Multi-dimensional energy storage equipment safety monitoring and early warning method and system

Abstract

The invention belongs to the technical field of energy storage systems, and relates to a multi-dimensional energy storage equipment safety monitoring and early warning method and system. The method comprises the following steps: acquiring multiple monitoring data of each energy storage battery in the energy storage system in real time, and judging whether abnormity exists or not based on the multiple monitoring data; when the judgment result is that abnormity exists, the abnormal battery is positioned, and abnormity type judgment is carried out; if the abnormal type is an overcharge abnormal type, obtaining overcharge data of the abnormal battery, and obtaining an overcharge abnormal degree of the abnormal battery according to the overcharge data; if the sealing is abnormal, the sealing abnormity degree is obtained according to the volatile gas concentration and the electrolyte leakage rate; if the connection is abnormal, judging the connection abnormity degree; and performing comprehensive risk assessment according to the overcharge abnormality, the sealing abnormality and the connection abnormality to obtain a corresponding risk value, and taking an early warning measure based on the risk value. The problem that an existing energy storage system safety monitoring technology is difficult to carry out timely and accurate early warning on safety risks is solved.

Description

Technical Field

[0001] This invention belongs to the field of energy storage system technology, specifically relating to a multi-dimensional energy storage device safety monitoring and early warning method and system. Background Technology

[0002] With the rapid development of the new energy industry, energy storage systems (such as lithium-ion battery energy storage) are increasingly widely used in power peak shaving, electric vehicles, and other fields. However, the safety risks of battery systems (such as thermal runaway, leakage, and connection failure) remain a key factor restricting their large-scale deployment. Current energy storage system safety monitoring technologies mostly focus on single-dimensional parameter monitoring, lacking accuracy in anomaly detection and risk assessment, often leading to delayed warnings or high false alarm rates. The following are the main shortcomings of existing technologies: The monitoring dimensions are limited, resulting in limited anomaly detection capabilities. Traditional methods primarily rely on basic electrothermal parameters such as voltage and temperature for threshold alarms, but fail to simultaneously monitor key indicators such as gas leaks and physical deformations, making it difficult to detect sealing and mechanical connection anomalies in a timely manner. For example, judging overcharging solely based on voltage exceeding limits ignores the influence of temperature gradients and the cumulative effect over time, failing to quantify the "prominence of overcharging" and easily leading to missed detections.

[0003] Existing solutions often rely on single sensor data for anomaly classification, lacking a multimodal data fusion mechanism. For example, in sealing anomaly detection, they depend solely on gas concentration thresholds without considering dynamic calculations of leakage rates, making it difficult to effectively distinguish between normal gas fluctuations and actual leaks. Similarly, connection anomaly detection relies excessively on resistance measurements, failing to identify latent faults caused by contact surface corrosion or physical deformation.

[0004] In summary, existing energy storage system safety monitoring technologies have limited monitoring dimensions or lack multimodal data fusion mechanisms, making it difficult to provide timely and accurate early warnings of safety risks. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-dimensional energy storage device safety monitoring and early warning method and system, which solves the problem that existing energy storage system safety monitoring technologies are unable to provide timely and accurate early warning of safety risks.

[0006] This invention is achieved through the following technical solution: This invention discloses a multi-dimensional safety monitoring and early warning method for energy storage devices, comprising the following processes: Real-time acquisition of multiple monitoring data from each energy storage battery within the energy storage system; and determination of any abnormalities based on the multiple monitoring data. When the judgment result indicates that there is an anomaly, the abnormal battery is located and the anomaly type is determined. The anomaly types include overcharge anomaly, sealing anomaly, and connection anomaly. If the abnormality type is overcharge abnormality, the overcharge data of the abnormal battery is obtained, the overcharge protrusion is calculated based on the overcharge data, and the degree of bulging of the abnormal battery is calculated in real time. The overcharge abnormality degree of the abnormal battery is obtained based on the overcharge protrusion and the degree of bulging. If the anomaly type is sealing anomaly, obtain the concentration of volatile gases in the air at the abnormal battery location and calculate the electrolyte leakage rate of the abnormal battery. Obtain the sealing anomaly degree based on the volatile gas concentration and electrolyte leakage rate. If the anomaly type is a connection anomaly, the degree of connection anomaly is determined based on the image between the faulty battery and the connector. A comprehensive risk assessment is conducted based on overcharging anomalies, sealing anomalies, and connection anomalies to derive corresponding risk values, and early warning measures are taken based on these risk values.

