-scale liquid metal battery energy storage safety management method, system, device and medium

Abstract

The application discloses a large-scale liquid metal battery energy storage safety management method, system, device and medium, the method comprises the following steps: predicting the short-term thermal behavior trend of the liquid metal battery through a prediction model, combining the prediction result with a comprehensive parameter set to evaluate the safety state of the liquid metal battery, and obtaining an evaluation result; determining a target battery through the evaluation result; comparing the evaluation result with a preset early warning condition to determine the risk level of the target battery, and executing a corresponding safety strategy according to the risk level; in the process of executing the safety strategy, continuously monitoring the multi-dimensional state parameters of the target battery, and when the monitoring result shows that the multi-dimensional state parameters all recover to a preset safety range, the early warning is released, otherwise, the safety strategy is upgraded to a higher level. The application solves the problems in the prior art that single parameter monitoring cannot accurately identify early abnormal characteristics, lacks abnormal prediction ability and hierarchical processing mechanism, and lacks closed-loop control.

Description

Technical Field

[0001] This invention relates to the field of battery energy storage technology, and in particular to a method, system, device and medium for the safe management of large-scale liquid metal battery energy storage. Background Technology

[0002] In recent years, large-capacity energy storage technology has been increasingly widely used in power systems. Liquid metal batteries have become an important choice in the field of large-scale energy storage due to their long cycle life. However, the operating temperature of liquid metal batteries is usually between 450℃ and 700℃, and this high-temperature environment brings considerable difficulties to safety management.

[0003] Existing safety management technologies for liquid metal batteries primarily rely on threshold monitoring of single parameters. While these methods can issue warnings when parameters are significantly abnormal, the complex relationships between multiple state parameters within a liquid metal battery, such as temperature, pressure, voltage, and current, make it difficult to accurately identify early abnormalities like short circuits and overheating based solely on threshold judgments for a single parameter. Furthermore, traditional monitoring methods lack the ability to predict future trends, only responding passively after an anomaly occurs. The countermeasures are also relatively simple, typically involving directly cutting off charging and discharging or activating the cooling system. This lacks the flexibility to classify and handle situations according to risk levels. After implementing safety strategies, there is also a lack of continuous monitoring and effectiveness evaluation of the battery's state, making it difficult to dynamically adjust strategies based on actual conditions. Consequently, the safety management process lacks closed-loop control capabilities. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method, system, device, and medium for the safety management of large-scale liquid metal battery energy storage, which solves the problems in existing battery safety management technologies, such as the difficulty in accurately identifying early abnormal characteristics through single-parameter monitoring, the lack of abnormal prediction capabilities and hierarchical processing mechanisms, and the lack of closed-loop control.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for safe management of large-scale liquid metal battery energy storage, comprising: real-time acquisition and fusion analysis of multi-dimensional state parameters of the liquid metal battery to obtain a comprehensive parameter set; short-term prediction of the thermal behavior trend of the liquid metal battery using a prediction model, and evaluation of the safety status of the liquid metal battery by combining the prediction results with the comprehensive parameter set to obtain an evaluation result, and determining the target battery based on the evaluation result; comparing the evaluation result with preset warning conditions to determine the risk level of the target battery, and executing a corresponding safety strategy according to the risk level; continuously monitoring the multi-dimensional state parameters of the target battery during the execution of the safety strategy, and lifting the warning when the monitoring results show that the multi-dimensional state parameters have all recovered to the preset safety range, otherwise upgrading to a higher level of safety strategy; recording the full-cycle data of the warning process, and generating an event analysis report based on the full-cycle data.

[0007] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the step of obtaining the comprehensive parameter set includes: real-time acquisition of multi-dimensional state parameters of the liquid metal battery through sensors and preprocessing them; using a multi-source data fusion algorithm to perform correlation analysis on the preprocessed multi-dimensional state parameters, and generating the comprehensive parameter set based on the correlation analysis results.

[0008] The beneficial effects of this preferred technical solution are as follows: By preprocessing the collected multi-dimensional state parameters, abnormal values ​​and missing data caused by sensor failure or communication interruption can be corrected and supplemented. The multi-source data fusion algorithm can comprehensively consider measurement data from multiple dimensions such as temperature, pressure, voltage, and current, and obtain parameter values ​​that are closer to the true state of the liquid metal battery through fusion processing, avoiding the influence of measurement errors from a single sensor. Correlation analysis can identify abnormal correlation patterns between multi-dimensional state parameters. For example, when the temperature rises while the voltage and pressure rise simultaneously, this synchronous change of multiple parameters can be identified as an early characteristic of an internal short circuit. Finally, the generated comprehensive parameter set contains the fusion results of each dimension and the abnormal patterns identified by correlation analysis, which can reflect the true operating state and potential risks of the battery, enabling anomalies to be identified before they develop into serious problems.

[0009] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the step of obtaining the evaluation result includes: inputting the historical state data of the liquid metal battery and the comprehensive parameter set into the prediction model, predicting the temperature and pressure change trends of the liquid metal battery within a preset future period, and comparing the temperature and pressure values ​​in the prediction results with the corresponding early warning thresholds to obtain a prediction evaluation; comparing the temperature and pressure values ​​in the comprehensive parameter set with the corresponding early warning thresholds to obtain a current evaluation; combining the current evaluation and the prediction evaluation to determine whether there is an abnormal risk in the battery, and generating an evaluation result.

