A dynamic early warning method for chemical plant risks

Abstract

The invention discloses a dynamic early warning method for chemical plant risk, comprising the following steps: using HAZOP to perform risk analysis, identifying risk factors, and screening out main risk factors; using an analytical structural model method to hierarchically divide the main risk factors to form a directed topological graph, whose hierarchical relationship reflects the importance of each risk factor; screening out risk factors that can be quantified and detected, forming risk factor indicators, scoring the risk factor indicators multiple times according to the importance of each risk factor, obtaining a scoring data set of each risk factor indicator, and then applying an OWA operator to calculate and weight the scoring data set; calculating the risk value of the risk factor indicator, performing comparative analysis on all involved risk factors, assessing the priority of the risk factor indicator, and monitoring the changes of each risk factor indicator in real time. The invention makes the early warning evaluation of chemical plants simple and easy, and monitors risks in real time.

Description

Technical Field

[0001] The invention relates to the technical field of chemical plant risk early warning, and in particular to a chemical plant risk dynamic early warning method. Background Art

[0002] Whether it is raw materials or products in petrochemical production, most of them have the characteristics of flammability, explosion, volatility, corrosiveness and toxicity. This makes the whole process of petrochemical enterprises have many potential safety risk factors in both production and storage. These risk factors threaten the life and property safety of people around at all times. If management is not strengthened, accidents will occur. Therefore, it is necessary to propose a simple and easy early warning assessment method for chemical equipment to analyze and evaluate these potential risk factors, so as to improve the safety of production processes, equipment and storage facilities.

[0003] At present, accidents occur frequently in the chemical industry, causing a large number of casualties and huge property losses. Poor risk control is one of the important reasons. For large petrochemical companies, there are a large number of risks at the same time. Various risks overlap in time and space and influence each other. The identification and monitoring of risks has become a major problem faced by companies. Summary of the invention

[0004] The present invention is completed in view of the above problems. The purpose of the present invention is to provide a dynamic early warning method for chemical plant risks, which overcomes the above defects, makes the early warning assessment of chemical plants simple and easy, and monitors the risks in real time.

[0005] To achieve the above object, the solution of the present invention is: a dynamic early warning method for chemical plant risk, comprising the following steps:

[0006] S1. Use HAZOP to conduct risk analysis on the processes and equipment in chemical production, storage and transportation, identify risk factors, and screen out the main risk factors;

[0007] S2. Use the analytical structural model method to hierarchically divide the main risk factors in S1 to form a directed topological graph, thereby reflecting the logical and hierarchical relationship between the risk factors in the chemical production and storage and transportation process, and reflecting the importance of each risk factor through the hierarchical relationship;

[0008] S3. Based on the HAZOP analysis results and directed topology graph, further screen out the risk factors that can be quantified and detected to form risk factor indicators. The risk factor indicators are scored multiple times according to the importance of each risk factor to obtain the scoring data set of each risk factor indicator. Then, the OWA operator is applied to calculate the scoring data set, and then the risk factor indicators are weighted to obtain the weight value of each risk factor indicator.

[0009] S4. Based on the screening and weighting results of S3, calculate the risk value of the risk factor indicators, conduct comparative analysis on all the risk factors involved, assess the priority of the risk factor indicators that affect the chemical production and storage and transportation process, and monitor the changes of each risk factor indicator in the chemical production and storage and transportation process in real time.

[0010] Further, in step S2, firstly, a risk factor set is established through the main risk factors, the correlation between the risk factors in the risk factor set is determined, and an adjacency matrix is ​​established:

[0011]

[0012] Where n represents the number of risk factors;

[0013] Perform a power operation on the adjacency matrix A when the following equation is satisfied:

[0014] (A+I) (k-1) ≠(A+I) k =(A+I) (k+1) =R

[0015] Where I is the identity matrix and k is a constant;

[0016] Then, the reachable matrix R is obtained, and then the reachable matrix R is decomposed and hierarchical, and the hierarchical extraction of elements is performed according to the following formula:

[0017] M(S i )={S i |R ij =1}

[0018] N(S j )={S j |R ij =1}

[0019] Among them, M(S i ) is a reachable set, representing element S i The reachable element set is the set of elements in the reachable matrix S i The set of elements corresponding to the columns of matrix elements whose rows are all 1; N(S j ) is the predecessor set, indicating the arrival element S j The element set of the reachable matrix S j The set of elements corresponding to the rows of matrix elements with all 1s in the columns;

[0020] A(S i ) represents the intersection of the reachable set and the predecessor set. When A(S i )=M(S i ), extract the S that meets this condition i, as the highest level factor set L1, the extracted rows and columns are deleted, and the next step of extraction is carried out to obtain all the hierarchical elements in turn;

[0021] Then, a hierarchical directed graph is drawn, and the extracted element set is summarized and divided into three levels, namely the top layer, the middle layer and the bottom layer, and a multi-level directed graph is drawn; finally, the ISM structure is analyzed.

