Event impact assessment management system based on artificial intelligence

Abstract

The invention discloses an event impact assessment management system based on artificial intelligence, and relates to the field of artificial intelligence, the system comprises a management center, the management center is in communication connection with a data acquisition module, a data processing module, a data assessment module and an information feedback module; the data acquisition module is used for performing data acquisition on a target storage tank area to obtain corresponding area information; the data processing module is used for performing data processing on the acquired regional information and constructing a corresponding risk prediction model according to a processing result; the data evaluation module is used for carrying out risk evaluation on the target storage tank area according to the constructed risk prediction model; obtaining a corresponding risk assessment result; the information feedback module is used for performing event influence prediction based on a risk assessment result and performing information feedback; according to the invention, the intellectualization and real-time performance of risk assessment and event influence prediction are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an event impact assessment management system based on artificial intelligence. Background Art

[0002] Traditional fire risk assessment methods usually include regular manual inspections, maintenance log records, and risk management judgments based on experience. These methods are not only time-consuming and labor-intensive, but also have certain limitations in data processing and analysis, making it difficult to accurately assess the risk of tank fires in real time after sudden earthquake events; these are the problems we need to solve, and for this reason we provide an event impact assessment management system based on artificial intelligence. Summary of the invention

[0003] The purpose of the present invention is to provide an event impact assessment management system based on artificial intelligence; The object of the present invention can be achieved by the following technical solutions: An event impact assessment management system based on artificial intelligence, characterized in that it includes a management center, and the management center is communicatively connected with a data acquisition module, a data processing module, a data assessment module and an information feedback module; The data acquisition module is used to collect data on the target storage tank area to obtain corresponding area information; The data processing module is used to process the collected regional information and build a corresponding risk prediction model based on the processing results; The data assessment module is used to perform risk assessment on the target storage tank area according to the constructed risk prediction model; and obtain corresponding risk assessment results; The information feedback module is used to predict the impact of events based on risk assessment results and to provide information feedback.

[0004] Furthermore, the data acquisition module collects data from the target tank area, and the process of obtaining corresponding area information includes: The data acquisition module is provided with a data capture node and a data collection node; The data capture node is used to capture data related to historical earthquakes in the target storage tank area to obtain corresponding historical earthquake data and fire data; The data acquisition node is used to collect data on the storage tank information in the target storage tank area to obtain corresponding storage tank data; The historical earthquake data and storage tank data obtained are summarized to obtain the corresponding regional information, and then uploaded to the management center for storage.

[0005] Furthermore, the process of the data processing module performing data processing on the collected regional information includes: Read the collected regional information and obtain the seismic data in the corresponding regional information; Obtain the tank data corresponding to each tank at different times during the corresponding earthquake cycle, and compare them to obtain the difference status of the corresponding tank data; Counting the difference states corresponding to all storage tanks, dividing them based on the types of materials stored in the tank data, the tank structure and other information, and correlating them with the seismic data at the corresponding time to obtain a corresponding correlation map; Based on the collected regional information, fire data corresponding to fire events within several earthquake cycles are obtained; based on the fire data, corresponding fire events are summarized and processed to obtain corresponding fire information, which includes but is not limited to the patterns and laws of the occurrence and spread of earthquake-induced fires.

[0006] Furthermore, the process of constructing a corresponding risk prediction model based on the processing results includes: An initial risk model is constructed based on deep learning technology, and corresponding test sets and training sets are constructed based on the association map and fire data. Then, based on the training set and the test set, a verification test is performed on the training of the corresponding initial risk model until it meets the requirements, thereby obtaining a corresponding risk prediction model.

[0007] Furthermore, the input parameters of the risk prediction model are geological data, environmental factors such as weather, and storage tank data; the risk prediction model is used to determine whether an earthquake will occur based on the input geological data. If not, no other operations are performed. If so, the probability of an earthquake is predicted based on the geological data, and corresponding earthquake information is output, including the earthquake location, magnitude, focal depth, etc.