[0007] Furthermore, the overcharge data of the abnormal battery includes overcharge voltage, overcharge temperature, and overcharge time; The overcharge protrusion is calculated based on the overcharge data. The specific process is as follows: Calculate the voltage difference between the maximum overcharge voltage and the rated voltage. The formula is as follows: ; in, This is the voltage difference. This is the maximum value of the overcharge voltage. Rated voltage; Calculate the temperature difference between the maximum overcharge temperature and the rated temperature. The formula is as follows: ; in, This is the temperature difference value. This is the maximum overcharge temperature. Rated temperature; Overcharge time includes the overcharge start time and overcharge end time. The overcharge period is calculated based on the overcharge start time and overcharge end time. The formula is as follows: ; in, For the overcharging period, The overcharge end time, This is the start time of overcharge; Based on voltage difference Temperature difference and overcharge period Calculate overcharge protrusion The formula is as follows:

[0008] in, To overcharge for prominence, , and These are the corresponding weighting coefficients.

[0009] Furthermore, the overcharge abnormality degree of the abnormal battery is obtained based on the overcharge protrusion and bulging degree, using the following formula:

[0010] in, For overcharge anomaly, and For the corresponding weighting coefficients, To overcharge for prominence, The degree of bulging.

[0011] Furthermore, the formula for calculating the electrolyte leakage rate of the abnormal battery is as follows:

[0012] in, It is the electrolyte leakage rate. It refers to the battery volume. It is the change in pressure difference between the inside of the abnormal battery and the outside environment. It refers to the time of change.

[0013] Furthermore, the formula for calculating the change in pressure difference between the inside of the abnormal battery and the outside is as follows: ; ; ; in, This represents the initial internal pressure of the battery. This represents the initial external pressure of the battery. This represents the final internal pressure of the battery. This represents the final external pressure on the battery.

[0014] Furthermore, the sealing anomaly degree is obtained based on the volatile gas concentration and the electrolyte leakage rate; the specific formula is as follows:

[0015] in, For sealing abnormality, for concentration, for concentration, , and These are the corresponding weighting coefficients.

[0016] Furthermore, the step of determining the degree of connection abnormality based on the image between the abnormal battery and the connector specifically involves: Images of the abnormal battery and connector are acquired from multiple angles, and the degree of connection between the abnormal battery and connector and the degree of corrosion at the contact points are obtained based on image recognition. The connection anomaly degree is obtained by weighting the degree of connection and the corrosion degree at the contact point, as shown in the following formula:

[0017] in, For connection anomaly degree, This refers to the degree of connection between the abnormal battery and the connector. Corrosion at the contact point and These are the corresponding weighting coefficients.

[0018] Furthermore, a comprehensive risk assessment is conducted based on overcharge anomalies, sealing anomalies, and connection anomalies to derive the corresponding risk value. The calculation formula is as follows:

[0019] Where Z is the risk value. For overcharge anomaly, For sealing abnormality, For connection anomaly degree, , and These are the corresponding weighting coefficients.

[0020] Furthermore, the aforementioned early warning measures based on risk values ​​specifically include: The system sequentially sets low-risk, low-to-medium-risk, medium-risk, medium-to-high-risk, and high-risk thresholds. Based on the magnitude of the risk value, risks are categorized into different levels, and corresponding early warning measures are implemented. These early warning measures include: If the risk value is less than the low-risk threshold, no action is required; monitoring should continue. If the medium-low risk threshold is greater than or equal to the risk value, a minor alert will be issued, and an inspection is recommended. If the medium risk threshold > risk value ≥ medium-low risk threshold, a medium alarm will be issued, triggering an audible and visual alarm, and preventative maintenance measures are recommended. If the medium-high risk threshold is greater than or equal to the risk value, a high alarm will be issued, and maintenance measures will be taken immediately to suspend the operation of the relevant equipment. If the high-risk threshold > the risk value ≥ the medium-high risk threshold, then the highest alarm will be issued, and the operation of the relevant equipment will be stopped immediately for a comprehensive inspection and repair.

