CN108171413B - Chemical industry park emergency resource allocation optimization method - Google Patents

Chemical industry park emergency resource allocation optimization method Download PDF

Info

Publication number
CN108171413B
CN108171413B CN201711433425.1A CN201711433425A CN108171413B CN 108171413 B CN108171413 B CN 108171413B CN 201711433425 A CN201711433425 A CN 201711433425A CN 108171413 B CN108171413 B CN 108171413B
Authority
CN
China
Prior art keywords
emergency
risk
source
decision
node
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201711433425.1A
Other languages
Chinese (zh)
Other versions
CN108171413A (en
Inventor
蒋鹏
门金坤
周硕
宋秋生
郑松
孔亚广
赵烨
沈刚
叶建刚
苏楠
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Dianzi University
Original Assignee
Hangzhou Dianzi University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hangzhou Dianzi University filed Critical Hangzhou Dianzi University
Priority to CN201711433425.1A priority Critical patent/CN108171413B/en
Publication of CN108171413A publication Critical patent/CN108171413A/en
Application granted granted Critical
Publication of CN108171413B publication Critical patent/CN108171413B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Business, Economics & Management (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Development Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Educational Administration (AREA)
  • General Engineering & Computer Science (AREA)
  • Game Theory and Decision Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Marketing (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Probability & Statistics with Applications (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to an emergency resource allocation optimization method for a chemical industry park. The method comprises the steps of establishing a dynamic risk assessment model of the risk source by considering the domino effect of accidents and risk factors of process flows on the basis of a traditional risk source grading model by combining the characteristics of the chemical industry park and obtaining a grading result by adopting a pedigree clustering method. Comprehensively considering rescue timeliness and economic applicability of configuration decision on the basis of the risk level of the hazard source, and establishing a multi-objective optimization emergency resource configuration model based on the risk level of the hazard source; and optimizing the multi-target emergency resource configuration model by using a genetic algorithm optimization method, and finally obtaining a configuration strategy meeting the expected target. The invention provides a configuration optimization method with strong global optimization capability aiming at some problems in emergency resource configuration optimization, and the optimization method has the characteristics of openness, robustness, parallelism, flexibility, global convergence, no special requirement on the mathematical form of problems and the like.

Description

Chemical industry park emergency resource allocation optimization method
Technical Field
The invention belongs to the technical field of information and control, relates to an automation technology, and particularly relates to an emergency resource allocation optimization method for a chemical industry park.
Background
With the continuous development of economy, the scale of the petrochemical industry is in a rapidly expanding period. Under the trend of industrial clustering development, the chemical industry park becomes a new mode of chemical industry development in China. Nowadays, chemical industry parks are basically located in sensitive areas with developed economy and dense population, so that the chemical industry parks have the characteristics of dense investment capital, dense technology, dense equipment and the like. In order to meet the production requirements, a great variety and quantity of flammable, explosive and toxic hazard sources are gathered in the chemical industry park, so that the possibility of major accidents in the chemical industry park is objectively existed. Throughout the development history of petrochemical industry, accidents caused by chemical plants at home and abroad are not rare. Emergency resources are key links for dealing with sudden accidents, and reasonable and effective configuration decisions are the premise and the basis for reducing accident loss. Although the probability of accidents occurring in the chemical industry park is higher than that in the general area, the accidents still occur with small probability in general, and the use frequency of emergency resources is not high, so that the unlimited investment of the emergency resources is not practical. The reasonable configuration decision idea needs to consider the emergency capacity of the configuration and the practical feasibility.
In the emergency system of the chemical industry park, the configuration decision of emergency resources determines the efficiency, cost and capacity of dealing with sudden accidents. At present, the configuration of emergency resources in a chemical industry park generally has a contradictory state, namely, areas with larger requirements lack emergency resources, and areas with smaller requirements have excess emergency resources. Therefore, the configuration optimization of the emergency resources has important significance for improving the emergency capacity of the park, reducing the configuration cost and realizing the optimal utilization of the resources. The method adopts the risk grade of the risk source as the basis for measuring the demand degree of the emergency resources of the region, combines the characteristics of the chemical industry park, considers the domino effect of accidents and the risk factors of the process flow on the basis of the traditional risk source grading model, establishes a dynamic risk evaluation model of the risk source, and adopts a pedigree clustering method to obtain the grading result. The configuration of the emergency resources is a multi-objective optimization problem, and a plurality of traditional mathematical optimization methods such as a simplex method, a conjugate gradient method, a geometric mean analysis method, an orthogonal design method and the like. Since these optimization methods lack robustness of global optimal search and most of the conventional optimizations require gradient information, it is very difficult to solve such a multi-objective optimization problem with complex mathematical forms.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an emergency resource allocation optimization method for a chemical industry park. The invention aims to solve the problems in emergency resource allocation optimization of a chemical industrial park, and provides an allocation optimization method with strong global optimization capability by dividing an emergency resource allocation strategy into an emergency node site selection strategy and an emergency material allocation strategy.
