Energy storage equipment safety protection early warning system applied to industrial park

Abstract

The invention belongs to the technical field of safety protection, and particularly relates to an energy storage equipment safety protection early warning system applied to an industrial park, which adopts a module-cluster-cabin three-level architecture to realize real-time monitoring and dynamic evaluation of thermal runaway risk, the architecture integrates multi-source data, and when the risk exceeds a preset threshold value, the energy storage equipment safety protection early warning system is used for early warning the thermal runaway risk. The system automatically starts graded fire extinguishing measures and a fire-fighting linkage mechanism, and meanwhile, secondary analysis is carried out on an early warning result through a cloud platform to optimize model parameters, so that the accuracy and the efficiency of the system are ensured; the system constructs an integrated monitoring, response and self-correction prevention and control process, and effectively overcomes the problems of early warning delay, high false alarm rate, single protection means, data islanding and the like in the prior art through a multi-parameter fusion early warning, accurate recognition, graded fire extinguishing and intelligent linkage mechanism. And a solid guarantee is provided for fire safety in a high-density energy storage environment.

Description

Technical Field

[0001] The present invention belongs to the field of safety protection, and particularly relates to an early warning system, and is specifically disclosed as an energy storage equipment safety protection early warning system applied to an industrial park. Background Art

[0002] With the advancement of the "dual carbon" strategy, electrochemical energy storage systems represented by lithium-ion batteries have rapidly become popular in industrial parks. However, frequent safety accidents such as fires and explosions caused by thermal runaway of energy storage equipment have become a core bottleneck restricting the large-scale development of the industry.

[0003] Currently, most industrial parks use a traditional safety protection system that relies on a BMS (Battery Management System) to monitor thresholds for single-dimensional parameters such as battery pack voltage and temperature. Smoke detectors (photoelectric / ionization) and temperature sensors (NTC thermistors) form a fire detection network. The trigger threshold is generally set above 150°C. When visible smoke or high temperatures are detected, a full-flood gas fire extinguishing device is activated. Most fire extinguishing systems activated after an alarm is triggered are full-flood. While this design can effectively extinguish fires during open flames, it suffers from four core flaws:

[0004] 1. Failure to identify risks in the early stages of thermal runaway, such as during the hydrogen evolution phase of the battery;

[0005] Second, a single sensor is susceptible to environmental interference, such as dust and humidity, which can lead to misjudgment;

[0006] 3. Unable to accurately suppress local thermal runaway, causing secondary damage to the equipment;

[0007] 4. There is a lack of data linkage between the battery management system, environmental monitoring and fire protection systems, making it difficult to achieve active protection.

[0008] Therefore, it is necessary to provide a safety protection system for energy storage equipment with multi-parameter fusion warning, accurate identification, graded fire extinguishing and intelligent linkage. Summary of the Invention

[0009] In view of this, the present invention proposes a safety protection and early warning system for energy storage equipment in industrial parks. It adopts a three-level "module-cluster-cabin" architecture to achieve real-time monitoring and dynamic assessment of thermal runaway risks. This architecture integrates multi-source data. When the risk exceeds the preset threshold, the system will automatically initiate graded fire extinguishing measures and fire linkage mechanisms. At the same time, the effectiveness of the protective measures is confirmed through verification of the safety feedback module. This system constructs an integrated monitoring, response and self-correction prevention and control process to solve the problems of delayed warning, high false alarm rate, single protection method and data silos in the existing technology.

[0010] The purpose of the present invention can be achieved by the following technical solution: an energy storage equipment safety protection early warning system applied to an industrial park, comprising:

[0011] Campus architecture planning module: Based on data location features, the campus map information is imported into the cloud and location zones are constructed, which are recorded as battery module zones, cluster-level zones, and cabin-level zones.