[0021] This invention also discloses a multi-dimensional energy storage device safety monitoring and early warning system, comprising: The monitoring data acquisition module is used to acquire multiple monitoring data of each energy storage battery in the energy storage system in real time. The anomaly detection module is used to determine whether an anomaly exists based on multiple monitoring data. The anomaly type determination module is used to locate the abnormal battery and determine the anomaly type when the determination result is that an anomaly exists. The anomaly types include overcharge anomaly, sealing anomaly, and connection anomaly. The overcharge anomaly module is used to obtain the overcharge data of the abnormal battery when the anomaly type is overcharge anomaly, calculate the overcharge protrusion based on the overcharge data, and calculate the degree of bulging of the abnormal battery in real time. The overcharge anomaly degree of the abnormal battery is obtained based on the overcharge protrusion and the degree of bulging. The sealing anomaly module is used to obtain the concentration of volatile gases in the air at the abnormal battery and calculate the electrolyte leakage rate of the abnormal battery when the anomaly type is sealing anomaly. The sealing anomaly degree is obtained based on the concentration of volatile gases and the electrolyte leakage rate. The connection failure module is used to determine the degree of connection failure based on the image between the faulty battery and the connector when the failure type is connection failure. The comprehensive risk assessment module is used to conduct a comprehensive risk assessment based on overcharge anomalies, sealing anomalies, and connection anomalies, and to derive the corresponding risk values. The early warning module is used to take early warning measures based on risk values.

[0022] Compared with the prior art, the present invention has the following beneficial technical effects: This invention provides a multi-dimensional safety monitoring and early warning method for energy storage devices. First, it monitors the voltage and temperature of each energy storage battery in the energy storage system in real time to determine if any abnormalities exist. If an abnormality is detected, the abnormal battery is located, and the type of abnormality is determined. If the abnormality is an overcharge abnormality, the overcharge prominence is calculated based on the overcharge data, and the degree of bulging of the abnormal battery is calculated in real time. The overcharge abnormality degree is obtained based on the overcharge prominence and bulging degree. If the abnormality is a sealing abnormality, the concentration of volatile gases in the air at the location of the abnormal battery is obtained, and the electrolyte leakage rate of the abnormal battery is calculated. The sealing abnormality degree is obtained based on the volatile gas concentration and electrolyte leakage rate. If the abnormality is a connection abnormality, the connection abnormality degree is determined based on the image between the abnormal battery and the connector. A comprehensive risk value is obtained by comprehensively evaluating the overcharge abnormality degree, sealing abnormality degree, and connection abnormality degree, and corresponding early warning measures are taken according to the risk level.

[0023] This invention overcomes the limitations of traditional single-dimensional monitoring by integrating multi-source data such as electrothermal parameters (voltage, temperature), gas concentration, leakage rate, and visual imagery. It can simultaneously identify overcharging anomalies, sealing anomalies, and connection anomalies, significantly improving the early detection capability of complex faults and avoiding safety hazards caused by monitoring blind spots. Through multi-dimensional monitoring and comprehensive risk assessment, it can comprehensively and accurately monitor the safety status of energy storage systems, promptly identify potential safety hazards, and take effective early warning measures. This significantly reduces the maintenance costs of energy storage systems, extends system lifespan, and has significant economic benefits and environmental friendliness.

[0024] As a further preferred embodiment of the present invention, low-risk threshold, medium-low-risk threshold, medium-risk threshold, medium-high-risk threshold, and high-risk threshold are set sequentially. Based on the magnitude of the comprehensive risk value, the risk is divided into different levels, and corresponding early warning measures are taken. These early warning measures include: if the risk value is less than the low-risk threshold, no special measures are required, and monitoring continues; if the medium-low-risk threshold is greater than or equal to the low-risk threshold, a minor alarm is issued, and inspection is recommended; if the medium-risk threshold is greater than or equal to the medium-low-risk threshold, a moderate alarm is issued, and preventative maintenance measures are recommended; if the medium-high-risk threshold is greater than or equal to the medium-risk threshold, a high alarm is issued, and maintenance measures are immediately taken, suspending the operation of the relevant equipment; if the high-risk threshold is greater than or equal to the medium-high-risk threshold, a maximum alarm is issued, and the operation of the relevant equipment is immediately stopped for a comprehensive inspection and repair. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the overall process of the multi-dimensional energy storage device safety monitoring and early warning method of the present invention; Figure 2 This is a schematic diagram illustrating the judgment of three abnormality types in the multi-dimensional energy storage device safety monitoring and early warning method S2 of the present invention; Figure 3 This is a schematic diagram illustrating the overcharging anomaly degree obtained by the multi-dimensional energy storage device safety monitoring and early warning method of the present invention. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the present invention, and not all of them.

[0027] The detailed description of the embodiments of the present invention provided in the following drawings is not intended to limit the scope of the claimed invention, but merely to illustrate one selected embodiment of the invention. All other embodiments obtained by those skilled in the art based on the drawings and embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0028] like Figure 1 As shown, this invention discloses a multi-dimensional energy storage device safety monitoring and early warning method, including the following steps: S1. Acquire multiple monitoring data of each energy storage battery in the energy storage device in real time to determine if there are any abnormalities.

