Chemical industrial park vulnerability analysis model deduction method based on cellular automaton

Abstract

The invention discloses a chemical industry park vulnerability analysis model deduction method based on a cellular automaton. The method comprises the steps of obtaining a chemical industry park data set; preprocessing the chemical industry park data set to obtain a preprocessed chemical industry park data set; training a cellular automaton initial model based on the preprocessed chemical industry park data set to obtain a chemical industry park vulnerability analysis model; setting different disaster scenes; deducing response and change processes of the chemical industrial park under different disaster situations based on the chemical industrial park vulnerability analysis model; through a cellular automaton model and targeted data set preprocessing and training, the performance of the chemical industry park in different disaster scenes can be simulated more accurately, so that the accuracy of vulnerability analysis is improved; different disaster situations are set for deduction, various possible disaster situations and the influence of the disaster situations on the chemical industry park can be comprehensively considered, and the risk and vulnerability can be evaluated more comprehensively.

Description

Technical Field

[0001] The invention relates to the field of chemical park vulnerability analysis, in particular to a chemical park vulnerability analysis model deduction method based on cellular automata. Background Art

[0002] Cellular Automata (CA) is a grid dynamics model in which time, space, and state are all discrete, spatial interactions and temporal causal relationships are local, and it has the ability to simulate the spatiotemporal evolution of complex systems. It consists of a regular grid of cells, each of which takes a finite number of discrete states, follows the same rules of action, and is updated synchronously according to certain local rules. This model can simulate a variety of complex dynamic behaviors, including self-organization, self-replication, and other phenomena.

[0003] The safety of chemical parks has always been the focus of public attention. With the rapid development of the economy and the acceleration of the industrialization process, chemical parks are complex systems that contain a variety of chemical substances, equipment, pipelines, etc. The interactions between them are complex, resulting in inaccurate vulnerability analysis of chemical parks; the vulnerability of chemical parks mainly stems from the complexity of their internal systems, the interference of the external environment, and the interaction within the system. This vulnerability may cause the system to fail or produce unpredictable consequences when it is subjected to external shocks. To conduct a vulnerability analysis of a chemical park, it is necessary to comprehensively consider the interactions between its internal components, the interference of the external environment, and the overall stability of the system.

[0004] In the existing technology, the multi-hazard coupling vulnerability assessment methods of chemical parks mainly include the index system method, the vulnerability curve method, the layer overlay method and the cellular automation method, which can quantitatively analyze the multi-hazard coupling vulnerability from multiple angles. The former methods have certain limitations to a certain extent:

[0005] 1. Indicator system method

[0006] 1) Subjectivity of indicator selection: The core of the indicator system method is to build a reasonable indicator system, but the selection of indicators is often affected by the subjective judgment of researchers, which may lead to inaccurate evaluation results.

[0007] 2) Determination of indicator weights: The weight allocation of different indicators in the evaluation system is a complex issue. There is currently no unified standard or method to determine it, which will also affect the objectivity of the evaluation results.

[0008] 3) Limitation of assessment scale: The indicator system method is usually applicable to regional scale assessments. It may not be detailed enough for local scale or detailed scale assessments, and it is difficult to analyze in detail the physical vulnerability indicators of specific disaster-prone objects such as building units.

[0009] 2. Fragility Curve Method

[0010] 1) Difficulty in data acquisition: The fragility curve method requires a large amount of historical disaster data and disaster-prone body loss data to construct the fragility curve, but these data are often difficult to obtain or incomplete, which limits the scope of application of this method.

[0011] 2) Limitations of disaster types: The vulnerability curves of different disasters may have different forms and parameters, so this method may not be applicable to all types of disaster assessments.

[0012] 3) Curve fitting error: When constructing the fragility curve, it is necessary to select a suitable mathematical model for fitting, but errors may exist in the fitting process, which may affect the accuracy of the assessment results.

[0013] 3. Layer Overlay

[0014] 1) Complexity of layer overlay: The layer overlay method requires overlay analysis of multiple layers, but the overlay relationship between different layers may be very complex and requires professional GIS technology and knowledge to support.

[0015] 2) Differences in the impact of disturbances: This method does not fully consider the differences in the impact of different disturbances on the vulnerability of the carrier, which may lead to deviations in the assessment results.

[0016] 3) Difficulty in interpreting results: The results of the layer overlay method are usually presented in a graphical manner, but it is a challenge to accurately interpret these graphical results and convert them into practical disaster risk management measures.

[0017] Cellular automata, which reflect the dynamic changes of cellular states based on transition rules, can be applied to vulnerability assessment and can well characterize the dynamic evolution of system vulnerability under the continuous action of multiple disturbances. The specific advantages are as follows:

[0018] 1. Ability to simulate complex systems: Cellular automata can simulate the spatiotemporal evolution of complex systems, which makes it suitable for systems with complex internal structures and interactions, such as chemical parks. By constructing a suitable cellular automaton model, the interaction between the components within the chemical park and the interference of the external environment can be simulated, thereby revealing the vulnerability of the system.

[0019] 2. Discrete states and local rules make it easy to build models and solve them: The discrete states and local rules of cellular automata make the models easy to build and solve. In the vulnerability analysis of chemical parks, the various components of the system can be regarded as cells, their states (such as normal operation, failure, etc.) can be regarded as discrete states, and local rules can be formulated according to the interaction rules of the system. In this way, the vulnerability of the system can be revealed by simulating the evolution of cells.