[0012] Detection data set acquisition module: used to collect data in the battery module area, cluster area and cabin area, and obtain primary data text, secondary data text and tertiary data text respectively;

[0013] Text data preprocessing module: used for integrating the primary data text, the secondary data text and the tertiary data text to obtain the primary integrated data text, the secondary integrated data text and the tertiary integrated data text;

[0014] Integrated data analysis module: used to import the first integrated data text, the second integrated data text, and the third integrated data text into the risk control data calculation model to obtain the thermal runaway risk index R;

[0015] Warning signal generation module: used to analyze the thermal runaway risk index R and generate a warning signal to the cloud;

[0016] Database secondary construction module: used to perform secondary analysis after receiving the warning signal and generate graded warning signals to the cloud;

[0017] Safety protection module: used to activate graded fire extinguishing devices according to graded early warning signals;

[0018] Fire linkage module: used to take different measures according to graded warning signals;

[0019] Safety feedback module: used to provide feedback on the safety protection and fire protection linkage results to determine whether the safety protection and fire protection linkage are successful;

[0020] Safety assessment module: used to analyze and process the safety assessment coefficients of each designated detection area.

[0021] Combining all the above technical solutions, the present invention has the following positive effects:

[0022] 1. This invention achieves early warning by overcoming the limitations of single physical quantity monitoring and enabling effective intervention before the irreversible stage of thermal runaway through the fusion of electrical, thermal, and gas multimodal data.

[0023] 2. The present invention achieves accurate judgment by adopting a multi-source data anti-interference verification mechanism, eliminating outliers and performing safety assessments to eliminate the risk of malfunction caused by environmental interference;

[0024] 3. This invention achieves precise fire prevention by designing differentiated fire suppression strategies based on the thermal runaway propagation path (module → cluster → cabin), balancing suppression efficiency and equipment protection.

[0025] 4. The present invention achieves the effect of fast collaboration, specifically: building a real-time interactive channel for cross-system data, and realizing millisecond-level linkage scheduling of BMS, environmental monitoring and fire protection resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0027] Attachment Figure 1 It is the workflow diagram of the present invention.

[0028] Attachment Figure 2 This is a schematic diagram of the device connection of the present invention.

[0029] Attachment Figure 3 It is a data collection step diagram of the present invention. DETAILED DESCRIPTION

[0030] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] See also Figure 1 and Figure 2 As shown, the present invention proposes an energy storage equipment safety protection early warning system applied to industrial parks, which is characterized by including a park architecture planning module, a detection data set acquisition module, a text data preprocessing module, an integrated data analysis module, a database secondary construction module, a safety protection module, a fire linkage module, a safety feedback module, and a safety assessment module.

[0032] In a more specific application of the present invention, the campus architecture planning module imports campus map information into the cloud based on data location features, constructs location areas, and records them as battery module areas, cluster-level areas, and cabin-level areas.

[0033] In the module-level area, an aerosol fire extinguishing device is installed inside the battery module to cover thermal runaway of a single battery cell; the module-level database stores the temperature, voltage and current data corresponding to a single battery module, with a data collection interval of ≤5 seconds.

[0034] In the cluster-level area, a perfluorohexanone directional injection device is deployed on the top of the battery cluster cabinet; the cluster-level database stores the gas concentration distribution matrix and thermal imaging data at the battery cluster level, with a data collection interval of ≤30 seconds.

[0035] The cabin area is equipped with a HFC-227ea full flooding system in the container; the cabin database stores the cabin structure stress monitoring data and pressure fluctuation curve, and the data collection interval is ≤2 minutes.

[0036] The detection data set acquisition module is used to collect data in the battery module area, cluster level area and cabin level area, and obtains the primary data text, secondary data text and tertiary data text respectively. The data collection steps are as follows: Figure 3 As shown, specifically:

[0037] A1. Set up data collection terminals in the battery module area, cluster area, and cabin area respectively.

[0038] A2. Collect data from the acquisition terminal to obtain the sum of temperature data, combustible gas data, smoke data, and current / voltage data;

[0039] A3. Data is constructed based on the data sum to obtain a module-level database, a cluster-level database, and a cabin-level database;

[0040] A4. The data in the module-level database, cluster-level database, and cabin-level database are recorded as primary data text, secondary data text, and tertiary data text, respectively.