[0010] The beneficial effects of this preferred technical solution are as follows: By inputting historical state data and a comprehensive parameter set into the prediction model, it is possible to predict the temperature and pressure change trends within a preset time period based on the past operating patterns of liquid metal batteries. This makes safety management no longer limited to monitoring the current state, but can identify potential deterioration trends before parameters reach dangerous levels. The combination of predictive assessment and current assessment constructs a dual judgment mechanism. The current assessment reflects whether the battery is currently in an abnormal state, while the predictive assessment reveals whether the battery will enter an abnormal state in the future. The comprehensive judgment of the two can cover more abnormal scenarios. Even if the current temperature and pressure are still within the safe range, if the prediction shows that they will exceed the warning threshold in the future, it is still possible to determine that there is an abnormal risk and intervene in advance, avoiding the passive situation of taking measures only after the parameters actually exceed the standard. Furthermore, by comparing the temperature and pressure values ​​with the corresponding warning thresholds, different judgment standards can be set for different parameters, making the assessment process more in line with the actual operating characteristics of liquid metal batteries.

[0011] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the step of determining the risk level of the target battery includes: comparing the evaluation results with preset warning conditions, determining the risk level of the target battery according to preset level classification rules; determining whether the evaluation results meet the jump-level triggering conditions, and adjusting the risk level to the highest level when the temperature value or pressure value in the evaluation results exceeds the preset emergency threshold.

[0012] The beneficial effects of this preferred technical solution are as follows: By comparing the evaluation results with preset warning conditions and determining the risk level according to the classification rules, appropriate measures can be taken based on the severity of the abnormal situation, avoiding over-intervention or under-intervention caused by using a uniform approach for all abnormal situations. The setting of the skip-level trigger condition allows the risk level to be directly adjusted to the highest level when the temperature or pressure value exceeds the preset emergency threshold. This mechanism can skip the conventional step-by-step escalation process when the battery is in an extremely dangerous state, shortening the time interval between anomaly identification and the implementation of the highest-intensity measures. The judgment of the emergency threshold is independent of the conventional classification rules, ensuring that even if the risk level is determined to be low according to conventional rules, the highest-level response can still be triggered if the temperature or pressure reaches the critical value that may lead to serious consequences. This mechanism, combining conventional classification with emergency skip-level, ensures that appropriate measures can be taken to maintain system operating efficiency in general abnormal situations, while also ensuring that the strongest safety protection can be quickly activated in dangerous situations.

[0013] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the method includes: executing a corresponding safety strategy according to the risk level, wherein the safety strategy includes a primary strategy and a secondary strategy; when the risk level is primary, the primary strategy is executed, which includes adjusting the charge and discharge state of the target battery and strengthening the cooling of the area where the target battery is located; when the risk level is secondary, the secondary strategy is executed, which includes closing a dedicated safety discharge circuit connected to both ends of the target battery and discharging the target battery in a controlled manner with a preset safety current.

[0014] The beneficial effects of this preferred technical solution are as follows: By setting two levels of safety strategies, namely primary and secondary strategies, intervention measures of different intensities can be taken according to the degree of risk. In the case of mild anomalies, the rising trend of temperature and pressure can be controlled by adjusting the charging and discharging state and strengthening cooling. This method has little impact on the normal operation of the battery and can maintain the availability of the system as much as possible while ensuring safety. When the primary strategy cannot effectively control the risk, the secondary strategy uses a dedicated safety discharge circuit to discharge the target battery in a controlled manner. This method is independent of the normal charging and discharging system, avoiding the impact of abnormal batteries on the entire system. At the same time, discharging with a preset safe current can reduce the state of charge of the battery to a safe level, fundamentally eliminating the safety hazards caused by excessive charge. This graded mechanism from mild intervention to deep intervention enables safety management to flexibly respond to anomalies of different degrees and take sufficiently strong measures to prevent accidents when necessary.

[0015] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the step of continuously monitoring the multi-dimensional state parameters of the target battery includes: setting a time window after implementing a safety strategy, and collecting the multi-dimensional state parameters of the target battery in real time at preset intervals within the time window; calculating the rate of temperature and pressure decrease respectively, and determining whether the rate of temperature and pressure decrease both reach the preset effective rate; when the rate of decrease reaches the effective rate, continuing to observe until the end of the time window, and determining whether the multi-dimensional state parameters have all recovered to the preset safe range; otherwise, determining that the current safety strategy is invalid, ending the observation early, and upgrading to a higher level of safety strategy.

[0016] The beneficial effects of this preferred technical solution are as follows: By setting a time window and collecting multi-dimensional state parameters of the target battery at intervals after the implementation of the safety strategy, a continuous data sequence of parameter changes over time can be formed. This makes the evaluation of the effectiveness of the safety strategy no longer dependent on a single measurement result, but based on the trend of parameter changes. Calculating the rate of temperature and pressure decrease and comparing it with the preset effective rate can identify whether the safety strategy has truly taken effect. Even if the temperature and pressure decrease, if the rate of decrease is too slow, it indicates that the current measures are insufficient to restore the battery state to a safe range within a reasonable time. When the rate of decrease reaches the effective rate, observation continues until the end of the time window, which can confirm whether the parameters have truly recovered to the safe range rather than just fluctuating briefly, avoiding repeated risks caused by prematurely lifting the warning. When the rate of decrease does not reach the effective rate, the observation ends early and the safety strategy is upgraded to a higher level, avoiding wasting time on ineffective strategies and delaying the control of abnormal situations. This timely adjustment mechanism forms a complete closed loop of execution, monitoring, evaluation, and adjustment in the safety management process, enabling flexible responses to different abnormal situations based on actual results.