[0008] Furthermore, the process of the data assessment module performing risk assessment on the target tank area according to the constructed risk prediction model includes: Based on the regional information, real-time geological information of the target tank area is obtained, and it is input into the constructed risk prediction model to obtain the corresponding earthquake probability; Setting a first threshold, comparing the obtained earthquake probability with the corresponding first threshold; and judging whether to perform risk assessment based on the comparison result; if not, not performing any other operations; if yes, obtaining corresponding earthquake information based on the output result of the risk prediction model; Based on the digital twin technology, a model of the corresponding tank area is constructed to obtain a corresponding tank twin model; and the geological data and tank data collected in real time are mapped into the tank twin model; First of all, the tank twin model is obtained by modeling the corresponding tank area based on digital twin technology. The tank data includes but is not limited to the type of stored material, tank structure, real-time status parameters of the tank and other information.

[0009] During mapping, the physical location information of the tank is accurately mapped to the corresponding spatial position in the tank twin model to ensure that the position in the virtual model is consistent with the position in the actual tank area.

[0010] The type of stored material is associated with the properties of the corresponding tank in the tank twin model so that accurate analysis can be performed based on the different properties of the stored material in subsequent simulations. For example, different stored materials may have different combustion characteristics and reactions when a fire occurs.

[0011] The tank structure information, such as the shape, size, material, etc. of the tank, is also accurately mapped to the tank twin model, so that the tank in the virtual model is as similar as possible to the actual tank in appearance and physical properties. This can more realistically reflect the response of the tank structure to events when simulating scenarios such as earthquakes and fires.

[0012] The real-time status parameters of the tank, such as temperature, pressure, liquid level, etc., are also updated in real time to the corresponding tank in the tank twin model. These parameters will affect the behavior and changes of the tank under various events during the simulation process.

[0013] Through this comprehensive and accurate mapping relationship, the tank twin model can reflect the actual situation of the target tank area as realistically as possible, providing a reliable basis for subsequent simulation and risk assessment.

[0014] After the mapping is completed, the geological information in the corresponding tank twin model is modified based on the obtained seismic information to simulate the scene state changes in the tank area under the corresponding seismic information, and information is collected for the corresponding simulation process to obtain corresponding simulation data; The obtained simulation data is read, and based on the obtained simulation data, a change state of the storage tank data corresponding to the corresponding storage tank in the corresponding simulation process is obtained.

[0015] First, when simulating the tank twin model, the changes in various parameters are recorded. These parameters include but are not limited to the location, shape, temperature, pressure and other tank data of the tank.

[0016] By comparing the differences in these data before and after the simulation, the changing state of the tank during the simulation process can be determined.

[0017] The simulation process of the corresponding storage tank is divided into cycles based on the obtained change state to obtain a number of sub-cycles.

[0018] The entire simulation process can be divided into different sub-periods according to the changing characteristics of the tank data, such as the amplitude and speed of the change. The state change of the tank in each sub-period is relatively stable, and the behavior of the tank at different stages can be analyzed in more detail.

[0019] Get the initial state of the corresponding tank in different sub-periods.

[0020] At the beginning of each sub-period, the parameters of the tank are recorded as initial states. These initial states will serve as the basis for subsequent analysis.

[0021] Based on the initial state of the corresponding sub-period, the total heat received by the storage tank in the corresponding sub-period is obtained.

[0022] By considering the impact of various factors on the tank, such as friction caused by earthquakes, heat radiation from fires, etc., the total heat received by the tank in each sub-period can be calculated. This requires combining the physical model with the actual situation for calculation.

[0023] Based on the simulation data, the triggering probability of various fire event scenarios in the target tank area under the corresponding earthquake information is obtained; and based on the simulation data, the fire probability corresponding to each tank in the target tank area under different fire event scenarios is obtained; Based on the digital twin technology, a model of the corresponding tank area is constructed to obtain the corresponding tank twin model.

[0024] The tank twin model is a digital simulation of the actual tank area, which can reflect the various characteristics of the tank and the influence of the surrounding environment.

[0025] Map the geological data and tank data collected in real time into the tank twin model.

[0026] Ensure that the data in the model is consistent with the actual situation to enable accurate simulation and analysis.

[0027] After the mapping is completed, the geological information in the twin model of the corresponding tank is modified based on the obtained seismic information to simulate the scene state changes in the tank area under the corresponding seismic information, and information is collected for the corresponding simulation process to obtain the corresponding simulation data.

[0028] By simulating the impact of an earthquake on the tank area, various data are collected for subsequent analysis.

[0029] Based on the simulation data, the triggering probability of various fire event scenarios in the target tank area under the corresponding earthquake information is obtained.