[0041] A multi-source sensor is installed on the acquisition end, and the multi-source sensor includes a temperature sensor, a hot spot sensor, a combustible gas sensor, a smoke sensor and a current / voltage sensor.

[0042] The text data preprocessing module is used to integrate the primary data text, the secondary data text and the tertiary data text. Before the integration, the data text is processed. The processing steps are as follows:

[0043] Data cleaning was performed to remove outliers, i.e. data points exceeding 3 times the standard deviation range;

[0044] Standardize the format and convert multi-source heterogeneous data into JSON format with unified timestamps;

[0045] Feature enhancement: add gradient change labels to temperature data, and mark the red warning segment when ΔT / Δt>5℃ / min.

[0046] After processing, the primary integrated data text, the secondary integrated data text and the tertiary integrated data text are obtained. The integration process of the primary integrated data text is specifically to import the primary data text into a primary calculation model, and the primary calculation model includes a temperature model, a combustible gas model, a smoke model, and a voltage / current parameter model.

[0047] It should be noted that the temperature model includes the battery module surface temperature change rate coefficient, denoted as K ΔT ; Cabin ambient temperature coefficient, recorded as K Tenv ; and the temperature difference coefficient within the battery cluster, denoted as K Tcluster ;

[0048] The battery module surface temperature change rate coefficient is specifically:

[0049]

[0050] Since t is the high-risk threshold of the industry standard, exceeding this value indicates accelerated thermal runaway, and T is the extreme limit, the range is divided into ΔT≤t, t<ΔT≤T and ΔT>T for segmented design.

[0051] Wherein, ΔT is the temperature change on the surface of the battery module, t = M°C / min, T = N°C / min, the specific value of M is any data between 5-8, and the specific value of N is any data between 15-18. The main value setting is based on ensuring the thermal safety of the battery module. The main purpose of this embodiment is to limit the range of possible settings of the high-risk threshold of the industry standard in actual situations. Therefore, the specific values ​​of M and N are not specifically limited.

[0052] In low-risk scenarios:

[0053] K ΔT =k1;

[0054] At this time, ΔT≤t is the slow self-heating stage, k1 is the contribution value of the temperature change rate at low risk, and the value is between 0.15-0.25. The specific value is determined according to the actual situation.

[0055] In medium- and high-risk scenarios:

[0056]

[0057] At this time, t<ΔT≤T, the risk increases with the temperature growth rate in a power law manner, the internal short circuit triggers a chain exothermic reaction, and the acceleration is nonlinear. k2 is the contribution value of the temperature change rate at medium and high risks, and the value is between 0.25-0.35. The specific value is determined according to the actual situation. The value of b is set between 1.2-1.5.

[0058] For example, when the battery module surface temperature changes ΔT = 10 ° C / min, take k1 = 0.2, k2 = 0.3, T = 15, t = 5, b = 1.5,

[0059]

[0060] That is, when ΔT = 10°C / min, the contribution value of the battery module surface temperature change rate is 0.3.

[0061] In extreme scenarios:

[0062] K ΔT =0.5;

[0063] At this time, ΔT>T, which is an extreme event such as deflagration, so a saturation limit is applied, that is, the maximum contribution of the temperature change rate is 0.5, accounting for 50% of the total weight of the temperature parameter.

[0064] The cabin ambient temperature coefficient is specifically: evn ≥F℃,

[0065]

[0066] Since the upper operating temperature limit of most energy storage systems is FG°C, exceeding this range may lead to performance degradation or thermal runaway. F°C is selected as the threshold for early warning. The higher the temperature, the greater the risk of thermal decomposition of materials in the cabin, equipment heat dissipation failure, or spontaneous combustion of combustibles. Therefore, an exponential function is introduced to enhance high temperature risk.

[0067] Where T evn is the real-time ambient temperature in the cabin, the threshold temperature is F℃, which is the critical value of triggering temperature risk. When the temperature is lower than this, K Tevn =0.