The method comprises the following specific steps:
step 1, acquiring dangerous source information and geographic information in a chemical industry park, process flow information of each chemical industry dangerous source engaged in production in the park, a place available for planning and the quantity of emergency materials to be distributed, wherein the dangerous source information comprises the type, state, stock and distribution condition of the dangerous source;
step 2: and (3) establishing a risk dynamic evaluation model of the risk source according to the information obtained in the step (1), wherein the considered risk source characteristic indexes are inherent risk coefficients of the risk source, domino effect risk coefficients and process flow risk coefficients.
Inherent danger coefficient of danger source
In a risk source evaluation model based on the R value, a value R obtained by correcting and summing the ratios of the actual existing quantity of various risk sources in the unit and the critical quantity specified in a file (GB18218) is used as a grading index; the calculation of the value of R is a coupling of the ease of occurrence of the accident and the severity of the consequences of the accident, and therefore this value of R represents in fact the inherent risk of the source of danger. The calculation formula of the inherent risk coefficient of the hazard source is as follows:
Figure BDA0001525339350000021
in the formula, betanIs the correction coefficient of the hazard source n, alpha is the correction coefficient of the exposure personnel outside the plant, qnIs the actual inventory of hazard n, QnThe critical quantity specified in the file (GB18218) for hazard source n.
Risk coefficient of domino effect-
The types of accidents that can cause the domino effect are mainly fires, explosions and toxic gas leaks. The damage of the secondary target equipment cannot be directly caused by the simple leakage and diffusion of the poison. Therefore, it is not discussed in the analysis of the domino effect.
The value of the risk coefficient of the domino effect is considered from the following three aspects:
1) forms of domino effect, including thermal radiation and shock wave overpressure;
2) probability of occurrence of domino effect;
3) the number of units in which a secondary accident may occur is affected by the minox effect.
The probability of the domino effect can be obtained by performing accident consequence simulation prediction on a danger source. The prediction process follows the principle of maximum risk as well as the principle of probability summation. The domino effect risk coefficient γ is calculated as follows:
Figure BDA0001525339350000031
wherein i and k represent the number of plants in which a secondary accident occurs, PblastTo impact the probability of domino effect, PheatIs the thermal radiation domino effect probability.
Risk coefficient of technological process
And (3) a method for calculating the risk coefficient of the process unit in the reference chemical fire and explosion index evaluation method. When calculating each of the process unit risk factors, selecting the most dangerous state of the material in the process unit, such as: start-up, continuous operation, and stop.
The calculation method of the process risk coefficient comprises the following steps:
F3＝F1×F2
in the formula, F1Is a general process flow risk factor, F2For a particular process risk factor, F3Is the risk coefficient of the process flow.
Step 3, the hazard source can be characterized as follows through three characteristic indexes of the hazard source:
Xi＝(xi1,xi2,xi3)T,i＝1,2,...,n
in the formula, XiIs a hazard source i, xi1、xi2And xi3Three characteristic indexes of the hazard source i are respectively represented.
The method comprises the following steps of (1) grading a hazard source sample by using a pedigree clustering method:
selecting definition of similarity measurement between samples and classes, and using distance between samples and distance between classes as similarity measurement between samples and classes.
The inter-sample distance is defined as the euclidean distance:
Figure BDA0001525339350000032
wherein m is the number of characteristic indexes of a research sample, dijIs the distance between sample i and sample j.
The inter-class distance is defined as the shortest distance, and the calculation formula is as follows:
Figure BDA0001525339350000041
in the formula, GpAnd GqAre of the classes p and q, DpqIs the distance between class p and class qAnd (5) separating.
Secondly, calculating the distance between the samples to generate a distance matrix D;
constructing each class which only contains one sample;
combining according to the definition of similarity measurement between classes to generate a new class;
calculating the distance between the new class and the existing class;
sixthly, if one category exists, drawing a pedigree clustering graph, otherwise, returning to the step four; the similarity relationship between the study samples is clearly expressed by the pedigree cluster map.
And seventhly, determining the number of the classes and obtaining a grading result.
