An intelligent lightning early warning system for large oil and gas storage bases

Abstract

This invention relates to the field of lightning early warning and discloses an intelligent lightning early warning system for large oil and gas storage bases. The system comprises a multi-source monitoring module, a data processing and analysis module, an early warning module, and a coordinated protection module. The multi-source monitoring module collects data from multiple dimensions, the data processing and analysis module preprocesses this data, and calculates the probability of lightning occurrence using a specific model. The early warning module issues corresponding early warning signals based on the hazard level, while the coordinated protection module automatically triggers corresponding protective measures based on the early warning signals. By employing multi-source data fusion, probabilistic model analysis, and a hierarchical linkage mechanism, this invention effectively solves the false alarm and missed alarm problems inherent in traditional single monitoring methods, achieving accurate early warning and intelligent protection for oil and gas storage bases, and significantly improving the lightning safety protection capabilities of large oil and gas storage bases.

Description

Technical Field

[0001] This invention relates to the field of lightning warning, and more specifically to an intelligent lightning warning system for large oil and gas storage bases. Background Technology

[0002] In the energy industry, large-scale oil and gas storage bases serve as crucial hubs for national energy security, storing vast quantities of flammable and explosive petroleum and natural gas, among other hazardous chemicals. However, lightning strikes pose a serious threat to the safe operation of these bases. Direct lightning strikes, induced lightning strikes, and lightning electromagnetic pulses can trigger fires and explosions in oil and gas storage tanks, damage electrical equipment, and paralysis of communication systems, causing enormous economic losses and casualties, and even severe environmental pollution and social impacts.

[0003] Currently, existing lightning warning systems on the market are mainly based on single monitoring methods, such as using ground electric field meters to monitor changes in the atmospheric electric field, or using satellite cloud images to make macroscopic predictions of lightning weather.

[0004] While it can provide lightning warnings, it still has some shortcomings, such as:

[0005] On the one hand, single monitoring methods are easily affected by environmental factors, and the accuracy and reliability of monitoring data are insufficient, which can easily lead to false alarms or missed alarms. On the other hand, the early warning system lacks targeted design for the complex scenarios of large oil and gas storage bases, and cannot achieve accurate early warning and coordinated protection of key areas within the base. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent lightning early warning system for large-scale oil and gas storage bases, thereby solving the above-mentioned technical problems.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] An intelligent lightning early warning system for large oil and gas storage bases includes:

[0009] The multi-source monitoring module includes an atmospheric electric field monitoring unit, a satellite cloud image receiving unit, a lightning location monitoring unit, and an environmental monitoring unit within the base. It is used to collect multi-dimensional data on changes in the atmospheric electric field, the macro-meteorological situation, lightning locations, and the environment of key areas within the base.

[0010] The data processing and analysis module is used to preprocess the data collected by the multi-source monitoring module, including data cleaning, spatiotemporal alignment and data standardization, and outputs the predicted overall lightning occurrence probability through the preset lightning occurrence probability analysis model.

[0011] The early warning module is used to issue early warning information to base management personnel according to different risk levels based on the prediction results of the data processing and analysis module.

[0012] The linkage protection module is used to automatically activate the corresponding lightning protection equipment and emergency protection measures within the base after receiving early warning information.

[0013] As a further technical solution, the expression for the overall lightning occurrence probability analysis model is as follows: ; This represents the probability of lightning occurring.

[0014] in, The rate of change of atmospheric electric field intensity is expressed as: , For a unit of time, This is the change in atmospheric electric field strength. As a reference rate of change of electric field, Used to reflect the effect of changes in electric field intensity on lightning. For lightning location density, Used to enhance the impact of high-density areas For reference lightning density, Temperature at the top of the cumulonimbus cloud. As the reference temperature, For temperature sensitivity coefficient, Used to simulate the properties of low temperature promoting lightning formation This is a curve showing the change in the environmental temperature and humidity abrupt change index over time. Used to accumulate the impact of environmental changes over a period of time. The start time, The current moment; , , , The weighting coefficients are determined based on historical data analysis.

[0015] As a further technical solution, the process for determining different hazard levels is as follows:

[0016] The overall probability of lightning occurrence obtained Compared with the preset lightning warning range Compare;

[0017] when When, it is judged as low risk; when When, it is judged as medium risk; when At that time, it was determined to be high risk.