[0022] Further, in step S3, after the risk factor indicators are formed, the risk factor indicators are divided into primary risk factor indicators and secondary risk factor indicators, and each primary risk factor indicator and each secondary risk factor indicator are scored multiple times, and the number of scoring times is recorded as n. After the scoring data are arranged in descending order, the scoring data sets (e0, e1, e2, ..., e n-1 ), calculate the weighted vector w:

[0023]

[0024] in, is the number of combinations after selecting i data from n-1 data;

[0025] Secondly, calculate the absolute weight value: weight the data in the scoring data set with the corresponding weight vector w to obtain the absolute weight value of each first-level risk factor indicator and each second-level risk factor indicator

[0026]

[0027] Among them, w∈[0,1], i∈[0,n-1];

[0028] Secondly, use the absolute weight value of each risk factor indicator The relative weight wi of each risk factor indicator is calculated:

[0029]

[0030] Where i = 0, 1, ..., n-1;

[0031] The final calculated weight value x ij :

[0032] xi j =wi·wi j

[0033] Among them, j is the number of secondary risk factor indicators corresponding to the primary risk factor indicator; the weight value of the secondary risk factor indicator is x ij .

[0034] Furthermore, in step S4, the risk level R and the index score V of the risk factor index are determined according to the severity of the accident caused by the risk factor and the ALARP principle.

[0035] Further, in step S4, the risk value M of the risk warning indicator is calculated according to the following formula:

[0036] r ij =V ij x ij ·R

[0037] M=∑r ij =∑V ij x ij ·R

[0038] Where: M is the risk value of the entire system, r ij is the risk value of a certain risk factor; R is the risk level of the risk indicator; V ij is the index score; x ij is the indicator weight;

[0039] Then, according to the risk value M, the priority of risk factor indicators affecting the chemical production, storage and transportation process is assessed, and the changes of various risk factor indicators in the chemical production, storage and transportation process are monitored in real time.

[0040] Furthermore, it also includes S5, dividing the risk factor indicators into static indicators and dynamic indicators, and using instruments and sensors to collect data on the dynamic indicators of the chemical plant, and presenting the collected data in a risk cloud map.

[0041] After adopting the above scheme, the beneficial effects of the present invention are:

[0042] (1) In S1, HAZOP is used to conduct risk analysis on the processes and equipment in the chemical production, storage and transportation process, identify risk factors, and screen out the main risk factors. A comprehensive analysis of the risks existing in the processes and equipment in the chemical production, storage and transportation process is conducted, and the main risk factors are screened out, making the subsequent analysis process more objective.

[0043] (2) In S2, the analytical structural model method is used to hierarchically divide the main risk factors in S1 to form a directed topological graph, thereby reflecting the logical and hierarchical relationship between the risk factors in the chemical production and storage and transportation process. The hierarchical relationship reflects the importance of each risk factor, and can intuitively reflect the logical relationship and importance between the main risk factors, thereby guiding the subsequent scoring process and reducing the subjective factors in the scoring process.

[0044] (3) The OWA operator calculates the scoring data set, reducing the negative impact of extreme values ​​in the scoring data set on the weighting results, and can ensure the scientific rationality of the weighting results.