And 4, step 4: abstracting the whole chemical industry park into a network topological structure, wherein the location of a hazard source is a demand node in a network topological graph, the location where emergency facilities and emergency materials can be stored is a potential emergency node in the network topological graph, a road connecting each node is an arc in the network topological graph, and the emergency range of the emergency node is an emergency jurisdiction; the expression of the chemical industry park network topological graph is as follows:
G＝{V,E}
in the formula, V is an arc set in the network, the weight of the arc is the arc length of the arc, E is a node set in the network, and the weight of a required node is the risk level of a hazard source;
and 5: dividing emergency resource allocation decisions into emergency node addressing decisions and emergency material allocation decisions, comprehensively considering the safety, economic applicability and rescue timeliness of the allocation decisions on the basis of risk levels of the hazard sources to establish a multi-target emergency resource allocation model, and summarizing optimization targets of the multi-target emergency resource allocation model into: the addressing strategy expects that the higher the danger weight, the larger the coverage degree of the demand node is, and simultaneously reduces the number of emergency nodes as much as possible. The distribution strategy expects that the emergency nodes with higher danger weight in the emergency jurisdiction distribute more materials and reduce rescue time as much as possible. And setting n as the number of the demand nodes, p as the number of the potential emergency nodes, and Q as the number of emergency materials to be distributed.
Safety of configuration decision
The configuration decision safety is measured by the weighted coverage degree of the demand nodes in the network and the weighted distribution degree of the emergency jurisdiction. Can be calculated by the following formula:
Figure BDA0001525339350000042
Figure BDA0001525339350000051
S＝S1+S2
in the formula, S1For addressing decision-making safety indicators, BiAnd the number of emergency nodes including the demand node i in the emergency jurisdiction. HiIs the danger weight of the demand node i. S2To assign a decision safety index, IuThe selected emergency node set is selected; kjDetermining the risk weight of the emergency node j in the emergency jurisdiction by the average risk weight of the demand nodes in the emergency jurisdiction; qjAllocating the stock of emergency materials for the emergency node j; and S is the overall safety index of emergency resource configuration decision.
② economic applicability of configuration decision
The economic applicability of the configuration decision is measured by the number of the selected emergency nodes. Can be calculated by the following formula:
Figure BDA0001525339350000052
in the formula, EA is an index of decision-making economic applicability; c, the number of the selected emergency points.
Thirdly, configuration decision rescue timeliness
The decision rescue timeliness is measured by emergency material storage in each emergency node and the average rescue distance, and the calculation method is as follows:
Figure BDA0001525339350000053
in the formula, T is the timeliness index of decision rescue, CjThe average rescue distance of the emergency node j is the average distance from the emergency node j to a demand node in the district;
and 6, optimizing the multi-target emergency resource configuration model by using a genetic algorithm optimization method, and finally solving a configuration strategy meeting the target expectation. The method comprises the following specific steps:
the method comprises the following steps that firstly, a classical binary coding mode is adopted to solve the multi-objective optimization emergency point addressing problem, continuous integers are used for endowing each potential emergency point with an ID number, for example, 10 potential emergency points exist, then the ID numbers 1 to 10 respectively represent an emergency service point, and a coding string 1011001010 represents that the potential emergency points with the ID numbers of 1,3,4,7 and 9 are selected. The distribution model adopts matrix coding; and respectively assigning an ID number to each emergency resource to be distributed and each emergency point by using continuous integers. For example, if the number of planned emergency points is 5, 20 emergency resources to be allocated exist, and an individual in the genetic algorithm is a 20-row and 5-column matrix X20×5The elements of the matrix are only 0 or 1. If the matrix element of the ith row and the jth column of the matrix is 1, the emergency resource with the ID number of i is allocated to the emergency point with the ID number of j.
And secondly, initializing, namely generating an initial population by the genetic algorithm. The population is composed of a certain number of individuals, so the pop _ size feasible solutions are randomly selected as the individuals to form the initial population pop.
And thirdly, carrying out dimensionless processing by using the extreme value of each index. And (3) constructing a genetic algorithm fitness function by combining the model expectation, wherein the fitness function of the site selection strategy is obtained by the following formula:
maximizeFx＝S1 *-EA*+C1
in the formula, FxDeciding a fitness function for site selection, S1 *Making a decision on a safety index, EA, for location after dimensionless processing*And deciding an economic applicability index for the site selection after the dimensionless processing. C1Is a constant.
The fitness function of the allocation policy is given by:
maximizeFy＝S2 *-T*+C2
in the formula, FyDeciding a fitness function for site selection, S2 *For the allocation of a decision-making safety index, T, after dimensionless processing*And deciding rescue timeliness indexes for the distribution after the non-dimensionalization processing. C2Is a constant.