[0018] As a further technical solution, the early warning module also includes:

[0019] The early warning information verification unit can divide the storage base into several key sub-regions based on the base's geographic information system (GIS), analyze the lightning occurrence probability of each key sub-region, and analyze whether the overall lightning occurrence probability risk level is true based on the lightning occurrence probability of each key sub-region. If it is true, the early warning information is issued normally; otherwise, the current risk level is adjusted before issuing the early warning information.

[0020] As a further technical solution, the formula for calculating the probability of lightning occurrence in each key sub-region is as follows: ;

[0021] in, The lightning location density for the current key sub-region, This represents the rate of change of atmospheric electric field intensity in the current key sub-region. This represents the risk weight coefficient for the current key sub-region.

[0022] As a further technical solution, the probability of lightning occurrence in each key sub-region is... After sorting in descending order, the critical sub-region ranked first is determined to be the critical sub-region with the highest probability of lightning occurrence;

[0023] The corresponding Compared with the preset lightning warning range After comparison, the probability risk level of lightning occurrence in the current key sub-region is determined;

[0024] If the lightning probability risk level of the current key sub-region is consistent with the overall lightning probability risk level, then the judgment is true;

[0025] If the lightning probability risk level of the current key sub-region is inconsistent with the overall lightning probability risk level, then the absolute difference between the lightning probability of the current key sub-region and the overall lightning probability is calculated. If the absolute difference is greater than the risk level difference threshold, then the lightning probability risk level of the current key sub-region is taken as the overall lightning probability risk level; otherwise, the overall lightning probability risk level remains unchanged.

[0026] As a further technical solution, the formula for calculating the environmental temperature and humidity abrupt change index is as follows: ,in, For the rate of temperature change, This represents the rate of change in humidity.

[0027] As a further technical solution, the emergency protection measures of the linkage protection module include:

[0028] During low-risk warnings, equipment inspection robots are activated to conduct infrared thermal imaging inspections on lightning protection equipment.

[0029] When a medium-risk warning is issued, the exposed interfaces of the oil and gas storage tank area are automatically shut off, and the exposed interfaces are purged with nitrogen.

[0030] When a high-risk warning is issued, a one-click power-off procedure is triggered to cut off the power supply to all non-emergency equipment and activate the uninterruptible power supply (UPS) to ensure the operation of the fire protection and monitoring systems.

[0031] The beneficial effects of this invention are:

[0032] (1) By integrating and monitoring multiple sources of data such as atmospheric electric field, satellite cloud image, lightning location and base environment, and combining the probability model analysis of multiple parameters such as comprehensive electric field change rate, lightning density and cumulonimbus temperature, the limitations of traditional single monitoring are overcome, the accuracy of lightning prediction is significantly improved, the false alarms and missed alarms are effectively reduced, and the goal of multi-dimensional accurate monitoring and reducing false alarm and missed alarm rates is achieved.

[0033] (2) Traditional early warning systems use a uniform early warning level for the entire base, which cannot identify local high-risk areas. This invention divides the base into gridded sub-regions, analyzes the probability of lightning occurrence in each region in real time, and automatically calibrates the global early warning level with the highest-risk sub-region. When abnormal data appears in a local area, the early warning level can be dynamically adjusted and the corresponding protective measures can be precisely triggered, avoiding the waste of resources caused by a one-size-fits-all response. This refines the early warning range from the entire base to key equipment units, minimizing interference with normal operations while ensuring safety. Attached Figure Description

[0034] The invention will now be further described with reference to the accompanying drawings.

[0035] Figure 1 This is a system structure diagram of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figure 1 As shown, this invention is an intelligent lightning early warning system for large-scale oil and gas storage bases, comprising:

[0038] The multi-source monitoring module includes an atmospheric electric field monitoring unit, a satellite cloud image receiving unit, a lightning location monitoring unit, and an in-base environmental monitoring unit. It is used to collect multi-dimensional data on atmospheric electric field changes, macro-meteorological conditions, lightning locations, and the environment of key areas within the base. For example: the atmospheric electric field monitoring unit deploys electric field sensors in a 1km×1km grid within the base to collect the rate of change of atmospheric electric field intensity at each grid point in real time; the satellite cloud image receiving unit acquires meteorological satellite data covering the base and a surrounding 20km radius, identifying the temperature and movement trajectory of cumulonimbus cloud tops; the lightning location monitoring unit connects to a wide-area lightning detection network and outputs the coordinates of cloud-to-ground lightning locations with a 500m accuracy in real time; and the in-base environmental monitoring unit deploys temperature and humidity sensors to collect environmental temperature and humidity abrupt change indices.