[0030] Analyze the factors related to fire events in the simulation data, such as tank rupture and leakage caused by earthquakes, and calculate the triggering probability of various fire event scenarios by combining historical data and statistical models. For example, if the simulation data shows that a certain type of storage tank is prone to leakage under a certain earthquake intensity, and the possibility of fire after leakage is high, then the triggering probability of this fire event scenario will be correspondingly high.

[0031] Then, the corresponding failure risk coefficient is obtained based on the trigger probability of each fire event scenario and the fire probability of each storage tank under the corresponding fire event scenario; Set a risk threshold range, compare the obtained risk threshold range with the corresponding failure risk coefficient, and obtain the corresponding risk level based on the comparison result.

[0032] Furthermore, the process of obtaining the failure risk coefficient includes: Read the obtained simulation data, and based on the obtained simulation data, obtain the change state of the corresponding tank data of the corresponding tank in the corresponding simulation process; Based on the obtained change state, the simulation process of the corresponding storage tank is divided into cycles to obtain a number of sub-cycles; Obtaining the initial state of the corresponding storage tank in different sub-periods; based on the initial state of the corresponding sub-period, obtaining the total heat received by the storage tank in the corresponding sub-period; Set a heat threshold and compare the total heat obtained with the corresponding heat threshold; If the total heat does not exceed the heat threshold, the total heat received by the corresponding tank is continuously monitored and re-evaluated; If the total heat exceeds the heat threshold, the corresponding storage tank i is marked as a fire storage tank, and the corresponding fire probability is obtained based on it; A heat threshold is determined based on factors such as the material of the tank, the characteristics of the stored material, and safety standards. This threshold represents the maximum heat that the tank can withstand. If this threshold is exceeded, the tank is more likely to catch fire.

[0033] The total heat obtained is compared with the corresponding heat threshold.

[0034] The total heat received by the tank calculated in each sub-period is compared with the set heat threshold.

[0035] If the total heat does not exceed the heat threshold, the total heat received by the corresponding tank is continuously monitored and re-evaluated.

[0036] If the total heat does not exceed the threshold, it means that the tank is currently relatively safe, but its heat changes still need to be continuously monitored so that potential risks can be discovered in time in subsequent simulations.

[0037] If the total heat exceeds the heat threshold, the corresponding tank i is marked as a fire tank, and the corresponding fire probability is obtained based on it.

[0038] Once the total heat exceeds the threshold, the tank is marked as a fire tank. Then, combined with other factors in the simulation data, such as the environment around the tank, the availability of fire-fighting facilities, etc., the fire probability of the tank is determined through statistical analysis and model calculation. For example, the fire probability of the current tank can be estimated based on the frequency of tank fires in similar situations in historical data.

[0039] The probability of various fire event scenarios and the fire probability under the corresponding fire scenarios are obtained, and based on them, the risk assessment of the target tank area is carried out to obtain the corresponding failure risk coefficient.

[0040] Furthermore, the information feedback module is used to predict the impact of events based on the risk assessment results, and the process of providing information feedback includes: When receiving a risk notification, obtain the trigger probability of the fire event scenario in the output result of the risk prediction model, and sort the scenarios based on it until the sorting is completed; According to the priority order, obtain the number of tanks marked as on fire in the corresponding fire event scene, and synchronously read the tank parameters corresponding to the corresponding tanks; Based on the tank parameters and in combination with the tank twin model, fire simulation is performed, and corresponding fire parameters are estimated based on the simulation results, and the results are fed back to the management center; The management center formulates a corresponding emergency response plan based on the fire parameters; The same method is used to obtain the emergency response plans corresponding to other fire event scenarios; and according to the trigger probability and risk level of the corresponding fire scenario, early warning information is sent to relevant staff; At the same time, a third threshold is set, and the burning storage tanks whose fire probability exceeds the third threshold are marked as key monitoring objects, and their corresponding storage tank positions are synchronously fed back.

[0041] Compared with the prior art, the present invention has the following beneficial effects: 1. By using deep learning technology to build a risk prediction model, it is possible to predict earthquake risks and secondary fire risks based on complex geological, meteorological and tank data, thereby improving the accuracy and scientificity of risk assessment and saving manpower and material resources; 2. By identifying potential risks early and taking preventive measures, the system helps improve the safety of the tank area and reduce direct and indirect losses caused by earthquakes or fires. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0043] like Figure 1 As shown, an event impact assessment management system based on artificial intelligence includes a management center, which is communicatively connected to a data acquisition module, a data processing module, a data assessment module, and an information feedback module; The data acquisition module is used to collect data on the target storage tank area to obtain corresponding area information; The data processing module is used to process the collected regional information and build a corresponding risk prediction model based on the processing results; The data assessment module is used to perform risk assessment on the target storage tank area according to the constructed risk prediction model; and obtain corresponding risk assessment results; The information feedback module is used to predict the impact of events based on risk assessment results and to provide information feedback.