[0020] 3. Easy to conduct visual analysis: The cellular automation model can also provide visual analysis methods. Through the graphical display of simulation results, the evolution process of the system and the distribution of vulnerability can be intuitively observed. This helps decision makers better understand the vulnerability of the system and take corresponding measures to reduce risks.

[0021] In the process of multi-hazard coupled vulnerability assessment of chemical parks, how to deduce the chemical park vulnerability analysis model based on cellular automata and improve the accuracy of the chemical park vulnerability analysis model has become an urgent problem that needs to be solved.

[0022] Therefore, a chemical park vulnerability analysis model deduction method based on cellular automata is urgently needed to solve the above problems. Summary of the invention

[0023] The present invention aims to solve at least one of the technical problems in the above-mentioned technology to a certain extent. To this end, the purpose of the present invention is to propose a method for deducing a chemical park vulnerability analysis model based on cellular automata, which can more accurately simulate the performance of the chemical park under different disaster scenarios through the cellular automata model and targeted data set preprocessing and training, thereby improving the accuracy of vulnerability analysis; setting different disaster scenarios for deduction can comprehensively consider various possible disaster situations and their impact on the chemical park, which is conducive to a more comprehensive assessment of risks and vulnerabilities.

[0024] To achieve the above-mentioned purpose, the embodiment of the present invention proposes a method for deriving a chemical park vulnerability analysis model based on cellular automata, comprising:

[0025] Get the chemical industry park dataset;

[0026] Preprocessing the chemical park data set to obtain a preprocessed chemical park data set;

[0027] Based on the preprocessed chemical park data set, the cellular automaton initial model is trained to obtain a chemical park vulnerability analysis model;

[0028] Set up different disaster scenarios;

[0029] Based on the chemical park vulnerability analysis model, the response and change process of the chemical park under different disaster scenarios are deduced.

[0030] Preferably, before obtaining the chemical park data set, the method further includes:

[0031] Determine the specific objectives of conducting a vulnerability analysis model for a chemical park based on cellular automata; the specific objectives include at least one of assessing the overall vulnerability of the chemical park, identifying key vulnerable areas, and predicting disaster spread paths.

[0032] Preferably, obtaining a chemical park data set includes:

[0033] Obtain the planning documents, equipment list and chemical safety data sheet of the chemical park to obtain the first data;

[0034] Acquire the spatial layout information of the chemical park based on geographic information system technology to obtain second data;

[0035] Obtain historical disaster data of the chemical park to obtain the third data;

[0036] The first data, the second data and the third data are combined to obtain a chemical park data set.

[0037] Preferably, the chemical park data set is preprocessed to obtain a preprocessed chemical park data set, including:

[0038] Performing data cleaning on the chemical park data set to obtain a cleaned chemical park data set;

[0039] The cleaned chemical park dataset is used as the preprocessed chemical park dataset.

[0040] Preferably, the chemical park data set is cleaned to obtain a cleaned chemical park data set, including:

[0041] Evenly dividing the chemical park data set into a plurality of chemical park sub-data sets;

[0042] Take any chemical park sub-dataset as the first data set;

[0043] Calculate the mean of the data values ​​corresponding to the data points in the first data set to obtain a first mean;

[0044] Calculate the mean of the data values ​​corresponding to the data points in all other chemical park sub-datasets except the first data set in the chemical park data set to obtain a second mean;

[0045] Calculating a first ratio of the first mean to the second mean, and when it is determined that the first ratio is greater than or equal to a first preset ratio threshold, taking a first data set corresponding to the first ratio as an abnormal data set;

[0046] Traverse all chemical park sub-datasets in the chemical park dataset to obtain several abnormal data sets;

[0047] Take any abnormal data set as the second data set;

[0048] Obtaining sequence data corresponding to the second data set;

[0049] Calculate the absolute value of the difference between the data values ​​corresponding to any two adjacent data points in the sequence data as the first absolute value;

[0050] Calculate the data fluctuation value corresponding to each data point based on the first absolute value to obtain a plurality of data fluctuation values;

[0051] Calculate the mean of several data fluctuation values ​​to obtain the data fluctuation mean;

[0052] Calculating a second ratio of the data fluctuation value corresponding to each data point to the data fluctuation mean, and when it is determined that the second ratio is greater than or equal to a second preset ratio threshold, taking the data point corresponding to the second ratio as an abnormal data point;

[0053] Traverse all abnormal data sets and obtain several abnormal data points;

[0054] Acquire data types corresponding to a number of abnormal data points, and classify the number of abnormal data points based on the data types to obtain a number of abnormal data classifications;

[0055] Obtain the target data cleaning rules corresponding to each abnormal data classification;

[0056] Data cleaning is performed on each abnormal data classification based on the target data cleaning rules corresponding to each abnormal data classification to obtain the cleaned chemical park data set.