[0039] The data processing and analysis module is used to preprocess the data collected by the multi-source monitoring module, including data cleaning, spatiotemporal alignment, and data standardization. It outputs the predicted overall lightning occurrence probability through a pre-set lightning occurrence probability analysis model. Data cleaning includes:

[0040] Outlier detection: Employ the 3σ criterion (data points exceeding the mean ± 3 standard deviations) or an outlier detection algorithm based on isolated forests.

[0041] Missing value handling: Fill missing values ​​using linear interpolation, Kalman filtering, or time series prediction models based on historical data;

[0042] Noise filtering: Moving average filtering is applied to atmospheric electric field data, and wavelet transform is used to denoise satellite cloud images;

[0043] Spatiotemporal alignment includes:

[0044] Time synchronization: Unify data from different sampling frequencies into the same time window, such as a 1-minute interval, using nearest neighbor interpolation or spline interpolation; Spatial registration: Match lightning location data with the base's geographic information system (GIS) based on latitude and longitude coordinates to establish a unified spatial grid, such as 1km×1km;

[0045] Data standardization includes: Z-score standardization and threshold normalization;

[0046] The early warning module is used to issue early warning information to base management personnel according to different risk levels based on the prediction results of the data processing and analysis module.

[0047] The linkage protection module is used to automatically activate the corresponding lightning protection equipment and emergency protection measures within the base after receiving early warning information.

[0048] In this embodiment, a multi-source heterogeneous data fusion monitoring system is constructed, integrating atmospheric electric fields, satellite cloud images, lightning location data, and base environment data. This overcomes the limitations of traditional single monitoring methods and enables full-dimensional perception of the lightning incubation process. Combined with data cleaning, spatiotemporal alignment, and standardization, noise and inconsistencies in multi-source data are eliminated, improving the quality of model input. Based on a probabilistic analysis model, a graded early warning mechanism can dynamically quantify lightning risk levels as low, medium, and high. The linkage protection module triggers differentiated emergency measures through preset thresholds, forming a closed loop of perception-assessment-response, thereby reducing false alarm rates and avoiding high-risk missed alarms, ensuring the safety of critical facilities such as oil and gas storage tanks and electrical equipment.

[0049] The expression for the overall lightning occurrence probability analysis model is as follows: ; This represents the probability of lightning occurring.

[0050] in, The rate of change of atmospheric electric field intensity is expressed as: , For a unit of time, This is the change in atmospheric electric field strength. As a reference rate of change of electric field, Used to reflect the effect of changes in electric field intensity on lightning. For lightning location density, Used to enhance the impact of high-density areas For reference lightning density, Temperature at the top of the cumulonimbus cloud. As the reference temperature, For temperature sensitivity coefficient, Used to simulate the properties of low temperature promoting lightning formation This is a curve showing the change in the environmental temperature and humidity abrupt change index over time. Used to accumulate the impact of environmental changes over a period of time. The start time, The current moment; , , , The weighting coefficients are determined based on historical data analysis.

[0051] In this embodiment, the atmospheric electric field change rate, lightning location density, cumulonimbus cloud top temperature, and ambient temperature and humidity abrupt change index are incorporated into the unified probability model: Through weighting coefficients , , , Dynamically characterize the coupling effect of different parameters on lightning formation;

[0052] Since lightning is essentially a discharge phenomenon resulting from the accumulation of atmospheric charge to a critical value, the rate of change of the electric field directly reflects the rate of charge accumulation; therefore, it is adopted... Characterizing the relative intensity of change, when Exceeding the reference value When, the denominator term Follow The increase leads to an increase in the overall probability of lightning occurrence. The value approaches 1, consistent with the rule that a stronger electric field corresponds to a higher probability of lightning; weighting coefficient By fitting historical data, the contribution of electric field changes to lightning is quantified, such as in plains areas. Take 1.2, and take 1.5 for mountainous areas due to the topographic lifting effect;

[0053] Since the number of lightning location occurrences per unit area reflects the regional discharge activity, high-density areas usually indicate thunderstorm centers; therefore, a squared term is used. Enhance the impact of high-density areas, for example when At that time, its contribution was four times the original value, consistent with the clustering effect of lightning activity; adjustments were made based on regional climate characteristics, such as high thunderstorm areas. Use 0.8 for low thunderstorm areas and 0.5 for low thunderstorm areas;