The target expectation of the fitness function is all the maximum value, and the value range of the fitness function of the genetic algorithm must be greater than 0, so C1＝C2＝1。
Adding constraint conditions, in practical application, there are often many different limitations to the configuration decision of the emergency resources, for example, all demand nodes in the campus are required to be in the emergency jurisdiction or constraints are added based on performance indexes of the configuration decision: the number of emergency nodes is limited, the coverage degree is limited, and the rescue timeliness is limited. The method used for processing the constraint conditions in the genetic algorithm optimization process is to eliminate individuals which do not meet the constraint conditions, i.e. to forcibly reset the fitness of the infeasible solution to 0.
Arranging individuals in the population from small to large according to the fitness, calculating the accumulated fitness:
Figure BDA0001525339350000061
wherein, Fitness (idv)i) And (4) the fitness obtained by the calculation is carried into the fitness function for the ith individual in the current population.
Sixthly, taking a random number r belonging to (0, FitnessT (pop _ size));
seventhly if r is the same as (FitnessT (i), FitnessT (i +1)]Then individual idviSelecting the selected plants;
repeating the step (c) until POP _ size-1 individuals are selected, and assigning the selected individuals to New POP; adopting an elite policy: reserving the last individual in the POP, emptying the other individuals, and assigning all the individuals in the New POP to the POP;
generating a real number a randomly in an interval (0,1) according to a given Cross _ rate, if a < Cross _ rate, then crossing two adjacent individuals, otherwise, not crossing. For binary coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, m), and exchanging all binary strings after the hybridization position between two chromosomes. For matrix coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, Q), and exchanging all row vectors below the hybridization position between the two chromosomes.
Ninthly, randomly generating a real number a in the interval (0,1) according to a given mutation probability Mutate _ rate, and if a is less than the Mutate _ rate, performing mutation operation on the individual. For binary coding, the mutation process is: randomly generating a positive integer b as a variation position between the intervals (0, m), and negating the binary number of the position. For matrix coding, the mutation process is: randomly generating a positive integer b as a variation position between the intervals (0, Q), setting all the b-th row elements of the individual matrix to be 0, randomly generating a positive integer d between the intervals (0, c), and taking the d-th row element of the b-th row of the individual matrix as 1.
And (c) taking iteration times G as an algorithm termination condition, returning to the third step when G is less than 200, outputting an individual corresponding to the optimal fitness when G is 200, and obtaining an optimal configuration decision according to a coding mode.
Has the advantages that: the method aims at the configuration decision of the optimized emergency resources of the chemical industrial park, so that the overall emergency capacity of the park is improved, the resource waste is reduced, and the optimization method has the characteristics of openness, robustness, parallelism, global convergence, flexibility, no special requirements on the mathematical form of problems and the like.
Drawings
FIG. 1 is a pedigree clustering diagram;
FIG. 2 is a schematic diagram of a crossover process;
FIG. 3 is a schematic diagram of a mutation process;
FIG. 4 is a flow chart of the present invention.
Detailed Description
As shown in FIG. 4, step 1, obtaining information of dangerous sources in a chemical industry park, geographic information, process flow information of each chemical industry dangerous source engaged in production in the park, places available for planning and the quantity of emergency materials to be distributed, wherein the information of the dangerous sources comprises the types, states, stock and distribution conditions of the dangerous sources;
step 2: and (3) establishing a risk dynamic evaluation model of the risk source according to the information obtained in the step (1), wherein the considered risk source characteristic indexes are inherent risk coefficients of the risk source, domino effect risk coefficients and process flow risk coefficients.
Inherent danger coefficient of danger source
In a risk source evaluation model based on the R value, the sum R of the correction ratios of the actual existing quantity of various risk sources in the unit and the critical quantity specified in a file (GB18218) is used as a grading index; the calculation of the value of R is a coupling of the ease of occurrence of the accident and the severity of the consequences of the accident, and therefore this value of R represents in fact the inherent risk of the source of danger. The calculation formula of the inherent risk coefficient of the hazard source is as follows:
Figure BDA0001525339350000081
in the formula, betanIs the correction coefficient of the hazard source n, alpha is the correction coefficient of the exposure personnel outside the plant, qnIs the actual inventory of hazard n, QnThe critical quantity specified in the file (GB18218) for hazard source n.
Risk coefficient of domino effect-
The types of accidents that can cause the domino effect are mainly fires, explosions and toxic gas leaks. The damage of the secondary target equipment cannot be directly caused by the simple leakage and diffusion of the poison. Therefore, it is not discussed in the analysis of the domino effect.