[0044] It should be further explained that, in the specific implementation process, the data acquisition module collects data from the target tank area, and the process of obtaining corresponding area information includes: The data acquisition module is provided with a data capture node and a data collection node; The data capture node is used to capture data related to historical earthquakes in the target storage tank area to obtain corresponding historical earthquake data and fire data, wherein the historical earthquake data includes but is not limited to earthquake duration, earthquake location, magnitude, focal depth, and geological data, temperature, wind speed and other environmental factors under the earthquake state; The data acquisition node is used to collect data on the storage tank information in the target storage tank area to obtain corresponding storage tank data, which includes but is not limited to the location, type, type and chemical characteristics of the storage tank, as well as the storage tank state, storage capacity and other related tank parameters; Summarize the historical earthquake data and storage tank data, obtain the corresponding regional information, and upload it to the management center for storage; It should be further explained that, in the specific implementation process, the fire data refers to the data corresponding to the fire time caused by the earthquake; if an earthquake does not cause the fire time, only the earthquake data is captured; It should be further explained that, in the specific implementation process, the data acquisition module is provided with a fixed acquisition frequency, acquisition cycle, etc., and the data acquisition module can be implemented through multiple channels such as the interface with the earthquake monitoring center database, the database interface of the tank management agency, satellite remote sensing data, and the field sensor network.

[0045] It should be further explained that, in the specific implementation process, the data processing module processes the collected regional information, and the process of constructing a corresponding risk prediction model based on the processing results includes: Read the collected regional information, obtain the earthquake data in the corresponding regional information, and construct the earthquake cycle based on it, where the earthquake cycle refers to the duration of the corresponding earthquake process; The tank data corresponding to each tank at different times during the corresponding earthquake cycle are obtained and compared to obtain the difference state of the corresponding tank data; the difference state is used to indicate the change of the corresponding tank data before and after the corresponding earthquake cycle; for example, a large cylindrical vertical liquid storage tank may generate dynamic pressure due to liquid sloshing during an earthquake, affecting its stability; Count the difference states corresponding to all storage tanks, and classify them based on the types of materials stored in the storage tank data, storage tank structure and other information, and correlate them with the seismic data at the corresponding time to obtain a corresponding correlation map, which is used to distinguish the impact of different seismic data on different types of storage tanks; Further, fire data corresponding to fire events existing within a number of earthquake cycles are obtained based on the collected regional information; Then, the corresponding fire events are summarized and processed based on the fire data to obtain corresponding fire information, which includes but is not limited to the patterns and laws of the occurrence and spread of earthquake-induced fires; Constructing an initial risk model based on deep learning technology, constructing a corresponding test set and training set based on the association map and fire data, and then performing a verification test on the corresponding initial risk model based on the training set and the test set until it meets the requirements, thereby obtaining a corresponding risk prediction model; The input parameters of the risk prediction model are geological data, environmental factors such as weather, and storage tank data; the risk prediction model is used to determine whether an earthquake will occur based on the input geological data. If not, no other operations are performed. If so, the earthquake probability is predicted based on the geological data and corresponding earthquake information is output, including the earthquake location, magnitude, focal depth, etc.