[0054] Because the lower the temperature at the top of cumulonimbus clouds, the more violent the collisions between ice crystals and water droplets within the cloud, and the easier it is for charge separation to occur; therefore, through... Simulating the characteristics of low temperature promoting lightning formation, when Below the reference temperature At that time, this value is positive and increases as the temperature decreases, thus increasing the overall probability of lightning occurrence. Rise; referencing meteorological research, take To balance the coupling effect of temperature with other parameters;

[0055] Since sudden changes in temperature and humidity, such as a sharp drop in temperature and a sharp rise in humidity caused by strong convection, are precursors to thunderstorms, integration is used... This is used to accumulate the impact of environmental changes over a period of time, reflecting the pattern that persistently unstable conditions are more likely to induce lightning; it is adjusted according to the season, such as the summer season when severe convection is more frequent. Take 0.3, and 0.1 in winter;

[0056] In the above formula, the parameters in the denominator are coupled through multiplication, reflecting the mechanism by which multiple factors work together to trigger lightning. For example, even with a strong electric field but no cumulonimbus clouds or low temperatures, the probability of lightning is still low. The overall formula... make It exhibits a nonlinear jump near the parameter threshold, which is consistent with the critical threshold characteristics of lightning occurrence.

[0057] The process for determining different hazard levels is as follows:

[0058] The overall probability of lightning occurrence obtained Compared with the preset lightning warning range Compare; This is the critical value for a false alarm rate of ≤10% in historical data. This is the critical value for a false alarm rate of ≤5% in historical data;

[0059] when When, it is judged as low risk; when When, it is judged as medium risk; when At that time, it was determined to be high risk.

[0060] In this embodiment, a preset probability interval is used. Risk levels are categorized to enable precise hierarchical management of early warning signals. Low risk (P < P1) triggers equipment inspections and status monitoring to avoid resource waste. Medium risk (P1 ≤ P ≤ P2) initiates the sealing of exposed interfaces and nitrogen purging to block the propagation path of flammable media. High risk (P > P2) executes one-button power cutoff and UPS backup, prioritizing lifeline equipment such as fire protection and monitoring systems. This tiered strategy improves the alignment between emergency response and risk levels while reducing unnecessary protective actions, such as avoiding nitrogen consumption in low-risk situations, thus balancing safety and economy.

[0061] The early warning module also includes:

[0062] The early warning information verification unit can divide the storage base into several key sub-regions based on the base's geographic information system (GIS), analyze the lightning occurrence probability of each key sub-region, and analyze whether the overall lightning occurrence probability risk level is true based on the lightning occurrence probability of each key sub-region. If it is true, the early warning information is issued normally; otherwise, the current risk level is adjusted before issuing the early warning information.

[0063] In this embodiment, based on the spatial partitioning verification mechanism of the base's Geographic Information System (GIS), the base is divided into gridded sub-regions, such as tank storage areas, oil pipeline areas, and power distribution rooms. Local high-risk hotspots are identified through weighted calculations of regional lightning location density and atmospheric electric field change rate. When the risk level of a sub-region is inconsistent with the overall warning, the global warning level is adjusted based on the local high risk, addressing the problem of traditional overall warning models being insensitive to spatial differences. For example, if a tank storage area triggers a high risk due to a local strong electric field, but the overall probability does not reach the threshold, a targeted warning will still be issued for that area to prevent critical facilities from being overlooked.

[0064] The formula for calculating the lightning occurrence probability of each key sub-region is as follows: ;

[0065] in, The lightning location density for the current key sub-region, This represents the rate of change of atmospheric electric field intensity in the current key sub-region. This represents the risk weight coefficient for the current key sub-region.

[0066] In this embodiment, the formula for the probability of regional lightning occurrence is... The lightning location density and the rate of change of atmospheric electric field intensity in the current key sub-region reflect the local lightning activity intensity and charge accumulation characteristics of the sub-region, compared with global parameters. , More accurately depicting spatial differences; through and Calculate the deviation of sub-region parameters relative to the global parameters; risk weight coefficient of the current key sub-region. The safety importance of a sub-area is determined by factors such as the flammability and explosiveness of the equipment, personnel density, and historical accident consequences. For example, a tank area: =1.2, which is considered high-risk; a lightning strike could trigger an explosion; oil pipeline area: =1.0, which is considered medium risk; leakage risk needs to be monitored. Office area: =0.8, which is considered low risk; the primary concern is personnel safety. The regional probability is based on the global probability, with local parameter deviations and risk weights adjusted to prevent local anomalies from being masked by global averaging; when a sub-region or When significantly higher than the global average, such as >2 or >2 , It may far exceed the global probability, triggering an upgrade of the warning level.