[0068] The higher the temperature, the greater the risk of thermal decomposition of materials in the cabin, failure of equipment heat dissipation, or spontaneous combustion of combustibles. The range of F is between 35-45, and the range of G is between 45-55. The main numerical settings are based on ensuring the safety of the battery cabin. The main purpose of this embodiment is to limit the upper limit of the operating temperature of the energy storage system to the range that may be set in actual conditions. Therefore, there is no specific limitation on the specific values ​​of F and G.

[0069] This coefficient converts the temperature rise (T evn -F) is mapped to the risk contribution value, ranging from Where a reflects the contribution of the real-time ambient temperature in the cabin to the temperature model, and the value is any data between 0.35-0.45. The specific value is determined according to the actual situation.

[0070] The coefficient It is used to control the exponential growth rate of temperature risk accumulation. When the ambient temperature rises by q℃, the exponential term The value of increases by 1.

[0071] For example, the coefficient is used to control the exponential growth rate of temperature risk accumulation. When the coefficient is 0.1, the exponential term 0.1 (T evn -F) increases by 1, and the formula increases by 1-e -1 ≈63%, which is consistent with the change trend of activation energy in the Arrhenius equation of lithium battery electrolyte decomposition reaction; when T evn =(F+10)℃ (i.e., exceeding the threshold by 10℃), the coefficient 0.1 makes the risk value It reaches about 63% of the maximum value (0.4×0.63≈0.25), which is consistent with the inflection point of the thermal runaway trigger probability measured experimentally.

[0072] The temperature difference coefficient within the battery cluster is specifically:

[0073]

[0074] Where T max is the maximum temperature in the cluster, T min is the lowest temperature in the cluster, is the average temperature within the cluster,

[0075]

[0076] n is the total number of temperature measurements, T i The temperature of any temperature measurement.

[0077] It should be noted that the combustible gas model is specifically the combustible gas concentration rising coefficient, denoted as K gas ;

[0078] The combustible gas concentration increase coefficient is specifically:

[0079]

[0080] Among them C gas Indicates the real-time concentration of combustible gas in the cabin, measured in percentage of lower explosion limit (%LEL). 1%LEL is 1% of the lower explosion limit, i.e., one percent of the minimum combustible concentration. For normalization, the actual concentration is converted into a multiple of 1% LEL to facilitate unified calculation (elimination of dimension).

[0081] The concentration term Reflects the linear relationship between the current concentration and the reference value (1% LEL). The coefficient m1 controls the sensitivity. The value of m1 is any data between 0.1-0.2. The specific value is determined according to the actual situation.

[0082] The temperature linkage term is m2×K ΔT A temperature variation coefficient, such as a temperature difference correction factor, is introduced to reflect the impact of temperature on gas diffusion and reaction rate. When the temperature change rate increases, the gas risk weight automatically increases. The value of m2 is any data between 0 and 0.1, and the specific value is determined based on actual conditions.

[0083] The upper limit constraint min (m, total calculated value) has a value of 0.3, which means that the maximum contribution of the combustible gas concentration increase rate is 0.3, avoiding excessive amplification of the risk value in extreme scenarios.

[0084] It should be noted that the smoke model is specifically the smoke particle concentration coefficient, denoted as K smoke :

[0085] The smoke particle concentration coefficient is specifically:

[0086]

[0087] Among them C t is the real-time smoke particle concentration, C0 is the baseline smoke particle concentration, C max is the upper limit of the sensor range;

[0088] It should be noted that the voltage / current parameter model is uniformly integrated into the electrical unbalance coefficient, denoted as K elec :

[0089] The electrical unbalance coefficient is specifically:

[0090]

[0091] Where V actual is the actual measured voltage, V nomial is the nominal voltage of the battery, I actual is the actual measured current, I rated is the rated current of the battery, and 0.5 means the weight of current and voltage parameters are half each.