[0045] (4) Assess the priority of risk factors that affect chemical production, storage and transportation, and monitor the changes of various risk factors in the process of chemical production, storage and transportation in real time, so as to focus on the changes of important risk factors and predict and eliminate risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0047] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] The present invention relates to a chemical plant risk dynamic early warning method, which mainly comprises the following steps:

[0049] S1. For the processes and equipment in the chemical production, storage and transportation process, the hazard and operability analysis method (HAZOP) is used to conduct risk analysis, identify risk factors, screen out the main risk factors, and apply the risk matrix method to determine the risk level, that is, in the screening process, a risk level table is generated based on the possibility of a specific risk and the severity of the accident consequences. In this embodiment, the possibility of an accident (i.e., the frequency of an accident) is used as the horizontal axis, and the severity of the accident consequences is divided into five levels: A, B, C, D, and E. These five levels are listed as the vertical axis, and the interval formed by the horizontal axis and the vertical axis is divided into four risk levels: major risk, relatively large risk, general risk, and low risk. That is, the higher the frequency of accidents and the higher the severity of accidents, the higher the corresponding risk level. The identified risks are classified into risk levels according to the risk level table, and then the main risks are identified;

[0050] S2. Use the analytical structural model method (ISM) to hierarchically divide the main risk factors in S1 to form a directed topological graph, thereby reflecting the logical and hierarchical relationship between the risk factors in the chemical production and storage and transportation process. The hierarchical relationship reflects the importance of each risk factor; specifically, determine the risk factor set, and from the risk factors, combine the process itself and the monitoring requirements, select the main risk factors, establish a risk factor set, and then, combine the actual production of the process unit and the previous experience of relevant experts to determine the correlation between the risk factors in the risk factor set, and establish an adjacency matrix as follows:

[0051]

[0052] Where n represents the number of risk factors;

[0053] Then, calculate and solve the reachable matrix, perform power operation on the adjacency matrix A, and when the following formula is satisfied, it means that R is the reachable matrix to be solved:

[0054] (A+I) (k-1) ≠(A+I) k =(A+I) (k+1) =R

[0055] Where I is the unit matrix, k = 1, 2, 3, ..., and k can be calculated using Matlab software, thereby obtaining the reachable matrix R;

[0056] Then, the hierarchical division is carried out. In order to analyze the impact of a single risk factor on the overall process and the degree of hierarchical correlation between various elements, the reachable matrix R should be decomposed and hierarchically divided, and the hierarchical extraction of elements is carried out according to the following formula:

[0057] M(S i )={S i |R ij =1}

[0058] N(S j )={S j |R ij =1}

[0059] Among them, M(S i ) is a reachable set, representing element S i The reachable element set is the set of elements in the reachable matrix S i The set of elements corresponding to the columns of matrix elements whose rows are all 1; N(S j ) is the predecessor set, indicating the arrival element S j The element set of the reachable matrix S j The set of elements corresponding to the rows of matrix elements with all 1s in the columns;

[0060] A(S i ) represents the intersection of the reachable set and the predecessor set. When A(S i )=M(S i ), extract the S that meets this condition i , as the highest level factor set L1, the extracted rows and columns are deleted, and the next step of extraction is carried out to obtain all the hierarchical elements in turn;

[0061] Then, a hierarchical directed graph is drawn, and the extracted element set is summarized and divided into three levels, namely the top layer, the middle layer and the bottom layer, and a multi-level directed graph is drawn;

[0062] Then, the ISM structure is analyzed, and the causal relationship between each level and each element can be obtained from the hierarchical directed graph. Among them, the top-level relationship is the most direct risk factor affecting the hazardousness of the process equipment, the middle-level relationship is the indirect risk factor affecting the hazardousness of the process equipment, and the middle-level relationship is also the key level that connects the upper and lower levels in the entire explanatory structure model. The bottom-level relationship is also the most basic risk factor, and its monitoring and management is also very important.

[0063] S3, component risk early warning index system and evaluation model, based on the HAZOP analysis results and directed topology diagram, and in accordance with expert experience, further screen out risk factors that can be quantified and detected to form risk factor indicators, and score the risk factor indicators multiple times according to the importance of each risk factor to obtain a scoring data set of each risk factor indicator, and then apply the OWA operator to calculate the scoring data set to weight the risk factor indicators to obtain the weight value of each risk factor indicator, thereby eliminating the negative effects of the extreme values ​​in the weakening scoring process to a certain extent; specifically, in step S3, after the risk factor indicators are formed, the risk factor indicators are divided into primary risk factor indicators and secondary risk factor indicators, where the primary risk indicator is the secondary risk factor indicator. The second-level risk factor index is a subdivision of the first-level risk factor index. Each first-level risk factor index and each second-level risk factor index are scored multiple times, and the number of scoring times is recorded as n. In this embodiment, n experts score each first-level risk factor index and each second-level risk factor index once. In this embodiment, the score range of the score is 0-10, and the score range is divided into five consecutive equal parts. The larger the score, the more important the risk factor index is. The corresponding score range is divided into five levels: generally important, important, relatively important, very important, and particularly important. After arranging the data in descending order, the score data sets (e0, e1, e2, ..., e n-1 ), calculate the weighted vector w:

[0064]

[0065] in, is the number of combinations after selecting i data from n-1 data;

[0066] Secondly, calculate the absolute weight value: weight the data in the scoring data set with the corresponding weight vector w to obtain the absolute weight value of each first-level risk factor indicator and each second-level risk factor indicator

[0067]

[0068] Among them, w∈[0,1], i∈[0,n-1];

[0069] Secondly, use the absolute weight value of each risk factor indicator The relative weight wi of each risk factor indicator is calculated:

[0070]

[0071] Where i = 0, 1, ..., n-1;

[0072] The final calculated weight value x ij :

[0073] x ij =w i ·w ij

[0074] Among them, j is the number of secondary risk factor indicators corresponding to the primary risk factor indicator; the weight value of the secondary risk factor indicator is x ij ;

[0075] S4. According to the screening and weighting results of S3, the risk value of the risk factor indicator is calculated, and all the risk factors involved are compared and analyzed to assess the priority of the risk factor indicators that affect the chemical production and storage and transportation process, and the changes of the risk factor indicators in the chemical production and storage and transportation process are monitored in real time. Specifically, in step S4, the risk value M of the risk warning indicator is calculated according to the following formula:

[0076] r ij =V ij x ij ·R

[0077] M=∑r ij =∑V ij x ij ·R

[0078] Where: M is the risk value of the entire system, r ij is the risk value of a certain risk factor; R is the risk level of the risk indicator; V ij is the index score; x ij is the indicator weight, V ij and R are obtained with reference to the safety risk classification standard. The safety risk classification standard is divided according to the severity of the accident caused by the risk factors and the minimum reasonable and feasible ALARP principle. The risk level is represented by R and the indicator score is represented by V.

[0079] S5. Divide the risk factor indicators into static indicators and dynamic indicators. For the dynamic indicators such as temperature, pressure, liquid level and flow, the existing instrument sensors in the factory are used to collect data, draw risk cloud maps, use computer databases for statistics, and establish cloud servers for data analysis to achieve dynamic monitoring. For the personnel interference risk indicators in the dynamic indicators, the distance between different operators and the monitoring point is analyzed, and the paths and operating status of the personnel are tracked to achieve the early warning effect of the impact of personnel interference. For different risk indicators, when an accident occurs, the severity of the accident consequences is considered, the actual situation on the scene is used as the benchmark, and the casualties are used as the formulation standard to define the scope of the consequences caused by different risks.

[0080] S6. Select the monitoring points involved in the dynamic indicators, conduct real-time monitoring according to the different impact ranges of different monitoring points in S4, generate a dynamic risk cloud map, and then assess the priority of risk factor indicators affecting chemical production and storage and transportation according to the risk value M, and monitor the changes of various risk factor indicators in the chemical production and storage and transportation process in real time.

[0081] The above description is only a preferred embodiment of the present invention and is not a limitation on the design of this case. Any equivalent changes made based on the design key of this case shall fall within the protection scope of this case.