[0067] The probability of lightning occurrence in each key sub-region After sorting in descending order, the critical sub-region ranked first is determined to be the critical sub-region with the highest probability of lightning occurrence;

[0068] The corresponding Compared with the preset lightning warning range After comparison, the probability risk level of lightning occurrence in the current key sub-region is determined;

[0069] If the lightning probability risk level of the current key sub-region is consistent with the overall lightning probability risk level, then the judgment is true;

[0070] If the lightning probability risk level of the current key sub-region is inconsistent with the overall lightning probability risk level, then the absolute difference between the lightning probability of the current key sub-region and the overall lightning probability is calculated. If the absolute difference is greater than the risk level difference threshold, then the lightning probability risk level of the current key sub-region is taken as the overall lightning probability risk level; otherwise, the overall lightning probability risk level remains unchanged.

[0071] In this embodiment, a cross-validation mechanism between sub-regional risk levels and overall early warning is used to address the misjudgment problem caused by averaging in traditional models; and a threshold is used to filter out minor fluctuations, such as absolute differences. ≤ Risk level difference threshold The system does not adjust settings periodically to prevent frequent changes in global early warnings due to local sensor noise or transient anomalies; it sets safety locks for high-risk levels, disallowing proactive downgrades, consistent with the safety principle of oil and gas bases that it is better to overestimate risks than underestimate threats; and it retains the ability to calibrate local risks, where local risks are significantly higher than global risks. When the risk level is forcibly upgraded to ensure timely activation of protective measures in critical areas, then, when the overall risk is misjudged as medium but the actual local risk is low, [the risk level will be lowered]. This allows for adjustments to the overall risk level based on localized low-risk situations, reducing unnecessary emergency response costs.

[0072] The formula for calculating the environmental temperature and humidity abrupt change index is as follows: ,in, For the rate of temperature change, This represents the rate of change in humidity.

[0073] In this embodiment, since thunderstorm formation requires a violent convergence of warm, moist air near the ground and dry, cold air at higher altitudes, sudden changes in temperature and humidity, such as a sudden drop in temperature and a sudden increase in humidity, are precursors to the development of strong convection. For example, sudden temperature changes... Reflecting the intensity of vertical air mass movement, rapid cooling may indicate cold air intrusion or descending air currents; sudden changes in humidity... Reflecting water vapor condensation efficiency, high humidity environments are more prone to forming cumulonimbus clouds; through real-time calculations... It can identify unstable environmental conditions during the incubation period of thunderstorms in advance. Compared with monitoring temperature or humidity alone, the comprehensive index can more sensitively capture critical conditions and advance the warning time.

[0074] The emergency protection measures of the linkage protection module include:

[0075] During low-risk warnings, equipment inspection robots are activated to conduct infrared thermal imaging inspections on lightning protection equipment.

[0076] When a medium-risk warning is issued, the exposed interfaces of the oil and gas storage tank area are automatically shut off, and the exposed interfaces are purged with nitrogen.

[0077] When a high-risk warning is issued, a one-click power-off procedure is triggered to cut off the power supply to all non-emergency equipment and activate the uninterruptible power supply (UPS) to ensure the operation of the fire protection and monitoring systems.

[0078] In this embodiment, the graded linkage protection strategy drives differentiated handling based on risk level: at low risk, an infrared thermal imaging inspection robot is activated to detect potential hazards such as abnormal grounding resistance and insulator cracks using AI image recognition technology, achieving non-intrusive prevention; at medium risk, exposed interfaces such as tank breathing valves and emergency shut-off valves are automatically closed, and flammable gases are replaced by nitrogen purging to suppress sparks that could cause an explosion; at high risk, a one-click power-off procedure is triggered to cut off non-emergency loads, such as lighting and pump sets, based on a priority list, while UPS is activated to ensure the continuous operation of critical equipment such as fire pumps, surveillance cameras, and emergency lighting.

[0079] This strategy dynamically matches protective measures with risk levels, reducing downtime losses by about 30% compared to the traditional "one-size-fits-all" power outage approach, while ensuring 100% availability of core safety systems.

[0080] It should be noted that the calculation formulas and all parameters involved in the calculations in this invention have been dimensionless beforehand. The process of dimensionless processing is well known in the industry and will not be described here.