[0017] As a preferred embodiment of the large-scale liquid metal battery energy storage safety management method of the present invention, the step of upgrading to a higher level of safety strategy includes: if the current execution is the first-level strategy, then upgrading to the second-level strategy for execution and monitoring; if the current execution is the second-level strategy, then fault isolation of the target battery, wherein the fault isolation includes disconnecting the electrical connection between the target battery and the charging and discharging system, and marking the target battery as a fault isolation state.

[0018] The beneficial effects of this preferred technical solution are as follows: By setting an upgrade path from a primary strategy to a secondary strategy, stronger controlled discharge measures can be taken when mild intervention measures cannot effectively control the anomaly. This step-by-step upgrade mechanism avoids excessive intervention in the battery at the initial stage, while ensuring that the intensity of intervention can be increased in a timely manner when necessary. When the secondary strategy still cannot restore the target battery to a safe state, the electrical connection between the target battery and the charging and discharging system is disconnected through fault isolation. This prevents the target battery from affecting the normal operation of the entire energy storage system and avoids the spread of a single target battery's fault to other liquid metal batteries or system equipment. Disconnecting the electrical connection means that the target battery no longer participates in the charging and discharging process, eliminating the risk of further deterioration or even safety accidents that may result from continued operation. Marking the target battery as being in a fault-isolated state allows the abnormal information of the target battery to be recorded at the system level, preventing the target battery from being put back into use without maintenance. This complete process from strategy upgrade to final isolation enables safety management to handle various situations from mild anomalies to severe faults.

[0019] Secondly, the present invention provides a large-scale liquid metal battery energy storage safety management system, comprising: Status monitoring module: used to collect multi-dimensional status parameters of liquid metal batteries in real time; Information processing module: used to perform fusion analysis on the multi-dimensional state parameters, make short-term predictions on the thermal behavior trend of the battery through the prediction model and assess the safety status of the battery, determine the risk level of the target battery, generate corresponding control commands, and record the full-cycle data of the early warning process and generate an event analysis report. Battery charging and discharging control module: used to adjust the charging and discharging state of the target battery according to the control command, and to control the dedicated safety discharge circuit to discharge the target battery in a controlled manner; Thermal management module: used to regulate the heat of the area where the target battery is located according to the control command.

[0020] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of a large-scale liquid metal battery energy storage safety management method.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the large liquid metal battery energy storage safety management method.

[0022] Compared with existing technologies, the beneficial effects of this invention are as follows: By employing a multi-source data fusion algorithm to perform correlation analysis on multi-dimensional state parameters of liquid metal batteries, such as temperature, pressure, voltage, and current, it can identify abnormal correlation patterns between multiple parameters. For example, the simultaneous occurrence of temperature rise, voltage drop, and pressure rise can be identified as an early characteristic of internal short circuit. This multi-dimensional comprehensive judgment has higher accuracy and sensitivity compared to single-parameter threshold monitoring. Furthermore, a predictive model is introduced to make short-term predictions of the thermal behavior trends of liquid metal batteries. The prediction results are combined with the current state for comprehensive evaluation, enabling safety management to shift from passive response to proactive prevention. Potential risks can be identified and intervened in advance before parameters reach dangerous levels. In addition, for… For anomalies of the same severity, a tiered safety strategy is designed. The first-level strategy addresses minor anomalies by adjusting the charging and discharging state and enhancing cooling, while the second-level strategy handles severe anomalies through controlled discharge via a dedicated safety discharge circuit. This tiered mechanism avoids the impact of excessive intervention on system operating efficiency and ensures that sufficiently strong measures can be taken in high-risk situations. More importantly, during the execution of the safety strategy, the multi-dimensional state parameters of the target battery are continuously monitored. The effectiveness of the strategy is evaluated by calculating the rate of temperature and pressure decrease. When the rate of decrease reaches the preset effective rate, observation continues until the parameters return to the safe range. When the strategy is ineffective, it is promptly upgraded to a higher-level strategy or fault isolation is performed, forming a complete closed-loop control mechanism. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the overall process of a large-scale liquid metal battery energy storage safety management method according to an embodiment of the present invention. Detailed Implementation

[0025] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0026] Example 1, referring to Figure 1As an embodiment of the present invention, a method for safe management of large-scale liquid metal battery energy storage is provided, including steps S100 to S500: S100: Real-time acquisition of multi-dimensional state parameters of liquid metal batteries and fusion analysis to obtain a comprehensive parameter set.

[0027] S200. The thermal behavior trend of liquid metal batteries is predicted in the short term using a prediction model. The prediction results are combined with a comprehensive parameter set to evaluate the safety status of the liquid metal batteries and obtain the evaluation results. The target battery is determined based on the evaluation results.

[0028] S300 compares the assessment results with the preset warning conditions to determine the risk level of the target battery and executes the corresponding safety strategy according to the risk level.

[0029] S400: During the execution of the safety strategy, the multi-dimensional state parameters of the target battery are continuously monitored. When the monitoring results show that the multi-dimensional state parameters have all returned to the preset safety range, the warning is lifted; otherwise, the safety strategy is upgraded to a higher level.

[0030] S500 records the entire lifecycle data of the early warning process and generates an event analysis report based on the full lifecycle data.

[0031] It should be noted that liquid metal batteries operate at temperatures as high as 450℃ to 700℃. During operation, the internal state parameters of liquid metal batteries, such as temperature, pressure, voltage, and current, are interconnected and vary significantly. Affected by factors such as charging and discharging power, ambient temperature, and heat dissipation conditions, the internal temperature and pressure of liquid metal batteries may experience large gradient changes, often resulting in delayed safety warnings. At the same time, due to the complex electrochemical reactions inside liquid metal batteries, abnormal situations such as short circuits and overheating are highly likely to occur, leading to safety accidents. The high temperature and high pressure environment has a significant impact on the state measurement of liquid metal batteries and can also damage them due to fatigue effects. Therefore, the safety monitoring and prediction of liquid metal batteries are also very important.