Claims

1. A chemical plant risk dynamic early warning method, characterized in that: The steps include: S1. Use HAZOP to conduct risk analysis on the processes and equipment in chemical production, storage and transportation, identify risk factors, and screen out the main risk factors; S2. Use the analytical structural model method to hierarchically divide the main risk factors in S1 to form a directed topological graph, thereby reflecting the logical and hierarchical relationship between the risk factors in the chemical production and storage and transportation process, and reflecting the importance of each risk factor through the hierarchical relationship; S3. Based on the HAZOP analysis results and directed topology graph, further screen out the risk factors that can be quantified and detected to form risk factor indicators. The risk factor indicators are scored multiple times according to the importance of each risk factor to obtain the scoring data set of each risk factor indicator. Then, the OWA operator is applied to calculate the scoring data set, and then the risk factor indicators are weighted to obtain the weight value of each risk factor indicator. S4. Calculate the risk value of risk factor indicators based on the screening and weighting results of S3, conduct comparative analysis on all risk factors involved, assess the priority of risk factor indicators that affect chemical production and storage and transportation, and monitor the changes of various risk factor indicators in the chemical production and storage and transportation process in real time; In step S3, after the risk factor indicators are formed, the risk factor indicators are divided into primary risk factor indicators and secondary risk factor indicators, and each primary risk factor indicator and each secondary risk factor indicator are scored multiple times, the number of scoring times is counted as n, and the scoring data are arranged in descending order to obtain the scoring data sets (e0, e1, e2, ..., e n-1 ), calculate the weighted vector w: in, is the number of combinations after selecting i data from n-1 data; Secondly, calculate the absolute weight value: weight the data in the scoring data set with the corresponding weight vector w to obtain the absolute weight value of each first-level risk factor indicator and each second-level risk factor indicator Among them, w∈[0,1], i∈[0,n-1]; Secondly, use the absolute weight value of each risk factor indicator Calculate the relative weight w of each risk factor indicator i : Where i = 0, 1, ..., n-1; The final calculated weight value x ij : x ij =w i ·w ij Among them, j is the number of secondary risk factor indicators corresponding to the primary risk factor indicator; the weight value of the secondary risk factor indicator is x ij ; In step S4, the risk level R and the index score V of the risk factor index are determined according to the severity of the accident caused by the risk factor and the ALARP principle; In step S4, the risk value M of the risk warning indicator is calculated according to the following formula: r ij =V ij x ij ·R M=∑r iｊ =∑V ij x ij ·R Where: M is the risk value of the entire system, r ij is the risk value of a certain risk factor; R is the risk level of the risk indicator; V ij is the index score; x ij is the indicator weight; Then, the priority of risk factor indicators affecting chemical production, storage and transportation is evaluated according to the risk value M, and the changes of various risk factor indicators in the chemical production, storage and transportation process are monitored in real time; In step S2, First, establish a risk factor set through the main risk factors, determine the correlation between the risk factors in the risk factor set, and establish an adjacency matrix: Where n represents the number of risk factors; Perform a power operation on the adjacency matrix A when the following equation is satisfied: (A+I) (k-1) ≠(A+I) k =(A+I) (k+1) =R Where I is the identity matrix and k is a constant; Then we get the reachable matrix R, and then decompose and hierarchically divide the reachable matrix R, and extract the hierarchical elements according to the following formula: M(S i )={S i |R ij =1} N(S j )={S j |R ij =1} Among them, M(S i ) is a reachable set, representing element S i The reachable element set is the set of elements in the reachable matrix S i The set of elements corresponding to the columns of matrix elements whose rows are all 1; N(S j ) is the predecessor set, indicating the arrival element S j The element set of the reachable matrix S j The set of elements corresponding to the rows of matrix elements with all 1s in the columns; A(S i ) represents the intersection of the reachable set and the predecessor set. When A(S i )=M(S i ), extract the S that meets this condition i , as the highest level factor set L1, the extracted rows and columns are deleted, and the next step of extraction is carried out to obtain all the hierarchical elements in turn; Then, a hierarchical directed graph is drawn to summarize and divide the extracted element set into three levels, namely the top layer, the middle layer and the bottom layer, and a multi-level directed graph is drawn; finally, the ISM structure is analyzed; S5. Divide the risk factor indicators into static indicators and dynamic indicators. For the temperature, pressure, liquid level and flow in the dynamic indicators, the existing instrument sensors in the factory are used to collect data, draw risk cloud maps, use computer databases for statistics, and establish cloud servers for data analysis to achieve dynamic monitoring. For the risk indicators of personnel interference in the dynamic indicators, the distance between different operators and the monitoring point is analyzed, and the paths and operating status of the personnel are tracked to achieve the early warning effect of the impact of personnel interference. For different risk indicators, when an accident occurs, the severity of the accident consequences is considered, and the actual situation on the scene is used as the benchmark, and the casualties are used as the formulation standard to define the scope of the consequences caused by different risks; S6. Select the monitoring points involved in the dynamic indicators, conduct real-time monitoring according to the different impact ranges of different monitoring points in S4, generate a dynamic risk cloud map, and then assess the priority of risk factor indicators affecting chemical production and storage and transportation according to the risk value M, and monitor the changes of various risk factor indicators in the chemical production and storage and transportation process in real time.

Images