[0046] It should be further explained that, in the specific implementation process, the data assessment module performs risk assessment on the target tank area based on the constructed risk prediction model; the process of obtaining the corresponding risk assessment result includes: Based on the data acquisition module, the geological information of the target area is obtained, and the geological information includes real-time geological data, temperature, wind speed and other environmental factors of the target storage tank area; the geological information is input into the constructed risk prediction model to obtain the corresponding earthquake probability; Setting a first threshold value, and comparing the obtained earthquake probability with the corresponding first threshold value; If the earthquake probability is less than the first threshold, no other operations are performed; If the first threshold is not less than the first threshold, obtaining corresponding earthquake information based on the output result of the risk prediction model; Furthermore, a model of the corresponding tank area is constructed based on the digital twin technology to obtain a corresponding tank twin model; and the geological data and tank data collected in real time are mapped into the tank twin model; After the mapping is completed, the geological information in the corresponding tank twin model is modified based on the obtained seismic information to simulate the scene state changes in the tank area under the corresponding seismic information, and information is collected for the corresponding simulation process to obtain corresponding simulation data; Based on the simulation data, the triggering probability of various fire event scenarios in the target tank area under the corresponding earthquake information is obtained; the fire event scenarios include sealing ring fire, full liquid surface fire, single tank fire causing chain fire of surrounding tanks, etc.; It should be further explained that, in the specific implementation process, the number of simulations is at least not less than Q times, where Q is a constant value determined by the staff according to actual needs; Taking a fire incident scenario as an example, the storage tanks in the target storage tank area are numbered and recorded as i, where i=1, 2, ..., n, n>0 and n is an integer; Based on the collected tank data, the volume and thermal radiation threshold of the corresponding tank i are obtained respectively; Read the obtained simulation data, and based on the obtained simulation data, obtain the change state of the corresponding tank data of the corresponding tank i in the corresponding simulation process, wherein the change state includes the leakage amount of the material stored in the tank i, the change of the tank radius and other related information; Based on the obtained change state, the simulation process of the corresponding tank i is divided into cycles to obtain L sub-cycles, where L is a natural number; it should be further explained that the corresponding cycle division process can be determined based on one or more indicators in the tank data, for example: division based on the spread range level of the material stored in the tank, the spread range level refers to the range of diffusion in the environment after the leaked material is released from the tank. This depends on many factors, including leakage volume, wind speed, terrain, meteorological conditions, etc. The initial state of the corresponding storage tank i in different adjacent sub-periods is obtained and marked as Si; wherein the volume and thermal radiation threshold of the corresponding storage tank i are respectively obtained based on the collected tank data; Read the obtained simulation data, and based on the obtained simulation data, obtain the change state of the corresponding tank data of the corresponding tank i in the corresponding simulation process, wherein the change state includes the leakage amount of the material stored in the tank i, the change of the tank radius and other related information; Based on the obtained change state, the simulation process of the corresponding tank i is divided into cycles to obtain L sub-cycles, where L is a natural number; it should be further explained that the corresponding cycle division process can be determined based on one or more indicators in the tank data, for example: division based on the spread range level of the material stored in the tank, the spread range level refers to the range of diffusion in the environment after the leaked material is released from the tank. This depends on many factors, including leakage volume, wind speed, terrain, meteorological conditions, etc. The initial state of the corresponding storage tank i in different adjacent sub-periods is obtained and marked as Si; wherein the volume and thermal radiation threshold of the corresponding storage tank i are respectively obtained based on the collected tank data; Read the obtained simulation data, and based on the obtained simulation data, obtain the change state of the corresponding tank data of the corresponding tank i in the corresponding simulation process, wherein the change state includes the leakage amount of the material stored in the tank i, the change of the tank radius and other related information; Based on the obtained change state, the simulation process of the corresponding tank i is divided into cycles to obtain L sub-cycles, where L is a natural number; it should be further explained that the corresponding cycle division process can be determined based on one or more indicators in the tank data, for example: division based on the spread range level of the material stored in the tank, the spread range level refers to the range of diffusion in the environment after the leaked material is released from the tank. This depends on many factors, including leakage volume, wind speed, terrain, meteorological conditions, etc. Get the initial state of the corresponding tank i in different adjacent sub-periods and mark it as Si; where, ; Where Si (L) represents the initial state of the corresponding tank i in the first sub-period, that is, the tank parameters corresponding to the tank i at the initial moment of the corresponding simulation process; Si(L-1) and DF(L-1) represent the initial state of tank i in the L-1th cycle and the changing state of tank parameters in the L-1 to L sub-cycles, respectively; Then, based on the initial state of the corresponding sub-period L, the total heat received by the storage tank i in the corresponding sub-period is obtained and recorded as Ti (L); the corresponding formula is as follows: ; In the formula, ROi represents the thermal radiation threshold corresponding to the corresponding storage tank i, which is determined by the storage tank data corresponding to the storage tank i; for example: if the stored material is flammable, the corresponding thermal radiation threshold is relatively low; Set a heat threshold and compare the total heat obtained with the corresponding heat threshold; If the total heat does not exceed the heat threshold, the total heat received by the corresponding storage tank i is continuously monitored and re-evaluated; If the total heat exceeds the heat threshold, the corresponding tank i is marked as a fire tank, and a corresponding fire probability is obtained based on it, where the fire probability refers to the proportion of the number of tanks marked as fire tanks in the tank data of the corresponding area; Obtain the probability of various fire event scenarios and the fire probability under the corresponding fire scenarios, and conduct risk assessment on the target tank area based on them to obtain the corresponding failure risk coefficient, which is recorded as SX; in, ; Where K represents the total number of fire event scenarios in the output results of the risk prediction model; P1SX and P2SX represent the trigger probability of the corresponding fire event scenario and the probability of fire under the corresponding fire event scenario, respectively; A risk threshold range is set, the obtained risk threshold range is compared with the corresponding failure risk coefficient, and a corresponding risk level is obtained according to the comparison result, wherein the risk level includes low risk, medium risk and high risk; and a risk notification is fed back to the management center.