The value of the risk coefficient of the domino effect is considered from the following three aspects:
1) forms of domino effect, including thermal radiation and shock wave overpressure;
2) probability of occurrence of domino effect;
3) the number of units in which a secondary accident may occur is affected by the minox effect.
The probability of the domino effect can be obtained by performing accident consequence simulation prediction on a danger source. The prediction process follows the principle of maximum risk as well as the principle of probability summation. The domino effect risk coefficient γ is calculated as follows:
Figure BDA0001525339350000082
wherein i and k represent the number of plants in which a secondary accident occurs, PblastTo impact the probability of domino effect, PheatIs the thermal radiation domino effect probability.
Risk coefficient of technological process
And (3) a method for calculating the risk coefficient of the process unit in the reference chemical fire and explosion index evaluation method. When calculating each of the process unit risk factors, selecting the most dangerous state of the material in the process unit, such as: start-up, continuous operation, and stop.
The calculation method of the process risk coefficient comprises the following steps:
F3＝F1×F2
in the formula, F1Is a general process flow risk factor, F2For a particular process risk factor, F3Is the risk coefficient of the process flow.
Step 3, the hazard source can be characterized as follows through three characteristic indexes of the hazard source:
Xi＝(xi1,xi2,xi3)T,i＝1,2,...,n
in the formula, XiIs a hazard source i, xi1、xi2And xi3Three characteristic indexes of the hazard source i are respectively represented.
The method comprises the following steps of (1) grading a hazard source sample by using a pedigree clustering method:
selecting definition of similarity measurement between samples and classes, and using distance between samples and distance between classes as similarity measurement between samples and classes.
The inter-sample distance is defined as the euclidean distance:
Figure BDA0001525339350000091
wherein m is the number of characteristic indexes of a research sample, dijIs the distance between sample i and sample j.
The inter-class distance is defined as the shortest distance, and the calculation formula is as follows:
Figure BDA0001525339350000092
in the formula, GpAnd GqAre of the classes p and q, DpqIs the distance between class p and class q.
Secondly, calculating the distance between the samples to generate a distance matrix D;
constructing each class which only contains one sample;
combining according to the definition of similarity measurement between classes to generate a new class;
calculating the distance between the new class and the existing class;
sixthly, if one category exists, drawing a pedigree clustering graph as shown in the figure 1, otherwise, returning to the step IV; the similarity relationship between the study samples is clearly expressed by the pedigree cluster map.
And seventhly, determining the number of the classes and obtaining a grading result.
And 4, step 4: abstracting the whole chemical industry park into a network topological structure, wherein the location of a hazard source is taken as a demand node in a network topological graph, the location where emergency facilities and emergency materials can be stored is taken as a potential emergency node in the network topological graph, a road connecting all the nodes is taken as an arc in the network topological graph, and the emergency range of the emergency node is taken as an emergency jurisdiction; the expression of the chemical industry park network topological graph is as follows:
G＝{V,E}
in the formula, V is an arc set in the network, the weight of the arc is the arc length of the arc, E is a node set in the network, and the weight of a required node is the risk level of a hazard source;
and 5: dividing emergency resource allocation decisions into emergency node addressing decisions and emergency material allocation decisions, comprehensively considering the safety, economic applicability and rescue timeliness of the allocation decisions on the basis of risk levels of the hazard sources to establish a multi-target emergency resource allocation model, and summarizing optimization targets of the multi-target emergency resource allocation model into: the addressing strategy expects that the higher the danger weight, the larger the coverage degree of the demand node is, and simultaneously reduces the number of emergency nodes as much as possible. The distribution strategy expects that the emergency nodes with higher danger weight in the emergency jurisdiction distribute more materials and reduce rescue time as much as possible. And setting n as the number of the demand nodes, m as the number of the potential emergency nodes, and Q as the number of emergency materials to be distributed.
Safety of configuration decision
The configuration decision safety is measured by the weighted coverage degree of the demand nodes in the network and the weighted distribution degree of the emergency jurisdiction. Can be calculated by the following formula:
Figure BDA0001525339350000101
Figure BDA0001525339350000102
S＝S1+S2
in the formula, S1For addressing decision-making safety indicators, BiAnd the number of emergency nodes including the demand node i in the emergency jurisdiction. HiIs the danger weight of the demand node i. S2To assign a decision safety index, IuThe selected emergency node set is selected; kjDetermining the risk weight of the emergency node j in the emergency jurisdiction by the average risk weight of the demand nodes in the emergency jurisdiction; qjAllocating the stock of emergency materials for the emergency node j; and S is the overall safety index of emergency resource configuration decision.