[0057] Preferably, the cellular automaton initial model is trained based on the preprocessed chemical park data set to obtain a chemical park vulnerability analysis model, including:

[0058] Based on the preprocessed chemical park data set, the cellular automaton initial model is trained to obtain an initial chemical park vulnerability analysis model;

[0059] Acquire a chemical park verification data set; the verification data set includes a disaster scenario data set and an expected deduction result data set;

[0060] Randomly select a verification data from the chemical park verification data set as the target verification data;

[0061] Inputting the disaster scenario data in the target verification data into the initialization chemical park vulnerability analysis model to obtain the chemical park vulnerability analysis deduction results;

[0062] Obtain the target deduction similarity between the vulnerability analysis deduction results of the chemical park and the expected deduction results corresponding to the disaster scenario data in the target verification data;

[0063] Determine the target scoring weight based on the target deduction similarity; obtain the target indicator characteristics and target indicator quantity in the chemical park vulnerability analysis deduction results;

[0064] Determine the target deduction result score based on the target score weight, target indicator characteristics and target indicator quantity;

[0065] Traverse all the verification data in the chemical park verification data set to obtain several target deduction result scores;

[0066] The average of the scores of several target deduction results is used as the evaluation value for initializing the vulnerability analysis model of the industrial park;

[0067] The evaluation value of the initialized chemical park vulnerability analysis model is compared with a preset model evaluation threshold, and when it is determined that the evaluation value of the initialized chemical park vulnerability analysis model is greater than or equal to the preset model evaluation threshold, a chemical park vulnerability analysis model is obtained.

[0068] Preferably, the method for constructing the initial model of the cellular automaton includes:

[0069] Based on the preprocessed chemical park data set, determining the size and shape of the cell and the state space of the cell;

[0070] Define the interaction rules between cells;

[0071] Set initial conditions;

[0072] According to the size and shape of cells, the state space of cells, the interaction rules between cells and the initial conditions, the model is constructed based on the cellular automaton software to obtain the initial model of the cellular automaton.

[0073] Preferably, after iteratively training the cellular automaton initial model on the preprocessed chemical park data set to obtain the initial chemical park vulnerability analysis model, the method further includes:

[0074] Evaluate the model parameters in the initial industrial park vulnerability analysis model based on the sensitivity analysis method; the evaluation includes the impact of the evaluation parameters on the deduction results;

[0075] The model parameters are adjusted based on the evaluation results.

[0076] Preferably, after the response and change process of the chemical park under different disaster scenarios are deduced based on the chemical park vulnerability analysis model, the method further includes:

[0077] After deducing the response and change process of the chemical park under different disaster scenarios based on the chemical park vulnerability analysis model, the deduction results of the response and change process of the chemical park under different disaster scenarios are obtained;

[0078] Record key data during the simulation;

[0079] The deduction results and key data in the deduction process are visualized.

[0080] Preferably, after the response and change process of the chemical park under different disaster scenarios are deduced based on the chemical park vulnerability analysis model, the method further includes:

[0081] Verify and correct the deduction results to obtain verified and corrected deduction results;

[0082] Based on the verified and revised simulation results, the overall vulnerability of the chemical park is assessed to identify the key vulnerable areas and potential risk points of the chemical park;

[0083] Compare the verified and revised simulation results under different disaster scenarios and analyze the interactions and impacts between different disasters;

[0084] Based on the key vulnerable areas, potential risk points and the interactions and impacts between different disasters in the chemical park, determine the corresponding risk management measures and disaster response strategies.

[0085] The present invention discloses a chemical park vulnerability analysis model deduction method based on cellular automata. Through the cellular automata model and targeted data set preprocessing and training, the performance of the chemical park under different disaster scenarios can be simulated more accurately, thereby improving the accuracy of vulnerability analysis; setting different disaster scenarios for deduction can comprehensively consider various possible disaster situations and their impact on the chemical park, which is conducive to a more comprehensive assessment of risks and vulnerabilities; understanding the response and change process of the chemical park under different situations is conducive to the early formulation of targeted risk management measures and disaster response strategies, and improving the safety and response capabilities of the park; based on accurate analysis results, resources can be more reasonably allocated to strengthen key areas, prevent potential risk points, and improve resource utilization efficiency.

[0086] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0087] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0089] Figure 1 is a flow chart of a method for deducing a chemical park vulnerability analysis model based on cellular automata according to an embodiment of the present invention;

[0090] Figure 2 is a flowchart of obtaining a chemical park data set according to an embodiment of the present invention;

[0091] Figure 3 It is a flowchart of a method for constructing an initial model of a cellular automaton according to an embodiment of the present invention. DETAILED DESCRIPTION

[0092] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0093] Example 1

[0094] like Figure 1 As shown in the figure, a method for deriving a vulnerability analysis model for a chemical park based on cellular automata includes S1-S5:

[0095] S1: Get the chemical park dataset;

[0096] S2: preprocessing the chemical park data set to obtain a preprocessed chemical park data set;

[0097] S3: training the cellular automaton initial model based on the preprocessed chemical park data set to obtain a chemical park vulnerability analysis model;

[0098] S4: Set different disaster scenarios;

[0099] S5: Based on the chemical park vulnerability analysis model, the response and change process of the chemical park under different disaster scenarios are simulated.

[0100] In this embodiment, the chemical park data set includes collecting basic data from sources such as planning documents, equipment lists, and chemical safety data sheets (MSDS) of the chemical park; using geographic information system (GIS) technology to obtain spatial layout information of the chemical park, including the geographical location and attributes of buildings, equipment, pipelines, etc.; collecting historical disaster data, including disaster type, occurrence time, impact range, loss situation, etc., for analyzing disaster laws and patterns.