[0032] Therefore, to address the aforementioned operational monitoring and safety prediction issues, steps S100-S500 are employed to utilize a multi-source data fusion algorithm to perform correlation analysis on multi-dimensional state parameters, identify abnormal correlation patterns, and accurately identify early abnormal characteristics such as internal short circuits in liquid metal batteries. Combined with a predictive model, short-term predictions of thermal behavior trends are made, enabling early warning of potential safety risks. Through a tiered safety strategy and continuous monitoring mechanism, the charging / discharging state and cooling intensity are dynamically adjusted according to the risk level, or controlled discharge and fault isolation are implemented, achieving timely intervention and effective control of abnormal situations. Simultaneously, by recording full-cycle data and generating event analysis reports, data support is provided for subsequent fault analysis and charging / discharging system optimization, achieving comprehensive safety management of the liquid metal battery energy storage system.

[0033] Example 2, refer to Figure 1 As an embodiment of the present invention, based on the above embodiment, a method for safe management of large-scale liquid metal battery energy storage is provided.

[0034] In this embodiment of the application, step S100 involves real-time acquisition of multi-dimensional state parameters of the liquid metal battery and fusion analysis to obtain a comprehensive parameter set; The steps for obtaining the comprehensive parameter set include A1~A2: A1. Real-time acquisition and preprocessing of multidimensional state parameters of liquid metal batteries using sensors; In this embodiment, multidimensional state parameters of the liquid metal battery are collected in real time using devices such as temperature sensors, voltage sensors, current sensors, and pressure sensors. The collection period is 1 second. After that, the collected multidimensional state parameters are subjected to noise reduction processing and data completion. During the noise reduction process, the current collected value is compared with the average of the most recent collected values. If the deviation exceeds a preset abnormal threshold, the current collected value is determined to be an abnormal value and replaced with the average of the most recent collected values. For example, the abnormal threshold is set to ±10℃. The collected temperature sequence is 350℃, 351℃, 352℃, 380℃, 353℃, 354℃, and 355℃. Among them, 380℃ deviates from the average of the two collected values ​​before and after it (351℃, 352℃, 353℃, and 354℃) of 352.5℃ by 27.5℃, which exceeds the abnormal threshold. Therefore, 380℃ is determined to be an abnormal value and replaced with 352.5℃. Data completion addresses the issue of missing raw data due to sensor malfunction or communication interruption by using linear interpolation. For example, if temperature data is missing in a certain acquisition cycle, with the previous cycle showing 352℃ and the next cycle showing 354℃, it can be completed to 353℃ through linear interpolation.

[0035] A2. A multi-source data fusion algorithm is used to perform correlation analysis on the preprocessed multi-dimensional state parameters, and a comprehensive parameter set is generated based on the correlation analysis results. In this embodiment, a Kalman filter algorithm is used to perform fusion analysis on the preprocessed multidimensional state parameters. By comprehensively considering the measurement data and error range of multiple sensors, the true state of the liquid metal battery is estimated. Simultaneously, correlation analysis is performed to identify abnormal correlation patterns between the multidimensional state parameters. For example, for the temperature sequence 350℃, 351℃, 352℃, 352.5℃, 353℃, 354℃, and 355℃ after noise reduction in step A1, the current measured value is 355℃. Fusion analysis integrates the changing trends of historical data and the current measured value, and through weighted calculation, the optimal temperature value at the current moment is obtained as 353.2℃. Simultaneously, a slight decrease of 0.3V in voltage and a slight increase in pressure are detected. Correlation analysis reveals that the simultaneous changes in these three parameters—temperature increase, voltage decrease, and pressure increase—identify as early characteristics of an internal short circuit in the battery. Finally, a comprehensive parameter set is generated based on the fusion and correlation analysis results, including the optimal temperature value of 353.2℃, voltage value, current value, pressure value, state of charge value, and comprehensive evaluation indicators.

[0036] In this embodiment of the application, step S200 involves making a short-term prediction of the thermal behavior trend of the liquid metal battery using a prediction model, and combining the prediction results with a comprehensive parameter set to evaluate the safety status of the liquid metal battery, thereby obtaining an evaluation result and determining the target battery based on the evaluation result. The steps to obtain the evaluation results include B1 to B2: B1. Input the historical state data and comprehensive parameter set of the liquid metal battery into the prediction model to predict the temperature and pressure change trends of the liquid metal battery within a preset time period in the future, and compare the temperature and pressure values ​​in the prediction results with the corresponding warning thresholds to obtain the prediction evaluation; In this embodiment, the historical state data includes the temperature sequence 350℃, 351℃, 352℃, 352.5℃, 353℃, 354℃, and 355℃ from step A1, as well as the corresponding voltage, current, pressure, and state of charge data. The prediction model employs a digital twin model, which establishes a virtual copy of the liquid metal battery, including its thermodynamic, electrochemical, and heat dissipation characteristics. Based on the input historical state data and the comprehensive parameter set generated in step A2, the model simulates the operating state of the liquid metal battery over the next 10 minutes, predicting the trends in temperature and pressure. For example, based on the current temperature of 353.2℃, a voltage drop of 0.3V, and a pressure increase, combined with parameters such as the liquid metal battery's heat capacity, heat dissipation coefficient, and charge / discharge power, the prediction model calculates that the temperature will reach 358.2℃ and the pressure will continue to rise within the next 10 minutes. Finally, the predicted temperature and pressure values ​​are compared with the corresponding warning thresholds to obtain a prediction assessment. For example, if the predicted temperature and pressure values ​​do not exceed the warning threshold, the prediction assessment is "the predicted temperature and pressure do not exceed the warning threshold".