[0047] It should be further explained that, in the specific implementation process, the information feedback module is used to predict the impact of events based on the risk assessment results, and the process of providing information feedback includes: When receiving a risk notification, obtain the trigger probability of the fire event scenario in the output result of the risk prediction model, and sort the scenarios based on it until the sorting is completed; According to the priority order, obtain the number of tanks marked as on fire in the corresponding fire event scene, and synchronously read the tank parameters corresponding to the corresponding tanks; Based on the tank parameters and in combination with the tank twin model, fire simulation is performed, and corresponding fire parameters are estimated based on the simulation results. The fire parameters include fire spread law, spread speed, property losses, including direct losses (such as damage to the tank itself and stored materials) and indirect losses (such as production interruption, obstruction of business activities), etc.; and the information is fed back to the management center; Furthermore, the management center formulates a corresponding emergency response plan based on the fire parameters, which includes but is not limited to strengthening inspections, restricting flammable items from approaching the storage tank area, and formulating emergency plans to guide relevant personnel to perform evacuation, fire extinguishing and rescue operations; The same method is used to obtain the emergency response plans corresponding to other fire event scenarios; and according to the trigger probability and risk level of the corresponding fire scenario, early warning information is sent to relevant staff through various channels such as SMS, application push, and broadcast; At the same time, a third threshold is set, and the burning storage tanks whose fire probability exceeds the third threshold are marked as key monitoring objects, and their corresponding storage tank positions are synchronously fed back.

[0048] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An event impact assessment management system based on artificial intelligence, characterized in that: It includes a management center, which is communicatively connected to a data acquisition module, a data processing module, a data evaluation module and an information feedback module; The data acquisition module is used to collect data on the target storage tank area to obtain corresponding area information; The data processing module is used to process the collected regional information and build a corresponding risk prediction model based on the processing results; The data assessment module is used to perform risk assessment on the target storage tank area according to the constructed risk prediction model; and obtain corresponding risk assessment results; The information feedback module is used to predict the impact of events based on risk assessment results and to provide information feedback.

2. The event impact assessment management system based on artificial intelligence according to claim 1 is characterized in that: The data acquisition module collects data from the target tank area, and the process of obtaining corresponding area information includes: The data acquisition module is provided with a data capture node and a data collection node; The data capture node is used to capture data related to historical earthquakes in the target storage tank area to obtain corresponding historical earthquake data and fire data; The data acquisition node is used to collect data on the storage tank information in the target storage tank area to obtain corresponding storage tank data; The historical earthquake data and storage tank data obtained are summarized to obtain the corresponding regional information, and then uploaded to the management center for storage.

3. The event impact assessment management system based on artificial intelligence according to claim 2 is characterized in that: The process of the data processing module performing data processing on the collected regional information includes: Read the collected regional information and obtain the seismic data in the corresponding regional information; Obtain the tank data corresponding to each tank at different times during the corresponding earthquake cycle, and compare them to obtain the difference status of the corresponding tank data; Counting the difference states corresponding to all storage tanks, and dividing them based on the types of materials stored in the tank data, the tank structure and other information, and correlating the division results with the seismic data at the corresponding time to obtain a corresponding correlation map; Based on the collected regional information, fire data corresponding to fire events within several earthquake cycles are obtained; based on the fire data, corresponding fire events are summarized and processed to obtain corresponding fire information, which includes the patterns and laws of the occurrence and spread of earthquake-induced fires.