[0101] In this embodiment, the chemical park data set is preprocessed, and the preprocessing includes but is not limited to data noise reduction and data cleaning.

[0102] In this embodiment, the initial cellular automaton model is constructed by using programming tools (such as Python, MATLAB, etc.) or professional cellular automaton software to construct the model.

[0103] In this embodiment, different disaster scenarios include fire, explosion, leakage, etc.

[0104] The beneficial effects of the above technical solution are: through the cellular automaton model and targeted data set preprocessing and training, the performance of the chemical park under different disaster scenarios can be simulated more accurately, thereby improving the accuracy of vulnerability analysis; setting different disaster scenarios for deduction can comprehensively consider various possible disaster situations and their impact on the chemical park, which is conducive to a more comprehensive assessment of risks and vulnerabilities; understanding the response and change process of the chemical park under different scenarios is conducive to the early formulation of targeted risk management measures and disaster response strategies, and improving the safety and response capabilities of the park; based on accurate analysis results, resources can be more reasonably allocated to strengthen key areas, prevent potential risk points, and improve resource utilization efficiency.

[0105] Example 2

[0106] Before obtaining the chemical park dataset, it also includes:

[0107] Determine the specific objectives of conducting a vulnerability analysis model for a chemical park based on cellular automata; the specific objectives include at least one of assessing the overall vulnerability of the chemical park, identifying key vulnerable areas, and predicting disaster spread paths.

[0108] In this embodiment, the method of determining the specific goals of conducting the chemical park vulnerability analysis model deduction based on cellular automata includes communicating with stakeholders such as chemical park managers and safety experts, clarifying the specific goals of the deduction, and formulating a detailed deduction plan and schedule based on the goals to ensure the orderly progress of the deduction process.

[0109] The beneficial effects of the above technical scheme are: by determining specific goals, the research and analysis work has a clear direction and avoids blindness; whether it is evaluating the overall vulnerability, identifying key areas or predicting the diffusion path, it can conduct in-depth analysis on the key aspects of chemical park safety and improve the efficiency and effectiveness of the analysis; it covers multiple important dimensions of chemical park vulnerability, which helps to fully understand the safety status of the park from different angles; the identification of key vulnerable areas can provide a basis for subsequent planning and improvement, and prepare response measures in advance; the prediction of disaster diffusion paths can help to formulate more reasonable emergency plans and improve the ability and efficiency of disaster response; overall it helps to improve the management and supervision of chemical parks and ensure the safe and stable operation of the parks.

[0110] Example 3

[0111] Obtaining a chemical park data set includes steps S11-S14:

[0112] S11: Obtain planning documents, equipment lists and chemical safety data sheets of the chemical park to obtain first data;

[0113] S12: Acquire spatial layout information of the chemical park based on geographic information system technology to obtain second data;

[0114] S13: Acquire historical disaster data of the chemical park to obtain third data;

[0115] S14: Merge the first data, the second data and the third data to obtain a chemical park data set.

[0116] In this embodiment, the spatial layout information of the chemical park includes the geographical location and attribute information of buildings, equipment, pipelines, etc.

[0117] In this embodiment, the historical disaster data of the chemical park includes disaster type, occurrence time, impact range, loss situation, etc.

[0118] The beneficial effects of the above technical scheme are as follows: by obtaining the first data such as planning documents, the second data of spatial layout information and the third data of historical disaster data, it can comprehensively cover the key information of various aspects of the chemical park and ensure the integrity and richness of the data set; planning documents and equipment lists are helpful to understand the basic situation and facilities of the park in detail; MSDS can clarify the characteristics and potential risks of chemicals; the spatial layout information obtained by using GIS technology provides intuitive data support for analyzing the rationality of the layout and risk distribution of the park, which is conducive to space-related assessment and planning; historical disaster data can provide a reference for assessing current and future risks, help discover potential problems and trends, so as to take more targeted preventive measures; different types of data are merged to form a unified chemical park data set, which is convenient for subsequent analysis and utilization and improves the efficiency of data utilization; it provides a solid scientific basis for the safety management, planning adjustment, risk prevention and control of the chemical park, and helps to make more informed decisions; the integration of multiple data can better predict and warn of possible risks and disasters, and make preparations in advance; ensuring the safe operation of the park is conducive to promoting the sustainable and healthy development of the chemical park.

[0119] Example 4

[0120] The chemical park data set is preprocessed to obtain a preprocessed chemical park data set, including:

[0121] Performing data cleaning on the chemical park data set to obtain a cleaned chemical park data set;

[0122] The cleaned chemical park dataset is used as the preprocessed chemical park dataset.

[0123] The beneficial effects of the above technical solution are: through data cleaning, noise, errors and duplicate information in the data are removed, thereby significantly improving the accuracy and reliability of the data set; the cleaned data is more standardized and consistent, making subsequent analysis and processing smoother and more effective, and improving the availability of data; avoiding misleading analysis results by erroneous or bad data, ensuring the scientificity and correctness of the analysis conclusions; providing better quality input for the model or algorithm built based on the data set, thereby improving the accuracy and performance of the model; clean data can reduce the time and computing resources spent on processing abnormal data in subsequent processing.