[0037] B2. Compare the temperature and pressure values ​​in the comprehensive parameter set with the corresponding warning thresholds to obtain the current assessment. Combine the current assessment with the predicted assessment to determine whether there is an abnormal risk to the battery and generate the assessment result. First, the current optimal temperature and pressure values ​​are extracted from the comprehensive parameter set generated in step A2 and compared with the corresponding warning thresholds to obtain the current assessment. For example, if neither the current optimal temperature nor the pressure value exceeds the warning threshold, the current assessment is "Current temperature and pressure do not exceed the warning threshold." Then, the current assessment and the predictive assessment in step B1 are combined to determine whether there is an abnormal risk in the liquid metal battery. In this embodiment, although both the current assessment and the predictive assessment show that the temperature and pressure do not exceed the warning threshold, the comprehensive assessment indicators identified by the correlation analysis in step A2 show that the three parameters of temperature rise, voltage drop, and pressure rise change simultaneously, which is consistent with the early characteristics of an internal short circuit in a liquid metal battery. Therefore, it is determined that the liquid metal battery has a potential abnormal risk. Based on the comprehensive judgment result, an assessment result is generated, including the abnormal risk type as "suspected internal short circuit risk," and the liquid metal battery is identified as the target battery, proceeding to the subsequent risk level determination and safety strategy execution steps.

[0038] In an optional implementation, the evaluation result obtained in step S200 can also employ a machine learning algorithm as a prediction model. This machine learning model is trained on a large amount of historical operating data and is capable of identifying complex nonlinear relationships between multidimensional state parameters. For example, a long short-term memory neural network model can be used, taking the temperature, voltage, current, pressure, and state of charge sequences from the most recent multiple acquisition cycles as input. The prediction model outputs predicted values ​​for temperature and pressure in the future time period. These predicted values ​​are then compared with corresponding warning thresholds to obtain a predictive evaluation. Finally, a comprehensive judgment is made by combining the current evaluation and the predictive evaluation to generate an evaluation result and determine the target battery.

[0039] In this embodiment of the application, step S300 involves comparing the evaluation results with preset warning conditions to determine the risk level of the target battery, and then implementing the corresponding safety strategy based on the risk level. The steps for determining the risk level of the target battery include C1 to C2: C1. Compare the assessment results with the preset warning conditions, and determine the risk level of the target battery according to the preset level classification rules; In this embodiment, the preset warning conditions include temperature warning thresholds and pressure warning thresholds. The risk level classification rules are based on the temperature and pressure values ​​and the type of abnormal risk in the assessment results. For example, if the assessment results in step B2 show that neither the temperature nor the pressure exceeds the warning thresholds, but there is a suspected internal short circuit risk, the risk level of the target battery is determined to be Level 1 according to the risk level classification rules. If the assessment results show that the temperature or pressure has exceeded the warning thresholds, or there are clear internal short circuit characteristics and the parameters continue to deteriorate, the risk level is determined to be Level 2.

[0040] C2. Determine whether the assessment results meet the trigger conditions for skipping levels. When the temperature or pressure value in the assessment results exceeds the preset emergency threshold, adjust the risk level to the highest level. In this embodiment, the skip-level trigger condition is that the temperature or pressure value in the assessment result generated in step B2 exceeds a preset emergency threshold. For example, the emergency threshold for temperature is 390°C. If the temperature value in the assessment result is 395°C, exceeding the emergency threshold of 390°C, then the skip-level trigger condition is met. At this time, the risk level is directly adjusted to the highest level, i.e., level two, skipping the level one risk level.

[0041] In one optional implementation, the risk level determination of the target battery in step S300 can also be based on the duration of the abnormality. First, the assessment results are compared with preset warning conditions, and the risk level of the target battery is initially determined according to preset level classification rules. Simultaneously, the historical monitoring records of the target battery are queried to calculate the duration of the target battery's abnormal state within a preset time period. If the abnormal duration exceeds a preset time threshold, for example, if the target battery has been continuously in an abnormal state where the temperature or pressure exceeds the warning threshold for the past 30 minutes, then the target battery is determined to have a persistent abnormal risk, and the initially determined risk level is increased by one level. By combining the abnormal duration, recurring or long-term abnormal situations can be identified, avoiding misjudgments caused by short-term abnormalities.

[0042] Among them, the corresponding security policy is executed according to the risk level, and the security policy includes primary policy and secondary policy; When the risk level is Level 1, the Level 1 strategy is implemented. The Level 1 strategy includes adjusting the charge and discharge state of the target battery and strengthening the cooling of the area where the target battery is located. In this embodiment, adjusting the charge / discharge state includes immediately pausing the charging process of the target battery, and reducing the discharge power to a safe range if the target battery is in a discharging state; for example, if the target battery is charging at 100kW, charging is stopped immediately; if it is discharging at 80kW, the discharge power is reduced to 30kW. Enhancing cooling includes increasing the power of the cooling system in the area where the target battery is located and increasing the flow rate of the cooling medium; for example, increasing the cooling fan speed from 50% to 100% of normal operation and increasing the coolant flow rate from the normal 10L / min to 20L / min to accelerate the heat dissipation process of the target battery.