② economic applicability of configuration decision
The economic applicability of the configuration decision is measured by the number of the selected emergency nodes. Can be calculated by the following formula:
Figure BDA0001525339350000103
in the formula, the number of the emergency points selected by the step c is increased.
Thirdly, configuration decision rescue timeliness
The decision rescue timeliness is measured by emergency material storage in each emergency node and the average rescue distance, and the calculation method is as follows:
Figure BDA0001525339350000104
in the formula, T is the timeliness index of decision rescue, CjThe average rescue distance of the emergency node j is the average distance from the emergency node j to a demand node in the district;
and 6, optimizing the multi-target emergency resource configuration model by using a genetic algorithm optimization method, and finally solving a configuration strategy meeting the target expectation. The method comprises the following specific steps:
the method comprises the following steps that firstly, a classical binary coding mode is adopted to solve the multi-objective optimization emergency point addressing problem, continuous integers are used for endowing each potential emergency point with an ID number, for example, 10 potential emergency points exist, then the ID numbers 1 to 10 respectively represent an emergency service point, and a coding string 1011001010 represents that the potential emergency points with the ID numbers of 1,3,4,7 and 9 are selected. The distribution model adopts matrix coding; and respectively assigning an ID number to each emergency resource to be distributed and each emergency point by using continuous integers. For example, if the number of planned emergency points is 5, 20 emergency resources to be allocated exist, and an individual in the genetic algorithm is a 20-row and 5-column matrix X20×5The elements of the matrix are only 0 or 1. If the matrix element of the ith row and the jth column of the matrix is 1, the emergency resource with the ID number of i is allocated to the emergency point with the ID number of j.
And secondly, initializing, namely generating an initial population by the genetic algorithm. The population is composed of a certain number of individuals, so the pop _ size feasible solutions are randomly selected as the individuals to form the initial population pop.
And thirdly, carrying out dimensionless processing by using the extreme value of each index. And (3) constructing a genetic algorithm fitness function by combining the model expectation, wherein the fitness function of the site selection strategy is obtained by the following formula:
maximizeFx＝S1 *-EA*+C1
in the formula, S1 *Making a decision on a safety index, EA, for location after dimensionless processing*And deciding an economic applicability index for the site selection after the dimensionless processing. C1Is a constant.
The fitness function of the allocation policy is given by:
maximize Fy＝S2 *-T*+C2
in the formula, S2 *For the allocation of a decision-making safety index, T, after dimensionless processing*And deciding rescue timeliness indexes for the distribution after the non-dimensionalization processing. C2Is a constant.
The target expectation of the fitness function is all the maximum value, and the value range of the fitness function of the genetic algorithm must be greater than 0, so C1＝C2＝1。
Adding constraint conditions, in practical application, there are often many different limitations to the configuration decision of the emergency resources, for example, all demand nodes in the campus are required to be in the emergency jurisdiction or constraints are added based on performance indexes of the configuration decision: the number of emergency nodes is limited, the coverage degree is limited, and the rescue timeliness is limited. The method used for processing the constraint conditions in the genetic algorithm optimization process is to eliminate individuals which do not meet the constraint conditions, i.e. to forcibly reset the fitness of the infeasible solution to 0.
Arranging individuals in the population from small to large according to the fitness, calculating the accumulated fitness:
Figure BDA0001525339350000121
wherein, Fitness (idv)i) And (4) the fitness obtained by the calculation is carried into the fitness function for the ith individual in the current population.
Sixthly, taking a random number r belonging to (0, FitnessT (pop _ size));
seventhly if r is the same as (FitnessT (i), FitnessT (i +1)]Then individual idviSelecting the selected plants;
repeating the step (c) until POP _ size-1 individuals are selected, and assigning the selected individuals to New POP; adopting an elite policy: reserving the last individual in the POP, emptying the rest individuals, and assigning all the individuals in the New POP to the POP
Generating a real number a randomly in an interval (0,1) according to a given Cross _ rate, if a < Cross _ rate, then crossing two adjacent individuals, otherwise, not crossing. For binary coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, m), and exchanging all binary strings after the hybridization position between two chromosomes, as shown in FIG. 2. For matrix coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, Q), and exchanging all row vectors below the hybridization position between the two chromosomes.
Ninthly, randomly generating a real number a in the interval (0,1) according to a given mutation probability Mutate _ rate, and if a is less than the Mutate _ rate, performing mutation operation on the individual. For binary coding, the mutation process is: randomly generating a positive integer b as a variation position between the intervals (0, m), and negating the binary number of the position. As shown in fig. 3. For matrix coding, the mutation process is: randomly generating a positive integer b as a variation position between the intervals (0, Q), setting all the b-th row elements of the individual matrix to be 0, randomly generating a positive integer d between the intervals (0, c), and taking the d-th row element of the b-th row of the individual matrix as 1.