[0124] Example 5

[0125] The chemical park data set is cleaned to obtain a cleaned chemical park data set, including:

[0126] Evenly dividing the chemical park data set into a plurality of chemical park sub-data sets;

[0127] Take any chemical park sub-dataset as the first data set;

[0128] Calculate the mean of the data values ​​corresponding to the data points in the first data set to obtain a first mean;

[0129] Calculate the mean of the data values ​​corresponding to the data points in all other chemical park sub-datasets except the first data set in the chemical park data set to obtain a second mean;

[0130] Calculating a first ratio of the first mean to the second mean, and when it is determined that the first ratio is greater than or equal to a first preset ratio threshold, taking a first data set corresponding to the first ratio as an abnormal data set;

[0131] Traverse all chemical park sub-datasets in the chemical park dataset to obtain several abnormal data sets;

[0132] Take any abnormal data set as the second data set;

[0133] Obtaining sequence data corresponding to the second data set;

[0134] Calculate the absolute value of the difference between the data values ​​corresponding to any two adjacent data points in the sequence data as the first absolute value;

[0135] Calculate the data fluctuation value corresponding to each data point based on the first absolute value to obtain a plurality of data fluctuation values;

[0136] Calculate the mean of several data fluctuation values ​​to obtain the data fluctuation mean;

[0137] Calculating a second ratio of the data fluctuation value corresponding to each data point to the data fluctuation mean, and when it is determined that the second ratio is greater than or equal to a second preset ratio threshold, taking the data point corresponding to the second ratio as an abnormal data point;

[0138] Traverse all abnormal data sets and obtain several abnormal data points;

[0139] Acquire data types corresponding to a number of abnormal data points, and classify the number of abnormal data points based on the data types to obtain a number of abnormal data classifications;

[0140] Obtain the target data cleaning rules corresponding to each abnormal data classification;

[0141] Data cleaning is performed on each abnormal data classification based on the target data cleaning rules corresponding to each abnormal data classification to obtain the cleaned chemical park data set.

[0142] In this embodiment, the mean of the data values ​​corresponding to the data points in the first data set is calculated. Assuming that there are three data points A, B, and C in the first data set, the data value corresponding to data point A is 10; the data value corresponding to data point B is 15; and the data value corresponding to data point C is 32; then the first mean is 19.

[0143] In this embodiment, the mean of the data values ​​corresponding to the data points in all the chemical park sub-datasets except the first data set in the chemical park data set is calculated to obtain the second mean; assuming that there are three data sets E, F, and G in the chemical park data set; when the E data set is the first data set, the first mean is 19, and the mean of the data values ​​corresponding to the data points in the F and G data sets is assumed to be 20; then the first ratio is Will The ratio is compared with a first preset ratio threshold value, and the first ratio is determined When the first ratio is greater than or equal to the first preset ratio threshold, the first ratio The corresponding first data set E is taken as an abnormal data set.

[0144] In this embodiment, the first preset ratio threshold is set in advance based on industry experience.

[0145] The second preset ratio threshold is set in advance based on industry experience.

[0146] In this embodiment, the data fluctuation value corresponding to each data point is calculated based on the first absolute value to obtain a plurality of data fluctuation values;

[0147]

[0148] Among them, f i represents the data fluctuation value corresponding to the i-th data point; B i represents the data value corresponding to the i-th data point; || represents the absolute value.

[0149] In this embodiment, the ratio of the data fluctuation value corresponding to each data point to the data fluctuation mean is used as the second ratio.

[0150] In this embodiment, the target data cleaning rule corresponding to each abnormal data classification is obtained, that is, the target data cleaning rule corresponding to the data type of each abnormal data classification is obtained. Assuming that the data type corresponding to the abnormal data classification is character type, the target data cleaning rule corresponding to the character data is to remove redundant spaces, unify uppercase and lowercase, and correct obvious spelling errors.

[0151] The beneficial effects of the above technical solution are: by calculating the mean and ratio of the data set, it is possible to more accurately find the data subsets and specific data points that may have anomalies; further analyzing the data fluctuation value of the abnormal data set can more accurately discover the specific characteristics of the data anomaly; based on the data type of the abnormal data point, classification is performed and corresponding specific cleaning rules are given, thereby improving the pertinence and effectiveness of data cleaning; effectively removing abnormal data, the quality of the chemical park data set is significantly improved, and a more reliable data foundation is provided for subsequent analysis and application; it can adapt to the complexity and diversity of the chemical park data set, and can flexibly apply corresponding cleaning methods according to different data characteristics and abnormal situations; after high-quality data cleaning, it can better guarantee the accuracy and reliability of various analyses and model construction based on the data set; this step-by-step and refined cleaning method is easy to expand and adjust to adapt to chemical park data sets of different sizes and characteristics.