[0092] The voltage anomaly item quantifies the degree to which the voltage deviates from the nominal value and reflects the risk of over-discharge;

[0093] When V nomial =V actual In the normal state, the numerator is 0 and the voltage term contributes 0;

[0094] When V actual>V nomial In the over-discharge state, the numerator is positive and the voltage abnormality value increases;

[0095] When the voltage is 20% lower than the nominal value, the over-discharge protection threshold is triggered.

[0096] The abnormal current value quantifies the proportion of the current exceeding the rated value, reflecting the risk of short circuit or overcurrent;

[0097] When I actual =I rated In the normal state, the numerator is 0 and the current term contributes 0;

[0098] When I actual >I rated In the overcurrent state, the numerator is positive and the abnormal current value increases;

[0099] If the current reaches 150% of the rated value, it indicates overcurrent.

[0100] The integrated data analysis module is used to import the first integrated data text, the second integrated data text and the third integrated data text into the risk control data calculation model to obtain the thermal runaway risk index R.

[0101] Based on the integrated data text aggregation results output by a model, the thermal runaway risk index R is dynamically calculated through the iterative weight coefficient of the risk control model. The formula is as follows:

[0102] R=α·(K ΔT +K Tenv +K cluster )+β·(K gas +K drop )+γ·K smoke +δ·K elec ;

[0103] Among them, α, β, γ and δ represent weight factors. The design principle of the weight distribution formula is as follows: temperature parameters reflect that thermal runaway is directly related to the risk of system explosion; combustible gas system coefficients reflect that sudden changes in gas concentration indicate extreme events; smoke coefficients reflect that the concentration of combustion products is positively correlated with the scale of the fire; abnormal electrical parameters reflect that overcharging / over-discharging triggers a chain reaction.

[0104] The dynamic adjustment range is: α∈[0.4,0.5],β∈[0.25,0.35],γ∈[0.15,0.25],δ∈[0.1,0.2];

[0105] The dynamic weight adjustment mechanism is:

[0106] In a high temperature environment (cabin temperature > X), the temperature parameter weight is increased by 10%, and other parameters are compressed proportionally. The conventional value of X is 40-50°C, which is not specifically limited in this embodiment.

[0107] When the detected CO concentration > Y, the weight of the smoke coefficient is increased by 10%, and the weight of the combustible gas is decreased by 5%. The conventional value of Y is between 50 - 60 ppm, and no specific limitation is set in this embodiment;

[0108] When the voltage is abnormal (> nominal value Z), the weight of the electrical parameter is increased by 10%, and the weight of the temperature is decreased by 5%. The conventional value of Z is between 20 - 30%, and no specific limitation is set in this embodiment.

[0109] The early warning signal generation module is used to analyze the thermal runaway risk index R and generate an early warning signal to the cloud. The working mechanism of the early warning signal generation module is as follows:

[0110] When the comprehensive risk index R reaches or exceeds the low-risk threshold r, the system will immediately trigger the early warning mechanism to ensure that timely measures are taken to prevent potential safety accidents. Here, the value of r is between 0.4 - 0.5.

[0111] Relatively, when the comprehensive risk index R is lower than r, the system will execute the conventional monitoring mode. In the conventional monitoring mode, the system will automatically generate a health report of the energy storage device every hour to continuously track the operation status of the device and ensure that it is in good working condition.

[0112] The database secondary construction module is used to perform secondary analysis after receiving the early warning signal and generate a hierarchical early warning signal to the cloud platform. After the fire is confirmed, the hierarchical early warning signal is triggered based on the thermal runaway risk index R:

[0113] When r ≤ R < r1, a first-level early warning is triggered; when r1 ≤ R < r2, a second-level early warning is triggered; when R ≥ r2, a third-level early warning is triggered. Here, r1 is the medium-risk threshold, and its value is between 0.6 - 0.7; r2 is the high-risk threshold, and its value is between 0.8 - 0.9;

[0114] The safety protection module is used to start the hierarchical fire extinguishing device according to the hierarchical early warning signal:

[0115] First-level early warning: Activate the cell-level compressed air foam fire extinguishing device, and spray the aerosol fire extinguishing agent containing K2CO3 within 3 seconds to quickly suppress the thermal runaway of a single cell and control the loss range;

[0116] Second-level early warning: Start the perfluoropentanone pre-positioned pipe network fire extinguishing device, and achieve precise coverage of the agent at 0.3 m

[0117] / min to avoid waste of the fire extinguishing agent;

[0118] Level 3 warning: Linked with the HFC-227ea full flooding system, the agent spraying is completed within 10 seconds and the fire extinguishing concentration is maintained for ≥30 minutes, meeting the UL 9540A standard's requirement for complete blocking of thermal runaway spread to prevent the spread of fire.

[0119] The fire linkage module is used to take different measures according to graded warning signals:

[0120] Level 1 warning: Start the exhaust system and push the operation and maintenance alarm;

[0121] Level 2 alarm: The power supply of the equipment is cut off and fire extinguishing is prepared;

[0122] Level 3 action: Automatically release the fire extinguishing medium and trigger the emergency broadcast.

[0123] The safety feedback module is used to provide feedback on the safety protection and fire protection linkage results and determine whether the safety protection and fire protection linkage are successful. The judgment logic is as follows:

[0124] (1) If the following indicators are not met within 60 seconds after the fire extinguishing is started, it will be considered as failure:

[0125] The temperature drop rate is ≥x, the smoke concentration drop is ≥y, and the combustible gas concentration is <z, triggering the secondary injection of the standby fire extinguishing agent storage tank and activating the emergency smoke exhaust fan at the same time, where the value of x is between 4-6°C / s, the value of y is between 60-70%, and the value of z is between 5-7%LEL, which is not specifically limited in this embodiment.

[0126] (2) False alarm filtering mechanism:

[0127] Kalman filter denoising is enabled for abnormal data with R value fluctuation less than 0.1 and duration less than 15 seconds;

[0128] When voltage / current parameters are abnormal, BMS data is forced to be called for cross-validation.

[0129] The safety assessment module is used to analyze and process the safety assessment coefficients of each designated detection area and evaluate the prediction accuracy.

[0130] The formula for calculating the safety assessment factor is:

[0131]

[0132] Where j represents the number of false alarms, j 总 represents the total number of warnings, and p represents the actual fire scene recognition rate.

[0133] When Q<w, the model parameter calibration procedure is automatically triggered, where u and v represent weights, the u value is between 0.6-0.7, and the v value is between 0.3-0.4. The specific values ​​are determined according to actual conditions.

[0134] w×100% represents the prediction accuracy, and the range of w is between 0.8-1. The specific value is selected according to the needs of the industrial park.

Claims

1. The energy storage equipment safety protection early warning system used in industrial parks is characterized by: include: Campus architecture planning module: Based on data location features, the campus map information is imported into the cloud and location zones are constructed, which are recorded as battery module zones, cluster-level zones, and cabin-level zones. Detection data set acquisition module: used to collect data in the battery module area, cluster area and cabin area, and obtain primary data text, secondary data text and tertiary data text respectively; Text data preprocessing module: used for integrating the primary data text, the secondary data text and the tertiary data text to obtain the primary integrated data text, the secondary integrated data text and the tertiary integrated data text; Integrated data analysis module: used to import the first integrated data text, the second integrated data text, and the third integrated data text into the risk control data calculation model to obtain the thermal runaway risk index R; Warning signal generation module: used to analyze the thermal runaway risk index R and generate a warning signal to the cloud; Database secondary construction module: used to perform secondary analysis after receiving the warning signal and generate graded warning signals to the cloud; Safety protection module: used to activate graded fire extinguishing devices according to graded early warning signals; Fire linkage module: used to take different measures according to graded warning signals; Safety feedback module: used to provide feedback on the safety protection and fire protection linkage results to determine whether the safety protection and fire protection linkage are successful; Safety assessment module: used to analyze and process the safety assessment coefficients of each designated detection area.