[0043] When the risk level is level 2, the level 2 strategy is implemented. The level 2 strategy includes closing a dedicated safety discharge circuit connected to both ends of the target battery and discharging the target battery in a controlled manner with a preset safety current. In this embodiment, the dedicated safety discharge circuit is a discharge path independent of the normal charge and discharge system, including a discharge resistor or a discharge load. Closing the dedicated safety discharge circuit means connecting the target battery to the dedicated safety discharge circuit through a control switch; for example, by closing a relay, the target battery is connected to the discharge resistor to form a discharge circuit. The preset safe current is determined based on the capacity and current state of charge of the target battery, ensuring that the discharge process is stable and controllable; for example, if the target battery capacity is 100Ah and the current state of charge is 80%, the safe current is set to 10A, and controlled discharge is performed at a discharge rate of 0.1C, expecting to reduce the battery charge to a safe level within 8 hours. During the controlled discharge process, the temperature and pressure changes of the target battery are continuously monitored to ensure the safety of the discharge process.

[0044] In this application embodiment, step S400 involves continuously monitoring the multi-dimensional state parameters of the target battery during the execution of the safety strategy. When the monitoring results show that the multi-dimensional state parameters have all returned to the preset safety range, the warning is lifted; otherwise, the safety strategy is upgraded to a higher level. The steps for continuously monitoring the multidimensional state parameters of the target battery include D1 to D3: D1. After executing the safety policy, set a time window and collect the multi-dimensional state parameters of the target battery in real time at preset intervals within the time window. In this embodiment, after executing the first-level strategy, the observation time window is set to 10 minutes. Within the observation time window, the multi-dimensional state parameters of the target battery are collected in real time every minute. The collected multi-dimensional state parameters include temperature, voltage, current, pressure, and state of charge. For example, after executing the first-level strategy (stop charging and enhance cooling), starting from time t0, the observation parameters are collected at t1 (1 minute), t2 (2 minutes), t3 (3 minutes)...t... 10 (10 minutes) Collect parameters such as temperature and pressure of the target battery at any time to form time series data.

[0045] D2. Calculate the rate of temperature and pressure decrease respectively, and determine whether the rate of temperature and pressure decrease both reach the preset effective rate. In this embodiment, based on the multidimensional state parameters collected in step D1, the rate of temperature and pressure decrease are calculated respectively. The rate of decrease is obtained by dividing the difference between the current collected value and the initial collected value by the time interval. For example, if the target battery temperature is 353.2℃ before implementing the first-level strategy, and the temperature collected 5 minutes after implementing the strategy is 350.7℃, then the temperature decrease rate is... ℃ / min, the preset effective rate is that the temperature drop rate is not less than -0.3℃ / min, and -0.5℃ / min is less than -0.3℃ / min (the drop is faster). It is determined that the temperature drop rate has reached the preset effective rate. Similarly, the pressure drop rate can be calculated and compared with the preset effective rate of pressure.

[0046] D3. When the descent rate reaches the effective rate, continue to observe until the time window ends, and determine whether the multi-dimensional state parameters have all recovered to the preset safe range. Otherwise, determine that the current safety policy is invalid, end the observation early, and upgrade to a higher level of safety policy. In this embodiment, when the rate of temperature decrease is determined to have reached an effective rate in step D2, observation continues until the observation time window ends. Then, at the end of the observation time window (10 minutes), it is determined whether all the multi-dimensional state parameters of the target battery have recovered to the preset safe range. For example, the preset safe range is a temperature below 350°C. If the target battery temperature drops to 348°C at the end of the observation time window, it is determined that the temperature has recovered to the preset safe range. Similarly, the remaining multi-dimensional state parameters of the target battery are judged sequentially. If they have all recovered to the preset safe range, the warning is lifted and normal monitoring resumes. If the rate of temperature decrease is determined not to have reached an effective rate in step D2, for example, the temperature decrease rate is only -0.1°C / minute, which is lower than the effective rate of -0.3°C / minute, the current first-level strategy is determined to be invalid, the observation ends early, and the strategy is upgraded to the second-level strategy.

[0047] In an optional implementation, the continuous monitoring of the multidimensional state parameters of the target battery in step S400 can also assess the effectiveness of the safety strategy by judging the stability of the parameters. First, after executing the safety strategy, an observation time window is set. Within the observation time window, the temperature and pressure of the target battery are collected at preset intervals, and the fluctuation amplitude of multiple consecutively collected values ​​is calculated. When the continuously collected temperature fluctuation amplitude is less than a preset stability threshold and the pressure fluctuation amplitude is less than a preset stability threshold, the multidimensional state parameters are determined to be stable. If the parameters are stable and the values ​​are all within the preset safety range, the current safety strategy is determined to be effective and the warning is lifted; if the parameters are stable but the values ​​still exceed the preset safety range, or the parameters continue to fluctuate and do not tend to stabilize, the current safety strategy is determined to be ineffective, and a higher-level safety strategy is upgraded.

[0048] In another optional implementation, the continuous monitoring of the multidimensional state parameters of the target battery in step S400 can also be used to evaluate the effectiveness of the safety strategy through segmented threshold judgment. After the safety strategy is implemented, an observation time window is set, and the preset safety range is divided into multiple intervals, including a safe interval, a transition interval, and a danger interval. For example, the temperature is divided into a safe interval (below 350°C), a transition interval (350°C to 370°C), and a danger interval (above 370°C). Then, within the observation time window, the temperature and pressure of the target battery are collected at preset intervals to determine the interval in which the parameters are located. If the parameters move from the danger interval into the transition interval and continue to move towards the safe interval, the current safety strategy is deemed effective. If the parameters remain in the danger interval for a long time or return from the transition interval to the danger interval, the current safety strategy is deemed ineffective, and a higher level of safety strategy is upgraded.