[0152] Example 6

[0153] The cellular automaton initial model is trained based on the preprocessed chemical park data set to obtain a chemical park vulnerability analysis model, including:

[0154] Based on the preprocessed chemical park data set, the cellular automaton initial model is trained to obtain an initial chemical park vulnerability analysis model;

[0155] Acquire a chemical park verification data set; the verification data set includes a disaster scenario data set and an expected deduction result data set;

[0156] Randomly select a verification data from the chemical park verification data set as the target verification data;

[0157] Inputting the disaster scenario data in the target verification data into the initialization chemical park vulnerability analysis model to obtain the chemical park vulnerability analysis deduction results;

[0158] Obtain the target deduction similarity between the vulnerability analysis deduction results of the chemical park and the expected deduction results corresponding to the disaster scenario data in the target verification data;

[0159] Determine the target scoring weight based on the target deduction similarity; obtain the target indicator characteristics and target indicator quantity in the chemical park vulnerability analysis deduction results;

[0160] Determine the target deduction result score based on the target score weight, target indicator characteristics and target indicator quantity;

[0161] Traverse all the verification data in the chemical park verification data set to obtain several target deduction result scores;

[0162] The average of the scores of several target deduction results is used as the evaluation value for initializing the vulnerability analysis model of the industrial park;

[0163] The evaluation value of the initialized chemical park vulnerability analysis model is compared with a preset model evaluation threshold, and when it is determined that the evaluation value of the initialized chemical park vulnerability analysis model is greater than or equal to the preset model evaluation threshold, a chemical park vulnerability analysis model is obtained.

[0164] In this embodiment, the cellular automaton initial model is a blank model constructed based on programming tools (such as Python, MATLAB, etc.) or professional cellular automaton software.

[0165] In this embodiment, the chemical park vulnerability analysis model is obtained based on iterative training of the cellular automaton initial model, and the construction basis of the chemical park vulnerability analysis model is the use of the cellular automaton grid dynamics model.

[0166] In this embodiment, the target deduction similarity between the chemical park vulnerability analysis deduction result and the expected deduction result corresponding to the disaster scenario data in the target verification data is obtained, and the similarity includes but is not limited to cosine similarity and Euclidean distance similarity.

[0167] In this embodiment, the target scoring weight is determined based on the target deduction similarity, that is, the target deduction similarity is used as the target scoring weight.

[0168] In this embodiment, the target deduction result score is determined based on the target scoring weight, target indicator characteristics and target indicator quantity, that is, the total target indicator characteristic value is determined based on the target indicator characteristic value and the target indicator quantity; the product of the total target indicator characteristic value and the target scoring weight is used as the target deduction result score.

[0169] In this embodiment, the preset model evaluation threshold is set in advance based on industry experience.

[0170] The beneficial effects of the above technical solution are: by training based on the preprocessed data set, the model is ensured to have a high-quality data foundation, thereby improving the accuracy of the model's vulnerability analysis of chemical parks; the verification data set is used to verify and adjust the model so that it can better adapt to different disaster scenarios and enhance the generalization ability and reliability of the model; by calculating the target deduction similarity, determining the scoring weights and other operations, the deduction results of the model are quantitatively evaluated, which can more scientifically measure the performance of the model; targeted adjustments and improvements are made according to the evaluation results, which helps to continuously improve the quality and effect of the model; the final model is determined only when the evaluation value of the model reaches the preset threshold, ensuring the rationality and availability of the model in practical applications.

[0171] Example 7

[0172] The method for constructing the initial model of the cellular automaton includes steps S311-S314:

[0173] S311: Determine the size, shape and state space of a cell based on the preprocessed chemical park data set;

[0174] S312: Define the interaction rules between cells;

[0175] S313: Setting initial conditions;

[0176] S314: constructing a model based on cellular automaton software according to the size and shape of the cells, the state space of the cells, the interaction rules between the cells and the initial conditions, and obtaining an initial model of the cellular automaton.

[0177] In this embodiment, the state space of the cell includes normal operation, failure, and leakage; the setting of the state space is based on various situations that may occur in the chemical park. The normal operation state is the basic state; the failure state can be subdivided into different types of equipment failures or system failures; the leakage state can be further subdivided according to the type and degree of the leaked substance.

[0178] In this embodiment, the interaction rules between cells are defined including the reaction between chemical substances, the mutual influence between devices, and the like.

[0179] In this embodiment, the initial conditions include the initial cell state and the initial location where the disaster occurs.

[0180] The beneficial effects of the above technical scheme are: determining the cell-related parameters based on the pre-processed data set of a specific chemical park, so that the model is more in line with the actual situation of the park and improves its pertinence; defining interaction rules helps to accurately simulate the dynamic relationship and influence between cells, and can better reflect the complex interactions in reality; reasonably setting the initial conditions provides a reliable starting point for subsequent simulation and analysis; using cellular automaton software for construction improves the efficiency and convenience of modeling; the model constructed by this method has certain flexibility and scalability, which is convenient for subsequent adjustment and improvement according to new needs; it helps to conduct in-depth and systematic analysis of various states and changes in chemical parks, and provide strong support for decision-making.

[0181] Example 8

[0182] After iteratively training the initial cellular automaton model on the preprocessed chemical park data set to obtain the initial chemical park vulnerability analysis model, the method further includes:

[0183] Evaluate the model parameters in the initial industrial park vulnerability analysis model based on the sensitivity analysis method; the evaluation includes the impact of the evaluation parameters on the deduction results;

[0184] The model parameters are adjusted based on the evaluation results.

[0185] In this embodiment, the parameters of the model include the probability of disaster occurrence, the speed of disaster spread, and the intensity of interaction between cells.

[0186] In this embodiment, sensitivity analysis and other methods are used to evaluate the impact of parameters on the deduction results.