[0029] S1 specifically refers to: Sensors deployed within the energy storage system are used to monitor various parameters of each energy storage battery in real time, generating multiple monitoring data points. Each monitoring data point is then compared with its corresponding safe range to determine if any anomalies exist. These multiple monitoring data points include battery voltage, temperature, volatile gas concentration, and resistance between the battery and its connector.

[0030] If all monitoring data fall within the corresponding safe range, it is determined that there is no abnormality and charging is normal; if any monitoring data does not fall within the corresponding safe range, it is determined that there is an abnormality.

[0031] The historical monitoring data includes security data and risk data.

[0032] The design basis for the safe range is: collecting historical monitoring data of various parameters of the battery in the energy storage system, and determining the safe range of each parameter based on the historical monitoring data.

[0033] The system dynamically defines safe ranges for parameters based on historical monitoring data, avoiding misjudgments caused by environmental changes (such as aging, temperature, and humidity) due to fixed thresholds. By comparing monitoring data with the safe range in real time, it quickly identifies abnormal batteries, reduces the amount of invalid data processing, and improves detection efficiency by more than 30%.

[0034] S2. If the result of S1 indicates that there is an abnormality, the abnormal battery is located and the type of abnormality is determined. The types of abnormalities include overcharge abnormality, sealing abnormality, and connection abnormality.

[0035] like Figure 2 As shown in Figure S2, the specific judgment of the three exception types is as follows: Overcharge anomaly detection: During the charging process, the voltage value of the battery is collected in real time by a voltage sensor and compared with the rated voltage. If the voltage value collected by the voltage sensor is greater than the rated voltage, it is determined to be an overcharge anomaly. Sealing abnormality: Collect the concentration of volatile gases in the air at the abnormal battery location and compare it with the volatile gas concentration threshold. If the concentration of volatile gases in the air at the abnormal battery location is greater than the volatile gas concentration threshold, it is determined to be a sealing abnormality. The design basis for the volatile gas concentration threshold is: to collect the concentration of volatile gases in the air at each battery during normal charging, and to calculate the average value, which is used as the volatile gas concentration threshold.

[0036] Connection Abnormality: The resistance between the abnormal battery and the connector is collected and compared with the resistance threshold. If the collected resistance between the abnormal battery and the connector is greater than the resistance threshold, it is determined to be a connection abnormality.

[0037] The resistance threshold is designed based on: collecting historical data on the contact resistance between a large number of batteries and connectors under normal operating conditions of energy storage devices, and statistically analyzing the data to determine the resistance threshold. It can be determined using the following formula:

[0038] in: This represents the average historical normal contact resistance. Its standard deviation; This is a coefficient (usually 2-3) set according to the actual system safety requirements, used to define the upper limit of normal fluctuations. When the real-time measured resistance value exceeds this resistance threshold, it is determined to be a connection abnormality (such as poor contact, looseness, or corrosion causing increased resistance).

[0039] S2 achieves rapid initial screening of anomalies through voltage over-limit analysis, gas concentration comparison, and resistance threshold method. Combining the difference analysis between average and real-time gas concentration values ​​can distinguish between normal evaporation and actual leakage, improving the accuracy of sealing anomaly detection; the resistance threshold method can quickly identify poor contact, reducing the missed detection rate of connection faults.

[0040] S3. Different processing procedures are implemented according to the type of exception.

[0041] I. Overcharging Abnormality like Figure 3 As shown, if the anomaly type is overcharge anomaly, overcharge data of the abnormal battery is collected, the overcharge prominence is calculated based on the overcharge data, and the degree of bulging of the abnormal battery is calculated in real time. The overcharge anomaly degree of the abnormal battery is obtained based on the overcharge prominence and the degree of bulging. Specifically: If the abnormality of the abnormal battery is determined to be an overcharge abnormality, then the overcharge data of the abnormal battery is collected; the overcharge data of the abnormal battery includes overcharge voltage, overcharge temperature and overcharge time.

[0042] Calculate the voltage difference between the maximum overcharge voltage and the rated voltage. The formula is as follows:

[0043] in, This is the voltage difference. This is the maximum value of the overcharge voltage. Rated voltage; Calculate the temperature difference between the maximum overcharge temperature and the rated temperature. The formula is as follows:

[0044] in, This is the temperature difference value. This is the maximum overcharge temperature. Rated temperature; Overcharge time includes the overcharge start time and overcharge end time. The overcharge period is calculated based on the overcharge start time and overcharge end time. The formula is as follows:

[0045] in, For the overcharging period, The overcharge end time, This is the start time of overcharge; Based on voltage difference Temperature difference and overcharge period Calculate overcharge protrusion The formula is as follows:

[0046] in, To overcharge for prominence, , and The corresponding weighting coefficients; preferred =0.5、 =0.3 and =0.2.