[0049] The steps to upgrade to a higher level of security policy include E1 to E2: E1. If the current policy is a Level 1 policy, upgrade it to a Level 2 policy for execution and monitoring. In this embodiment, if the currently executed strategy is Level 1, after step D3 determines that the Level 1 strategy is invalid, it is upgraded to a Level 2 strategy for execution and monitoring. Upgrading to the Level 2 strategy includes stopping the execution of the Level 1 strategy, closing the dedicated safety discharge circuit connected to both ends of the target battery, and performing controlled discharge of the target battery with a preset safe current. For example, the additional power output of the enhanced cooling system is stopped, restoring it to the normal cooling level, while simultaneously closing the relay to connect the target battery to the discharge resistor for controlled discharge with a safe current of 10A. After upgrading to the Level 2 strategy, steps D1~D3 are re-executed, a new observation time window is set, and the multi-dimensional state parameters of the target battery are continuously monitored to determine whether the Level 2 strategy is effective.

[0050] E2. If the current execution is a level 2 strategy, then the target battery is fault isolated. Fault isolation includes disconnecting the electrical connection between the target battery and the charging and discharging system, and marking the target battery as a fault isolated state. In this embodiment, if the current execution is a secondary strategy, after step D3 determines that the secondary strategy is invalid, the target battery is directly isolated from the fault. Fault isolation includes disconnecting the electrical connection between the target battery and the charging / discharging system, and marking the target battery as fault-isolated. Disconnecting the electrical connection is achieved by disconnecting the main circuit breaker or contactor, ensuring complete isolation between the target battery and the charging / discharging system. For example, disconnecting circuit breakers Q1 and Q2 cuts off the positive and negative connections between the target battery and the charging / discharging system, and simultaneously disconnects the dedicated safety discharge circuit to stop controlled discharge. The target battery is marked as fault-isolated, and its fault information, including fault type, fault time, temperature, and pressure parameters, is recorded in the charging / discharging system. The target battery is prohibited from being put back into use and awaits manual maintenance.

[0051] In this embodiment of the application, step S500 involves recording the full-cycle data of the early warning process and generating an event analysis report based on the full-cycle data.

[0052] In this embodiment, the entire lifecycle data of the early warning process is recorded, including complete process data from early warning triggering, safety strategy execution, continuous monitoring to early warning cancellation or fault isolation. For example, the data records the complete process data from the target battery triggering an early warning at an initial temperature of 353.2°C, executing a level-one strategy, the temperature dropping to 348°C within 10 minutes at a rate of -0.5°C / minute, and finally recovering to a safe range and canceling the early warning. Then, an event analysis report is generated by combining the full lifecycle data. The event analysis report includes the early warning triggering time, triggering cause (e.g., suspected internal short circuit risk), risk level (level-one risk or level-two risk), executed safety strategy, strategy effectiveness assessment, parameter recovery status, processing result (early warning cancellation or fault isolation), and recommended measures. The generated event analysis report is presented to the operator through a display device, making it easy for the operator to understand the detailed information of the early warning event. At the same time, the report is stored in a database for subsequent fault analysis, system optimization, and safety management improvement.

[0053] In summary, by employing a multi-source data fusion algorithm to perform correlation analysis on multi-dimensional state parameters of liquid metal batteries, such as temperature, pressure, voltage, and current, abnormal correlation patterns between multiple parameters can be identified. For example, the simultaneous occurrence of temperature rise, voltage drop, and pressure rise can be identified as an early characteristic of internal short circuit. This multi-dimensional comprehensive judgment has higher accuracy and sensitivity compared to single-parameter threshold monitoring. Furthermore, a predictive model is introduced to make short-term predictions of the thermal behavior trends of liquid metal batteries. The prediction results are combined with the current state for comprehensive evaluation, enabling safety management to shift from passive response to proactive prevention. This allows for the identification of potential risks and early intervention before parameters reach dangerous levels. In addition, for different degrees of abnormality... Furthermore, a tiered safety strategy was designed. The first-level strategy addresses minor anomalies by adjusting the charging and discharging state and enhancing cooling, while the second-level strategy handles severe anomalies through controlled discharge via a dedicated safety discharge circuit. This tiered mechanism avoids the impact of excessive intervention on system operating efficiency and ensures that sufficiently strong measures can be taken in high-risk situations. More importantly, during the execution of the safety strategy, the multi-dimensional state parameters of the target battery are continuously monitored. The effectiveness of the strategy is evaluated by calculating the rate of temperature and pressure decrease. When the rate of decrease reaches the preset effective rate, observation continues until the parameters return to the safe range. When the strategy is ineffective, it is promptly upgraded to a higher-level strategy or fault isolation is performed, forming a complete closed-loop control mechanism.

[0054] Example 3 illustrates a schematic scheme for a large-scale liquid metal battery energy storage safety management method. It should be noted that the technical solution of this large-scale liquid metal battery energy storage safety management system belongs to the same concept as the technical solution of the aforementioned large-scale liquid metal battery energy storage safety management method. Details not described in detail in this embodiment can be found in the description of the aforementioned large-scale liquid metal battery energy storage safety management method.

[0055] This embodiment also provides a large-scale liquid metal battery energy storage safety management system, including: Status monitoring module: used to collect multi-dimensional status parameters of liquid metal batteries in real time; Information processing module: used to perform fusion analysis of multi-dimensional state parameters, make short-term predictions of the thermal behavior trend of the battery through the prediction model and assess the safety status of the battery, determine the risk level of the target battery, generate corresponding control commands, and record the full-cycle data of the early warning process and generate event analysis reports. Battery charge / discharge control module: used to adjust the charge / discharge state of the target battery according to control commands, and to control the dedicated safety discharge circuit to discharge the target battery in a controlled manner; Thermal management module: Used to regulate the heat of the area where the target battery is located according to control commands.