[0081] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. An intelligent lightning early warning system for large-scale oil and gas storage bases, characterized in that, include: The multi-source monitoring module includes an atmospheric electric field monitoring unit, a satellite cloud image receiving unit, a lightning location monitoring unit, and an environmental monitoring unit within the base. It is used to collect multi-dimensional data on changes in the atmospheric electric field, the macro-meteorological situation, lightning locations, and the environment of key areas within the base. The data processing and analysis module is used to preprocess the data collected by the multi-source monitoring module, including data cleaning, spatiotemporal alignment and data standardization, and outputs the predicted overall lightning occurrence probability through the preset lightning occurrence probability analysis model. The early warning module is used to issue early warning information to base management personnel according to different risk levels based on the prediction results of the data processing and analysis module. The linkage protection module is used to automatically activate the corresponding lightning protection equipment and emergency protection measures within the base after receiving early warning information; The expression for the overall lightning occurrence probability analysis model is as follows: ; This represents the probability of lightning occurring. in, The rate of change of atmospheric electric field intensity is expressed as: , For a unit of time, This is the change in atmospheric electric field strength. As a reference rate of change of electric field, Used to reflect the effect of changes in electric field intensity on lightning. For lightning location density, Used to enhance the impact of high-density areas For reference lightning density, Temperature at the top of the cumulonimbus cloud. As the reference temperature, For temperature sensitivity coefficient, Used to simulate the properties of low temperature promoting lightning formation This is a curve showing the change in the environmental temperature and humidity abrupt change index over time. Used to accumulate the impact of environmental changes over a period of time. The start time, The current moment; , , , The weighting coefficients are determined based on historical data analysis. The process for determining different hazard levels is as follows: The overall probability of lightning occurrence obtained Compared with the preset lightning warning range Compare; when When, it is judged as low risk; when When, it is judged as medium risk; when At that time, it was determined to be high risk; The early warning module also includes: The early warning information verification unit can divide the storage base into several key sub-regions based on the base's geographic information system (GIS), analyze the lightning occurrence probability of each key sub-region, and analyze whether the overall lightning occurrence probability risk level is true based on the lightning occurrence probability of each key sub-region. If it is true, the early warning information is issued normally; otherwise, the current risk level is adjusted before issuing the early warning information.

2. The intelligent lightning early warning system for large-scale oil and gas storage bases according to claim 1, characterized in that, The formula for calculating the lightning occurrence probability of each key sub-region is as follows: ; in, The lightning location density for the current key sub-region, This represents the rate of change of atmospheric electric field intensity in the current key sub-region. This represents the risk weight coefficient for the current key sub-region.

3. The intelligent lightning early warning system for large-scale oil and gas storage bases according to claim 2, characterized in that, The probability of lightning occurrence in each key sub-region After sorting in descending order, the critical sub-region ranked first is determined to be the critical sub-region with the highest probability of lightning occurrence; The corresponding Compared with the preset lightning warning range After comparison, the probability risk level of lightning occurrence in the current key sub-region is determined; If the lightning probability risk level of the current key sub-region is consistent with the overall lightning probability risk level, then the judgment is true; If the lightning probability risk level of the current key sub-region is inconsistent with the overall lightning probability risk level, then the absolute difference between the lightning probability of the current key sub-region and the overall lightning probability is calculated. If the absolute difference is greater than the risk level difference threshold, then the lightning probability risk level of the current key sub-region is taken as the overall lightning probability risk level; otherwise, the overall lightning probability risk level remains unchanged.

4. The intelligent lightning early warning system for large-scale oil and gas storage bases according to claim 1, characterized in that, The formula for calculating the environmental temperature and humidity abrupt change index is as follows: ,in, For the rate of temperature change, This represents the rate of change in humidity.

5. The intelligent lightning early warning system for large-scale oil and gas storage bases according to claim 1, characterized in that, The emergency protection measures of the linkage protection module include: During low-risk warnings, equipment inspection robots are activated to conduct infrared thermal imaging inspections on lightning protection equipment. When a medium-risk warning is issued, the exposed interfaces of the oil and gas storage tank area are automatically shut off, and the exposed interfaces are purged with nitrogen. When a high-risk warning is issued, a one-click power-off procedure is triggered to cut off the power supply to all non-emergency equipment and activate the uninterruptible power supply (UPS) to ensure the operation of the fire protection and monitoring systems.

Images