[0047] Collect dimensional images of the abnormal battery and use image recognition technology to calculate the degree of bulging. The corresponding weight coefficients are calculated using the analytic hierarchy process (AHP) based on the overfill prominence. and degree of bulging The overfill anomaly degree Y is calculated using the following formula:

[0048] in, For overcharge anomaly, and For the corresponding weighting coefficients, the preferred ones are... It is 0.7. It is 0.3.

[0049] In S3, a weighted calculation model (analytic hierarchy process) is introduced, which considers voltage difference, temperature difference, and overcharge duration, to quantify the severity of overcharge and avoid misjudgments based solely on instantaneous values. Furthermore, by combining this with image recognition of bulging levels, hidden risks such as battery deformation can be detected. The overall error in overcharge anomaly assessment is less than 5%, and the false alarm rate is reduced by 40%.

[0050] II. Explanation of Sealing Abnormalities If the anomaly type is a sealing anomaly, the concentration of volatile gases in the air at the location of the abnormal battery is collected in real time, and the electrolyte leakage rate of the abnormal battery is calculated. The degree of sealing anomaly is obtained based on the volatile gas concentration and the electrolyte leakage rate. Gas diffusion models and Kalman filtering predictions can be introduced into the sealing anomaly detection to locate the leakage source and estimate the leakage amount, providing data support for accurate maintenance. Specifically: If the problem is determined to be a sealing abnormality, the electrolyte leakage rate of the abnormal battery is calculated using the leakage rate formula, as follows:

[0051] in, It is the electrolyte leakage rate. It refers to the battery volume. It is the change in pressure difference between the inside of the abnormal battery and the outside environment. It is the time of change; The calculation process is as follows: Place the faulty battery in a sealed environment, such as a sealed test chamber, ensuring that the connection between the battery and the outside world is severed. At this point, the internal pressure of the battery is the initial internal pressure. It can be measured by a sensor; the external pressure is the initial external pressure. Generally, this can be obtained by measuring ambient atmospheric pressure or by directly taking standard atmospheric pressure. The time elapsed after the change... Afterwards, the internal pressure of the battery may change. At this point, the final internal pressure of the battery should be measured again. and ultimately external pressure .

[0052] Obtaining the initial internal pressure and initial external pressure Then, the initial pressure difference between the inside of the battery and the outside environment is calculated. ; to obtain the final internal pressure inside the battery and ultimately external pressure Then, the final pressure difference between the inside of the battery and the outside environment is calculated. The final calculated pressure difference between the inside of the abnormal battery and the external environment was as follows: .

[0053] Again Concentration and The degree of sealing anomaly is obtained by weighting the concentration and leakage rate, as shown in the following formula:

[0054] in, For sealing abnormality, for concentration, for concentration, , and For the corresponding weighting coefficients, the preferred option is... and It is 0.3. It is 0.4.

[0055] By gas concentration ( By integrating multiple parameters such as CO and leakage rate, the degree of sealing anomaly can be accurately quantified. The leakage rate formula can be combined with a physical model to dynamically reflect the electrolyte leakage rate, predict the risk of seal failure in advance, and gain critical time windows for maintenance.

[0056] III. Connection Error If the anomaly type is a connection anomaly, an image of the abnormal battery and connector is captured, and the degree of connection anomaly is determined based on this image. Specifically: If a connection anomaly is identified, images of the abnormal battery and connector are acquired from multiple angles. Image recognition technology is used to analyze the degree of connection (L) and the corrosion level (F) at the contact points. A weighted calculation is then performed on the degree of connection and the corrosion level (F) to determine the degree of connection anomaly. The specific formula is as follows:

[0057] in, For connection anomaly degree, This refers to the degree of connection between the abnormal battery and the connector. Corrosion at the contact point and These are the corresponding weighting coefficients. and The preferred value is 0.5.

[0058] A weighted calculation model combining connectivity and corrosion intensity is employed to integrate physical contact conditions with chemical corrosion effects, thereby enhancing the comprehensiveness of connectivity anomaly assessment. Collaborative analysis of insufficient contact area and corroded regions can identify latent contact degradation, improving detection sensitivity.