[0187] The beneficial effects of the above technical solution are: by evaluating the model parameters through sensitivity analysis, we can more accurately understand which parameters have a greater impact on the results, so as to make targeted adjustments to improve the accuracy of the model; by reasonably adjusting the parameters according to the evaluation results, we can continuously optimize the performance and performance of the model to make it more in line with actual conditions and needs; ensure the reliability and stability of the model in different scenarios, and reduce deviations and errors caused by unreasonable parameters; help to deeply understand the internal mechanism of the model and the role of each parameter, and provide a basis for further improvement and perfection of the model; enable the model to better adapt to various possible changes and complex situations, and enhance its adaptability and flexibility; provide a more reliable model basis for the vulnerability analysis of chemical parks, so as to better support relevant decision-making.

[0188] Example 9

[0189] After the response and change process of the chemical park under different disaster scenarios are simulated based on the chemical park vulnerability analysis model, it also includes:

[0190] After deducing the response and change process of the chemical park under different disaster scenarios based on the chemical park vulnerability analysis model, the deduction results of the response and change process of the chemical park under different disaster scenarios are obtained;

[0191] Record key data during the simulation;

[0192] The deduction results and key data in the deduction process are visualized.

[0193] In this embodiment, key data in the simulation process include the path of disaster spread, the area of ​​the affected area, the loss situation, etc.

[0194] In this embodiment, the visual display includes generating a disaster diffusion map, a loss distribution map, etc.

[0195] The beneficial effects of the above technical solution are: complex deduction results and key data are displayed in a visual way, so that people can understand and grasp the status and changes of the chemical park under different disaster scenarios more intuitively and clearly; it is convenient for relevant personnel to conduct in-depth analysis and research on responses under different scenarios and quickly find out patterns and problems; the visual display is conducive to efficient communication and information sharing between different departments or personnel, and reduces understanding bias; it provides decision makers with a stronger basis to help them make more reasonable decisions based on intuitive information; through visual display, the effectiveness and accuracy of the chemical park vulnerability analysis model can be further verified.

[0196] Example 10

[0197] After the response and change process of the chemical park under different disaster scenarios are simulated based on the chemical park vulnerability analysis model, it also includes:

[0198] Verify and correct the deduction results to obtain verified and corrected deduction results;

[0199] Based on the verified and revised simulation results, the overall vulnerability of the chemical park is assessed to identify the key vulnerable areas and potential risk points of the chemical park;

[0200] Compare the verified and revised simulation results under different disaster scenarios and analyze the interactions and impacts between different disasters;

[0201] Based on the key vulnerable areas, potential risk points and the interactions and impacts between different disasters in the chemical park, determine the corresponding risk management measures and disaster response strategies.

[0202] In this embodiment, the simulation results are verified and corrected to obtain verified and corrected simulation results. The specific implementation method is: collect historical disaster data, actual response conditions, etc. of the chemical park, and compare them with the simulation results to see whether they are consistent; invite experts in related fields to evaluate and judge the rationality and accuracy of the simulation results; use other verified similar models or methods to simulate the same scenario and compare the consistency of the results.

[0203] Based on the differences found in the verification results, analyze the model parameters that may be affected and make appropriate adjustments and optimizations; if the data is found to be insufficient or inaccurate, promptly supplement with more comprehensive and accurate data to improve the model; review the algorithms and logic in the model and promptly correct any unreasonableness; incorporate the factors that were not taken into account during the verification process into the model, and re-deduce and correct it; through repeated verification and correction processes, gradually improve the accuracy and reliability of the deduction results.

[0204] The beneficial effects of the above technical scheme are: verifying and correcting the deduction results in combination with actual conditions to make the results closer to reality and improve the accuracy and reliability of the model; being able to accurately assess the overall vulnerability of the chemical park, identify key vulnerable areas and potential risk points, and help to take targeted measures; comparing the results under different disaster scenarios, deeply analyzing the interactions and impacts between different disasters, and forming a comprehensive understanding; determining risk management measures and response strategies based on key areas, risk points and interactions, and providing strong support for scientific decision-making; facilitating the formulation of more effective disaster response plans and enhancing the ability of chemical parks to respond to various disasters; timely discovering and responding to potential risk points to reduce the possibility and severity of disasters.

[0205] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for deriving a vulnerability analysis model for a chemical park based on cellular automata, characterized in that: include: Get the chemical industry park dataset; Preprocessing the chemical park data set to obtain a preprocessed chemical park data set; Based on the preprocessed chemical park data set, the cellular automaton initial model is trained to obtain a chemical park vulnerability analysis model; Set up different disaster scenarios; Based on the chemical park vulnerability analysis model, the response and change process of the chemical park under different disaster scenarios are deduced.

2. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: Before obtaining the chemical park dataset, it also includes: Determine the specific objectives of conducting a vulnerability analysis model for a chemical park based on cellular automata; the specific objectives include at least one of assessing the overall vulnerability of the chemical park, identifying key vulnerable areas, and predicting disaster spread paths.

3. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: Get the Chemical Park Dataset, including: Obtain the planning documents, equipment list and chemical safety data sheet of the chemical park to obtain the first data; Acquire the spatial layout information of the chemical park based on geographic information system technology to obtain second data; Obtain historical disaster data of the chemical park to obtain the third data; The first data, the second data and the third data are combined to obtain a chemical park data set.