4. The event impact assessment management system based on artificial intelligence according to claim 3 is characterized in that: The process of building a corresponding risk prediction model based on the processing results includes: An initial risk model is constructed based on deep learning technology, and corresponding test sets and training sets are constructed based on the association map and fire data. Then, based on the training set and the test set, a verification test is performed on the training of the corresponding initial risk model until it meets the requirements, thereby obtaining a corresponding risk prediction model.

5. The event impact assessment management system based on artificial intelligence according to claim 3 is characterized in that: The input parameters of the risk prediction model are geological data, weather and other environmental factors, and storage tank data; the risk prediction model is used to determine whether an earthquake will occur based on the input geological data. If not, no other operations are performed. If so, the earthquake probability is predicted based on the geological data and corresponding earthquake information is output, including the earthquake location, magnitude, and focal depth.

6. The event impact assessment management system based on artificial intelligence according to claim 5 is characterized in that: The process of the data assessment module performing risk assessment on the target tank area according to the constructed risk prediction model includes: Based on the regional information, real-time geological information of the target tank area is obtained, and it is input into the constructed risk prediction model to obtain the corresponding earthquake probability; Setting a first threshold, comparing the obtained earthquake probability with the corresponding first threshold; and judging whether to perform risk assessment based on the comparison result; if not, not performing any other operations; if yes, obtaining corresponding earthquake information based on the output result of the risk prediction model; Based on the digital twin technology, a model of the corresponding tank area is constructed to obtain a corresponding tank twin model; and the geological data and tank data collected in real time are mapped into the tank twin model; After the mapping is completed, the geological information in the corresponding tank twin model is modified based on the obtained seismic information to simulate the scene state changes in the tank area under the corresponding seismic information, and information is collected for the corresponding simulation process to obtain corresponding simulation data; Based on the simulation data, the triggering probability of various fire event scenarios in the target tank area under the corresponding earthquake information is obtained; and based on the simulation data, the fire probability corresponding to each tank in the target tank area under different fire event scenarios is obtained; Then, the corresponding failure risk coefficient is obtained based on the trigger probability of each fire event scenario and the fire probability of each storage tank under the corresponding fire event scenario; Set a risk threshold range, compare the obtained risk threshold range with the corresponding failure risk coefficient, and obtain the corresponding risk level based on the comparison result.

7. The event impact assessment management system based on artificial intelligence according to claim 4 is characterized in that: The process of obtaining the failure risk factor includes: Read the obtained simulation data, and based on the obtained simulation data, obtain the change state of the corresponding tank data of the corresponding tank in the corresponding simulation process; Based on the obtained change state, the simulation process of the corresponding storage tank is divided into cycles to obtain a number of sub-cycles; Obtaining the initial state of the corresponding storage tank in different sub-periods; based on the initial state of the corresponding sub-period, obtaining the total heat received by the storage tank in the corresponding sub-period; Set a heat threshold and compare the total heat obtained with the corresponding heat threshold; If the total heat does not exceed the heat threshold, the total heat received by the corresponding tank is continuously monitored and re-evaluated; If the total heat exceeds the heat threshold, the corresponding storage tank i is marked as a fire storage tank, and the corresponding fire probability is obtained based on it; The probability of various fire event scenarios and the fire probability under the corresponding fire scenarios are obtained, and based on them, the risk assessment of the target tank area is carried out to obtain the corresponding failure risk coefficient.

8. The event impact assessment management system based on artificial intelligence according to claim 4 is characterized in that: The information feedback module is used to predict the impact of events based on the risk assessment results, and the process of providing information feedback includes: When receiving a risk notification, obtain the trigger probability of the fire event scenario in the output result of the risk prediction model, and sort the scenarios based on it until the sorting is completed; According to the priority order, obtain the number of tanks marked as on fire in the corresponding fire event scene, and synchronously read the tank parameters corresponding to the corresponding tanks; Based on the tank parameters and in combination with the tank twin model, fire simulation is performed, and corresponding fire parameters are estimated based on the simulation results, and the results are fed back to the management center; The management center formulates a corresponding emergency response plan based on the fire parameters; The same method is used to obtain the emergency response plans corresponding to other fire event scenarios; and according to the trigger probability and risk level of the corresponding fire scenario, early warning information is sent to relevant staff; At the same time, a third threshold is set, and the burning storage tanks whose fire probability exceeds the third threshold are marked as key monitoring objects, and their corresponding storage tank positions are synchronously fed back.

Images