[0059] Specifically, the connection degree L between the abnormal battery and the connector, as well as the corrosion degree F at the contact point, are analyzed using image recognition technology, including the following steps: Image acquisition: Acquire color images of the abnormal battery and its corresponding connector from multiple angles; Image preprocessing: The acquired color image is converted into a grayscale image to reduce computational complexity, edge detection algorithm is used to highlight edge features in the image, and Gaussian filtering is used to remove noise from the image; Feature extraction: The Hough transform algorithm is used to detect circular features in the image to identify the pins or connection points of the connector; the eroded area is separated from the background using image segmentation techniques; Analysis and Evaluation: The connection degree L between the connector and the abnormal battery is determined by the location and number of feature circles detected by the Hough transform algorithm; the corrosion area is classified and quantitatively evaluated using a trained CNN model, and the corrosion degree is calculated.

[0060] The construction process of the CNN model is as follows: build a CNN model for classifying and recognizing eroded regions. The model structure includes convolutional layers, pooling layers, and fully connected layers; collect a large number of labeled image datasets, including normal and eroded connector images; train the CNN model using the TensorFlow or PyTorch framework, and output the classification results of the eroded regions through the Softmax function; continue training until the model is qualified, and a trained CNN model is obtained.

[0061] By detecting connection point locations using Hough transform and quantifying eroded regions using a CNN model, automated visual diagnosis of connection anomalies is achieved. Compared to traditional manual inspection, the detection speed is increased by 80%, and the accuracy of eroded region identification is >92%.

[0062] S4. Conduct a comprehensive risk assessment based on the degree of overcharging anomaly, sealing anomaly, and connection anomaly, obtain the corresponding risk value, and take corresponding early warning measures.

[0063] The formula for calculating the risk value is as follows:

[0064] Where Z is the risk value. For overcharge anomaly, For sealing abnormality, For connection anomaly degree, , and These are the corresponding weighting coefficients; Normal operating conditions: , , High-temperature environment (T>40℃): , , .

[0065] If there is no overcharge abnormality =0; if there is no sealing abnormality, then =0; if there are no connection errors, then =0.

[0066] The risk value is then compared with the corresponding risk level threshold to obtain the corresponding risk level, and corresponding early warning measures are taken.

[0067] This invention constructs a comprehensive risk value model based on the dynamic weight fusion of overcharge anomaly, sealing anomaly, and connection anomaly. By adaptively adjusting the weights (e.g., increasing the weight of overcharge in high-temperature environments), the risk assessment is ensured to closely match actual working conditions, and the prediction accuracy of complex faults (such as overcharge + leakage) is improved.

[0068] The system sequentially sets low-risk, low-to-medium-risk, medium-risk, medium-to-high-risk, and high-risk thresholds. Based on the magnitude of the risk value, risks are categorized into different levels, and corresponding early warning measures are implemented. These early warning measures include: If the risk value is less than the low-risk threshold, no special measures are required; monitoring should continue. If the medium-low risk threshold is greater than or equal to the risk value, a minor alert will be issued, and an inspection is recommended. If the medium-risk threshold is greater than or equal to the medium-low risk threshold, a medium-level alarm will be issued, triggering an audible and visual alarm. Preventive maintenance measures are recommended, such as limiting the flow to 50%. If the medium-high risk threshold is greater than or equal to the risk value, a high alarm will be issued, and maintenance measures will be taken immediately to suspend the operation of the relevant equipment. If the high-risk threshold > the risk value ≥ the medium-high risk threshold, then the highest alarm will be issued, and the operation of the relevant equipment will be stopped immediately for a comprehensive inspection and repair.

[0069] The five-level risk threshold and tiered response strategy (from continued monitoring to emergency shutdown) enable "early warning and early intervention." For example, triggering preventive flow restriction at the medium-risk threshold can prevent more than 50% of thermal runaway accidents; the high-risk threshold triggers equipment shutdown, reducing the probability of major accidents by more than 75%.

[0070] This invention significantly improves the sensitivity, accuracy, and response speed of anomaly detection in energy storage systems through multi-dimensional data fusion, physical-chemical-visual model collaboration, and dynamic risk assessment mechanisms, while reducing operation and maintenance costs and safety risks, providing a highly reliable proactive protection solution for large-scale energy storage applications.

[0071] By integrating multi-dimensional data such as electrothermal parameters, gas concentration, and visual imagery, the system enables simultaneous detection and quantitative assessment of overcharging anomalies, sealing anomalies, and connection anomalies, solving the problem of missed detections caused by single-dimensional monitoring in traditional methods. The comprehensive risk assessment mechanism can dynamically quantify the overall risk level, triggering tiered early warnings and significantly improving the real-time performance and accuracy of energy storage system safety protection.