And (c) taking iteration times G as an algorithm termination condition, returning to the third step when G is less than 200, outputting an individual corresponding to the optimal fitness when G is 200, and obtaining an optimal configuration decision according to a coding mode.

Claims (1)

1. A chemical industry park emergency resource allocation optimization method is characterized by specifically comprising the following steps:
step 1, acquiring dangerous source information and geographic information in a chemical industry park, process flow information of each chemical industry dangerous source engaged in production in the park, a place available for planning and the quantity of emergency materials to be distributed, wherein the dangerous source information comprises the type, state, stock and distribution condition of the dangerous source;
step 2: establishing a risk dynamic evaluation model of the risk source according to the information obtained in the step 1, wherein the considered risk source characteristic indexes are inherent risk coefficients of the risk source, domino effect risk coefficients and process flow risk coefficients;
inherent danger coefficient of danger source
In a risk source evaluation model based on an R value, a value R obtained by correcting and summing the ratios of the actual existing quantities of various risk sources in a unit and the critical quantities specified in a document GB18218 is used as a grading index; the calculation process of the R value is a coupling process of the easiness of accidents and the severity of the accident consequence, so the R value actually represents the inherent danger of the danger source; the calculation formula of the inherent risk coefficient of the hazard source is as follows:
Figure FDA0001525339340000011
in the formula, betanIs the correction coefficient of the hazard source n, alpha is the correction coefficient of the exposure personnel outside the plant, qnIs the actual inventory of hazard n, QnThe critical amount specified in document GB18218 for hazard source n;
risk coefficient of domino effect-
The probability of the domino effect is obtained by performing accident consequence simulation prediction on a hazard source; the prediction process follows a maximum risk principle and a probability summation principle; the domino effect risk coefficient γ is calculated as follows:
Figure FDA0001525339340000012
wherein i and k represent the number of plants in which a secondary accident occurs, PblastTo impact the probability of domino effect, PheatIs the thermal radiation domino effect probability;
risk coefficient of technological process
Calculating the risk coefficient of a process unit in a reference chemical fire and explosion index evaluation method; selecting the most dangerous state of the material in the process unit when calculating each coefficient in the process unit danger coefficients;
the calculation method of the process risk coefficient comprises the following steps:
F3＝F1×F2
in the formula, F1Is a general process flow risk factor, F2For a particular process risk factor, F3Is the risk coefficient of the process flow;
step 3, the hazard source can be characterized as follows through three characteristic indexes of the hazard source:
Xi＝(xi1,xi2,xi3)T,i＝1,2,...,n
in the formula, XiIs a hazard source i, xi1、xi2And xi3Three characteristic indexes respectively representing hazard sources i;
the method comprises the following steps of (1) grading a hazard source sample by using a pedigree clustering method:
selecting definitions of similarity measurement between samples and classes, and using distances between samples and between classes as the similarity measurement between samples and classes;
the inter-sample distance is defined as the euclidean distance:
Figure FDA0001525339340000021
wherein m is a study sampleNumber of characteristic indexes of (d)ijIs the distance between sample i and sample j;
the inter-class distance is defined as the shortest distance, and the calculation formula is as follows:
Figure FDA0001525339340000022
in the formula, GpAnd GqAre of the classes p and q, DpqIs the distance between class p and class q;
secondly, calculating the distance between the samples to generate a distance matrix D;
constructing each class which only contains one sample;
combining according to the definition of similarity measurement between classes to generate a new class;
calculating the distance between the new class and the existing class;
sixthly, if one category exists, drawing a pedigree clustering graph, otherwise, returning to the step four; the similarity relation between research samples is clearly expressed by a pedigree cluster map;
seventhly, determining the number of the classes and obtaining a grading result;
and 4, step 4: abstracting the whole chemical industry park into a network topological structure, wherein the location of a hazard source is a demand node in a network topological graph, the location for storing emergency facilities and emergency materials is a potential emergency node in the network topological graph, a road for connecting each node is an arc in the network topological graph, and the emergency range of the emergency node is an emergency jurisdiction; the expression of the chemical industry park network topological graph is as follows:
G＝{V,E}
in the formula, V is an arc set in the network, the weight of the arc is the arc length of the arc, E is a node set in the network, and the weight of a required node is the risk level of a hazard source;
and 5: dividing the emergency resource allocation decision into an emergency node site selection decision and an emergency material allocation decision, and establishing a multi-target emergency resource allocation model based on the risk level of the hazard source; setting n as the number of demand nodes, p as the number of potential emergency nodes, and Q as the number of emergency materials to be distributed;