2. The energy storage equipment safety protection early warning system for industrial parks according to claim 1, characterized in that: The acquisition steps of the detection data set acquisition module are as follows: A1. Set up data collection terminals in the battery module area, cluster area, and cabin area respectively. A2. Collect data from the acquisition terminal to obtain the sum of temperature data, combustible gas data, smoke data, and current / voltage data; A3. Data is constructed based on the data sum to obtain a module-level database, a cluster-level database, and a cabin-level database; A4. The data in the module-level database, cluster-level database, and cabin-level database are recorded as primary data text, secondary data text, and tertiary data text, respectively.

3. The energy storage equipment safety protection early warning system for industrial parks according to claim 1 is characterized in that: The integration process of the primary integrated data text is specifically to import the primary data text into a primary calculation model, and the primary calculation model includes a temperature model, a combustible gas model, a smoke model, and a voltage / current parameter model.

4. The energy storage equipment safety protection early warning system for industrial parks according to claim 3 is characterized in that: The temperature model includes the battery module surface temperature change rate coefficient, denoted as K ΔT ; Cabin ambient temperature coefficient, recorded as K Tenv ; and the temperature difference coefficient within the battery cluster, denoted as K Tcluster ; The battery module surface temperature change rate coefficient is specifically: Where ΔT is the temperature change on the battery module surface, k1 and k2 are the contribution values ​​of the temperature change rate, and the maximum contribution of the temperature change rate is 0.5; The cabin ambient temperature coefficient is specifically: evn ≥F℃, Where T evn is the real-time ambient temperature in the cabin, the threshold temperature is F℃, and the critical value of triggering temperature risk at this time; pass Increase the temperature (T evn -F) is mapped to the risk contribution value, ranging from The coefficient To control the exponential growth rate of temperature risk accumulation, when the ambient temperature rises by q℃, the exponential term The value of increases by 1; The temperature difference coefficient within the battery cluster is specifically: Where T max is the maximum temperature in the cluster, T min is the lowest temperature in the cluster, is the average temperature within the cluster, n is the total number of temperature measurements, T i The temperature of any temperature measurement.

5. The energy storage equipment safety protection early warning system for industrial parks according to claim 3 is characterized in that: The combustible gas model is specifically the combustible gas concentration rising coefficient, denoted as K gas ; The combustible gas concentration increase coefficient is specifically: Among them C gas Indicates the real-time concentration of combustible gas in the cabin. For normalization, the actual concentration is converted into a multiple of 1% LEL to facilitate unified calculation (elimination of dimension).

6. The energy storage equipment safety protection warning system for industrial parks according to claim 3, characterized in that: The smoke model is specifically the smoke particle concentration coefficient, denoted as K smoke : The smoke particle concentration coefficient is specifically: Among them C t is the real-time smoke particle concentration, C0 is the baseline smoke particle concentration, C max The upper limit of the sensor range.

7. The energy storage equipment safety protection early warning system for industrial parks according to claim 3 is characterized in that: The voltage / current parameter model is unified into the electrical unbalance coefficient, denoted as K elec : The electrical unbalance coefficient is specifically: Where V actual is the actual measured voltage, V nomial is the nominal voltage of the battery, I actual is the actual measured current, I rated is the rated current of the battery; is the voltage anomaly term, which quantifies the degree to which the voltage deviates from the nominal value; The current abnormal value quantifies the ratio of the current exceeding the rated value.

8. The energy storage equipment safety protection early warning system for industrial parks according to claim 1, characterized in that: The formula of the thermal runaway risk index R is as follows: R=α·(K ΔT +K Tenv +K cluster )+β·K gas +γ·K smoke +δ·K elec ; where α, β, γ and δ represent weight factors.

9. The energy storage equipment safety protection warning system for industrial parks according to claim 1, characterized in that: The safety assessment coefficient formula is as follows: Where j represents the number of false alarms, j 总 Represents the total number of warnings, p represents the actual fire scene recognition rate, and when Q<w, the model parameter calibration procedure is automatically triggered.

Images