[0072] This invention overcomes the limitations of traditional single-dimensional monitoring by integrating multi-source data such as electrothermal parameters (voltage, temperature), gas concentration, leakage rate, and visual imagery. It can simultaneously identify overcharge anomalies, sealing anomalies, and connection anomalies, significantly improving the early detection capability of complex faults and avoiding safety hazards caused by monitoring blind spots. Dynamic calculation models (such as overcharge protrusion, leakage rate, and connection anomaly degree) are designed for different anomaly types, combining time cumulative effects and multi-parameter correlation analysis to achieve quantitative assessment of the anomaly degree. For example, the overcharge anomaly degree, through weighted fusion of voltage, temperature extremes, and bulging degree, can distinguish between instantaneous fluctuations and true overcharge risk, reducing the false alarm rate by more than 40%. Based on the synergistic effect of overcharge anomaly degree, sealing anomaly degree, and connection anomaly degree, a dynamic weighted risk assessment model is constructed to accurately quantify the overall risk level of the system (low / medium / high). Image recognition technology (such as contact area calculation and corrosion area segmentation) replaces manual inspection, achieving automated diagnosis of connection anomalies and improving detection efficiency by more than 70%. Compared to traditional isolated alarm mechanisms, this invention can predict complex faults such as thermal runaway 10-15 minutes in advance and trigger tiered response strategies (such as early warning, current limiting, and power outage) to minimize losses. Simultaneously, it reduces unnecessary downtime for inspections and the waste of maintenance resources caused by false alarms. In summary, this invention, through multi-dimensional data fusion, dynamic quantification models, and intelligent decision-making mechanisms, achieves accurate perception, real-time assessment, and proactive protection of energy storage system safety risks, providing a highly reliable safety assurance solution for large-scale energy storage applications.

[0073] In summary, this invention achieves accurate perception, real-time assessment, and proactive protection of safety risks in energy storage systems through multi-dimensional data fusion, dynamic quantification models, and intelligent decision-making mechanisms, providing a highly reliable safety assurance solution for large-scale energy storage applications.

[0074] Finally, 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 the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for safety monitoring and early warning of a multi-dimensional energy storage device, characterized in that, The method comprises the following steps: Real-time acquisition of multiple monitoring data of each energy storage battery in the energy storage system, judgment of whether there is an abnormality based on the multiple monitoring data; When the judgment result is that there is an abnormality, the abnormal battery is located and the abnormal type is judged, and the abnormal type includes overcharge abnormality, sealing abnormality and connection abnormality; If the abnormal type is overcharge abnormality, the overcharge data of the abnormal battery is acquired, the overcharge prominence is calculated according to the overcharge data, and the bulging degree of the abnormal battery is calculated in real time, and the overcharge abnormality degree of the abnormal battery is obtained according to the overcharge prominence and the bulging degree; If the abnormal type is sealing abnormality, the volatile gas concentration in the air at the abnormal battery is acquired, and the electrolyte leakage rate of the abnormal battery is calculated, and the sealing abnormality degree is obtained according to the volatile gas concentration and the electrolyte leakage rate; If the abnormal type is connection abnormality, the connection abnormality degree is judged according to the image between the abnormal battery and the connector; Comprehensive risk assessment is carried out according to the overcharge abnormality degree, the sealing abnormality degree and the connection abnormality degree, and a corresponding risk value is obtained, and a warning measure is taken based on the risk value.

2. The method of claim 1, wherein the method further comprises: The overcharge data of the abnormal battery includes overcharge voltage, overcharge temperature and overcharge time; The overcharge prominence is calculated according to the overcharge data, and the specific process is as follows: The voltage difference between the maximum value of the overcharge voltage and the rated voltage is calculated and is given by the following equation: ； wherein, is a voltage difference, is a maximum value of overcharge voltage, is a rated voltage; The temperature difference between the maximum value of the overcharge temperature and the rated temperature The formula is as follows: ； wherein, is the temperature difference, is the maximum value of the overcharge temperature, is the rated temperature; The overcharge time includes an overcharge start time and an overcharge end time, and the overcharge time period is calculated according to the overcharge start time and the overcharge end time , as follows: ； wherein, is the overcharge period, is the overcharge end time, is the overcharge start time; Based on voltage difference Temperature difference and overcharge period Calculate overcharge protrusion The formula is as follows: wherein, is the overcharge prominence, , and are corresponding weight coefficients.

3. The method of claim 1, wherein the method further comprises: The overcharge abnormality degree of the abnormal battery is obtained according to the overcharge prominence and the bulging degree, and the formula is as follows: wherein, is overcharge abnormality degree, and is corresponding weight coefficient, is overcharge prominence degree, is bulge degree.