4. The method for deducing a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: The chemical park data set is preprocessed to obtain a preprocessed chemical park data set, including: Performing data cleaning on the chemical park data set to obtain a cleaned chemical park data set; The cleaned chemical park dataset is used as the preprocessed chemical park dataset.

5. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 4, characterized in that: The chemical park data set is cleaned to obtain a cleaned chemical park data set, including: Evenly dividing the chemical park data set into a plurality of chemical park sub-data sets; Take any chemical park sub-dataset as the first data set; Calculate the mean of the data values ​​corresponding to the data points in the first data set to obtain a first mean; Calculate the mean of the data values ​​corresponding to the data points in all other chemical park sub-datasets except the first data set in the chemical park data set to obtain a second mean; Calculating a first ratio of the first mean to the second mean, and when it is determined that the first ratio is greater than or equal to a first preset ratio threshold, taking a first data set corresponding to the first ratio as an abnormal data set; Traverse all chemical park sub-datasets in the chemical park dataset to obtain several abnormal data sets; Take any abnormal data set as the second data set; Obtaining sequence data corresponding to the second data set; Calculate the absolute value of the difference between the data values ​​corresponding to any two adjacent data points in the sequence data as the first absolute value; Calculate the data fluctuation value corresponding to each data point based on the first absolute value to obtain a plurality of data fluctuation values; Calculate the mean of several data fluctuation values ​​to obtain the data fluctuation mean; Calculating a second ratio of the data fluctuation value corresponding to each data point to the data fluctuation mean, and when it is determined that the second ratio is greater than or equal to a second preset ratio threshold, taking the data point corresponding to the second ratio as an abnormal data point; Traverse all abnormal data sets and obtain several abnormal data points; Acquire data types corresponding to a number of abnormal data points, and classify the number of abnormal data points based on the data types to obtain a number of abnormal data classifications; Obtain the target data cleaning rules corresponding to each abnormal data classification; Data cleaning is performed on each abnormal data classification based on the target data cleaning rules corresponding to each abnormal data classification to obtain the cleaned chemical park data set.

6. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: The cellular automaton initial model is trained based on the preprocessed chemical park data set to obtain a chemical park vulnerability analysis model, including: Based on the preprocessed chemical park data set, the cellular automaton initial model is trained to obtain an initial chemical park vulnerability analysis model; Acquire a chemical park verification data set; the verification data set includes a disaster scenario data set and an expected deduction result data set; Randomly select a verification data from the chemical park verification data set as the target verification data; Inputting the disaster scenario data in the target verification data into the initialization chemical park vulnerability analysis model to obtain the chemical park vulnerability analysis deduction results; Obtain the target deduction similarity between the chemical park vulnerability analysis deduction results and the expected deduction results corresponding to the disaster scenario data in the target verification data; Determine the target scoring weight based on the target deduction similarity; obtain the target indicator characteristics and target indicator quantity in the chemical park vulnerability analysis deduction results; Determine the target deduction result score based on the target score weight, target indicator characteristics and target indicator quantity; Traverse all the verification data in the chemical park verification data set to obtain several target deduction result scores; The average of the scores of several target deduction results is used as the evaluation value for initializing the vulnerability analysis model of the industrial park; The evaluation value of the initialized chemical park vulnerability analysis model is compared with a preset model evaluation threshold, and when it is determined that the evaluation value of the initialized chemical park vulnerability analysis model is greater than or equal to the preset model evaluation threshold, a chemical park vulnerability analysis model is obtained.

7. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 6, characterized in that: The method for constructing the initial model of cellular automaton includes: Based on the preprocessed chemical park data set, determining the size and shape of the cell and the state space of the cell; Define the interaction rules between cells; Set initial conditions; According to the size and shape of cells, the state space of cells, the interaction rules between cells and the initial conditions, the model is constructed based on the cellular automaton software to obtain the initial model of the cellular automaton.

8. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 7, characterized in that: After iteratively training the initial cellular automaton model on the preprocessed chemical park data set to obtain the initial chemical park vulnerability analysis model, the method further includes: Evaluate the model parameters in the initial industrial park vulnerability analysis model based on the sensitivity analysis method; the evaluation includes the impact of the evaluation parameters on the deduction results; The model parameters are adjusted based on the evaluation results.

9. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: After the response and change process of the chemical park under different disaster scenarios are simulated based on the chemical park vulnerability analysis model, it also includes: After deducing the response and change process of the chemical park under different disaster scenarios based on the chemical park vulnerability analysis model, the deduction results of the response and change process of the chemical park under different disaster scenarios are obtained; Record key data during the simulation; The deduction results and key data in the deduction process are visualized.

10. The method for deriving a chemical park vulnerability analysis model based on cellular automata according to claim 1, characterized in that: After the response and change process of the chemical park under different disaster scenarios are simulated based on the chemical park vulnerability analysis model, it also includes: Verify and correct the deduction results to obtain verified and corrected deduction results; Based on the verified and revised simulation results, the overall vulnerability of the chemical park is assessed to identify the key vulnerable areas and potential risk points of the chemical park; Compare the verified and revised simulation results under different disaster scenarios and analyze the interactions and impacts between different disasters; Based on the key vulnerable areas, potential risk points and the interactions and impacts between different disasters in the chemical park, determine the corresponding risk management measures and disaster response strategies.

Images