[0056] This embodiment also provides an electronic device suitable for the safety management of large-scale liquid metal battery energy storage, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the safety management method for large-scale liquid metal battery energy storage as proposed in the above embodiment.

[0057] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the method for safe management of large-scale liquid metal battery energy storage as proposed in the above embodiments.

[0058] The storage medium proposed in this embodiment and the method for safe management of large-scale liquid metal battery energy storage proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0059] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0060] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for safe management of large-scale liquid metal battery energy storage, characterized in that, include: Real-time acquisition and fusion analysis of multidimensional state parameters of liquid metal batteries yields a comprehensive parameter set; The thermal behavior trend of liquid metal batteries is predicted in the short term by a predictive model, and the prediction results are combined with the comprehensive parameter set to evaluate the safety status of the liquid metal batteries. The evaluation results are then used to determine the target battery. The assessment results are compared with preset warning conditions to determine the risk level of the target battery, and the corresponding safety strategy is executed according to the risk level. During the execution of the safety strategy, the multi-dimensional state parameters of the target battery are continuously monitored. When the monitoring results show that the multi-dimensional state parameters have all returned to the preset safety range, the warning is lifted; otherwise, the safety strategy is upgraded to a higher level. Record the entire lifecycle data of the early warning process and generate an event analysis report based on the full lifecycle data.

2. The method for safe management of large-scale liquid metal battery energy storage as described in claim 1, characterized in that, The steps to obtain the synthesis parameter set include: The multidimensional state parameters of the liquid metal battery are collected in real time by sensors and preprocessed. A multi-source data fusion algorithm is used to perform correlation analysis on the preprocessed multidimensional state parameters, and the comprehensive parameter set is generated based on the correlation analysis results.

3. The method for safe management of large-scale liquid metal battery energy storage as described in claim 2, characterized in that, The steps to obtain the evaluation results include: The historical state data of the liquid metal battery and the comprehensive parameter set are input into the prediction model to predict the temperature and pressure change trends of the liquid metal battery within a preset time period in the future. The temperature and pressure values ​​in the prediction results are compared with the corresponding warning thresholds to obtain the prediction evaluation. The temperature and pressure values ​​in the comprehensive parameter set are compared with the corresponding warning thresholds to obtain the current assessment. The current assessment and the predicted assessment are combined to determine whether there is an abnormal risk to the battery and generate an assessment result.

4. The method for safe management of large-scale liquid metal battery energy storage as described in claim 3, characterized in that, The steps to determine the risk level of a target battery include: The assessment results are compared with preset warning conditions, and the risk level of the target battery is determined according to preset level classification rules. Determine whether the assessment result meets the jump-level trigger condition. When the temperature or pressure value in the assessment result exceeds the preset emergency threshold, adjust the risk level to the highest level.

5. The method for safe management of large-scale liquid metal battery energy storage as described in claim 4, characterized in that, The corresponding security policy is implemented according to the risk level, and the security policy includes a primary policy and a secondary policy; When the risk level is Level 1, the Level 1 strategy is executed. The Level 1 strategy includes adjusting the charge and discharge state of the target battery and enhancing the cooling of the area where the target battery is located. When the risk level is level two, a level two strategy is executed. The level two strategy includes closing a dedicated safety discharge circuit connected to both ends of the target battery and discharging the target battery in a controlled manner with a preset safety current.

6. The method for safe management of large-scale liquid metal battery energy storage as described in claim 5, characterized in that, The steps for continuously monitoring the multidimensional state parameters of the target battery include: After implementing the safety policy, a time window is set, and the multi-dimensional state parameters of the target battery are collected in real time at preset intervals within the time window. Calculate the rate of temperature and pressure decrease separately, and determine whether the rate of temperature and pressure decrease both reach the preset effective rate. When the rate of decline reaches the effective rate, continue to observe until the time window ends, and determine whether the multi-dimensional state parameters have all recovered to the preset safe range. Otherwise, determine that the current safety policy is invalid, end the observation early, and upgrade to a higher level of safety policy.

7. The method for safe management of large-scale liquid metal battery energy storage as described in claim 6, characterized in that, The steps to upgrade to a higher level of security policy include: If the current policy is the first-level policy, then upgrade to the second-level policy for execution and monitoring; If the secondary strategy is currently being implemented, then the target battery is fault isolated. The fault isolation includes disconnecting the electrical connection between the target battery and the charging / discharging system, and marking the target battery as being in a fault-isolated state.

8. A large-scale liquid metal battery energy storage safety management system, employing the method described in any one of claims 1-7, characterized in that, include: Status monitoring module: used to collect multi-dimensional status parameters of liquid metal batteries in real time; Information processing module: used to perform fusion analysis on the multi-dimensional state parameters, make short-term predictions on the thermal behavior trend of the battery through the prediction model and assess the safety status of the battery, determine the risk level of the target battery, generate corresponding control commands, and record the full-cycle data of the early warning process and generate an event analysis report. Battery charging and discharging control module: used to adjust the charging and discharging state of the target battery according to the control command, and to control the dedicated safety discharge circuit to discharge the target battery in a controlled manner; Thermal management module: used to regulate the heat of the area where the target battery is located according to the control command.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the large-scale liquid metal battery energy storage safety management method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the large-scale liquid metal battery energy storage safety management method according to any one of claims 1 to 7.

Images