4. The method of claim 1, wherein the method further comprises: The formula for calculating the electrolyte leakage rate of the abnormal battery is as follows: wherein, is the electrolyte leakage rate, is the battery volume, is the abnormal battery internal and external pressure difference change amount, is the change time.

5. The method of claim 4, wherein the method further comprises: The calculation formula of the abnormal pressure difference change between the inside and outside of the battery is: ; ； ； wherein, P0 is an initial internal pressure of the battery; P1 is an initial external pressure of the battery; P2 is a final internal pressure of the battery, P3 is a final external pressure of the battery.

6. The method of claim 4, wherein the method further comprises: The sealing abnormality degree is obtained according to the volatile gas concentration and the electrolyte leakage rate, and the specific formula is as follows: wherein, is the sealing abnormality degree, is concentration, is concentration, , and is the corresponding weight coefficient.

7. The method of claim 1, wherein the method further comprises: The connection abnormality degree is judged according to the image between the abnormal battery and the connector, and the specific process is as follows: The image between the abnormal battery and the connector is collected from multiple angles, and the connection degree of the abnormal battery and the connector and the corrosion degree of the contact point are obtained based on image recognition; The connection degree and the corrosion degree of the contact point are weighted and calculated to obtain the connection abnormality degree, and the specific formula is as follows: wherein, is a connection abnormality degree, is a connection degree of the abnormal battery and the connector, is a corrosion degree of the contact point, and is a corresponding weight coefficient.

8. The method of claim 1, wherein the method further comprises: The comprehensive risk assessment is carried out according to the overcharge abnormality degree, the sealing abnormality degree and the connection abnormality degree, and a corresponding risk value is obtained, and the calculation formula is as follows: wherein Z is a risk value, is an overfill abnormality degree, is a seal abnormality degree, is a connection abnormality degree, , and are corresponding weight coefficients.

9. The method of claim 1, wherein the method further comprises: The warning measure is taken based on the risk value, and the specific process is as follows: Low risk threshold, medium-low risk threshold, medium risk threshold, medium-high risk threshold and high risk threshold are set in sequence, the risk is divided into different levels according to the size of the risk value, and corresponding warning measures are taken, and the warning measures include: If the risk value is less than the low risk threshold, no measures need to be taken, and the monitoring continues; If the medium-low risk threshold is greater than the risk value and the low risk threshold is greater than or equal to the risk value, a slight warning is given, and it is suggested to check; If the medium risk threshold is greater than the risk value and the medium-low risk threshold is greater than or equal to the risk value, a medium warning is given, a sound and light alarm is triggered, and it is suggested to take preventive maintenance measures; If the medium-high risk threshold is greater than the risk value and the medium risk threshold is greater than or equal to the risk value, a high warning is given, maintenance measures are immediately taken, and the operation of the related equipment is temporarily suspended; If the high risk threshold is greater than the risk value and the medium-high risk threshold is greater than or equal to the risk value, the highest warning is given, the operation of the related equipment is immediately stopped, and comprehensive inspection and repair are carried out.

10. A multi-dimensional energy storage device safety monitoring and early warning system for implementing the multi-dimensional energy storage device safety monitoring and early warning method of any one of claims 1-9, characterized in that, It comprises: A monitoring data acquisition module for acquiring multiple monitoring data of each energy storage battery in the energy storage system in real time; An abnormality judgment module for judging whether there is an abnormality based on the multiple monitoring data; The abnormal type judgment module is configured to, when the judgment result is that there is an abnormality, locate the abnormal battery and perform abnormal type judgment. The abnormal type includes overcharge abnormality, sealing abnormality, and connection abnormality. The overcharge abnormality module is configured to, when the abnormal type is overcharge abnormality, obtain overcharge data of the abnormal battery, calculate overcharge prominence based on the overcharge data, and calculate the bulging degree of the abnormal battery in real time. The overcharge abnormality degree of the abnormal battery is obtained based on the overcharge prominence and the bulging degree. The sealing abnormality module is configured to, when the abnormal type is sealing abnormality, obtain the volatile gas concentration in the air around the abnormal battery, calculate the electrolyte leakage rate of the abnormal battery, and obtain the sealing abnormality degree based on the volatile gas concentration and the electrolyte leakage rate. The connection abnormality module is configured to, when the abnormal type is connection abnormality, determine the connection abnormality degree based on the image between the abnormal battery and the connector. The comprehensive risk assessment module is configured to perform comprehensive risk assessment based on the overcharge abnormality degree, the sealing abnormality degree, and the connection abnormality degree, and obtain a corresponding risk value. The early warning module is configured to take early warning measures based on the risk value.

Images