safety of configuration decision
The configuration decision safety is measured by two aspects of the weighted coverage degree of the demand nodes in the network and the weighted distribution degree of the emergency jurisdiction; calculated by the following formula:
Figure FDA0001525339340000031
Figure FDA0001525339340000032
S＝S1+S2
in the formula, S1For addressing decision-making safety indicators, BiThe number of emergency nodes including a demand node i in an emergency jurisdiction; hiThe danger weight of the demand node i; s2To assign a decision safety index, IuThe selected emergency node set is selected; kjDetermining the risk weight of the emergency node j in the emergency jurisdiction by the average risk weight of the demand nodes in the emergency jurisdiction; qjAllocating the stock of emergency materials for the emergency node j; s is an emergency resource allocation decision overall safety index;
② economic applicability of configuration decision
The economic applicability of the configuration decision is measured by the number of the selected emergency nodes; calculated by the following formula:
Figure FDA0001525339340000033
in the formula, EA is an index of decision-making economic applicability; c, the number of the selected emergency points is counted;
thirdly, configuration decision rescue timeliness
The decision rescue timeliness is measured by emergency material storage in each emergency node and the average rescue distance, and the calculation method is as follows:
Figure FDA0001525339340000041
in the formula, T is the timeliness index of decision rescue, CjThe average rescue distance of the emergency node j is the average distance from the emergency node j to a demand node in the district;
step 6, optimizing the multi-target emergency resource configuration model by using a genetic algorithm optimization method, and finally solving a configuration strategy meeting the target expectation; the method comprises the following specific steps:
solving the problem of multi-target optimization emergency point addressing by adopting a classical binary coding mode, wherein each potential emergency point is endowed with an ID number by using continuous integers, and each emergency resource to be distributed and each emergency point are respectively endowed with an ID number by using the continuous integers;
initializing process, namely generating an initial population in the initialization process of the genetic algorithm; the population is composed of a certain number of individuals, so that pop _ size feasible solutions are randomly selected as the individuals to form an initial population pop;
carrying out dimensionless processing by using the extreme value of each index; and (3) constructing a genetic algorithm fitness function by combining the model expectation, wherein the fitness function of the site selection strategy is obtained by the following formula:
maximize Fx＝S1 *-EA*+C1
in the formula, FxDeciding a fitness function for site selection, S1 *Making a decision on a safety index, EA, for location after dimensionless processing*Deciding an economic applicability index for the site selection after the dimensionless processing; c1Is a constant;
the fitness function of the allocation policy is given by:
maximize Fy＝S2 *-T*+C2
in the formula, FyDeciding a fitness function for site selection, S2 *For the allocation of a decision-making safety index, T, after dimensionless processing*Deciding rescue timeliness indexes for distribution after non-dimensionalization processing; c2Is a constant;
the target expectation of the fitness function is all the maximum value, and the value range of the fitness function of the genetic algorithm must be greater than 0, so C1＝C2＝1；
Adding constraint conditions; the method for processing the constraint conditions in the genetic algorithm optimization process is to eliminate individuals which do not meet the constraint conditions, namely, the fitness of the infeasible solution is forcibly reset to 0;
arranging individuals in the population from small to large according to the fitness, calculating the accumulated fitness:
Figure FDA0001525339340000042
wherein, Fitness (idv)i) The fitness obtained by calculation is brought into a fitness function for the ith individual in the current population;
sixthly, taking a random number r belonging to (0, FitnessT (pop _ size));
seventhly if r is the same as (FitnessT (i), FitnessT (i +1)]Then individual idviSelecting the selected plants;
repeating the step (c) until POP _ size-1 individuals are selected, and assigning the selected individuals to New POP; adopting an elite policy: reserving the last individual in the POP, emptying the other individuals, and assigning all the individuals in the New POP to the POP;
generating a real number a randomly in an interval (0,1) according to a given Cross _ rate, if a < Cross _ rate, crossing two adjacent individuals, otherwise not crossing; for binary coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, m), and exchanging all binary strings after the hybridization position between two chromosomes is exchanged; for matrix coding, the hybridization process is: randomly generating a positive integer b as a hybridization position between intervals (0, Q), and exchanging all row vectors below the hybridization position between the two chromosomes;
ninthly, randomly generating a real number a in an interval (0,1) according to a given mutation probability Mutate _ rate, and if a is less than the Mutate _ rate, performing mutation operation on the individual; for binary coding, the mutation process is: randomly generating a positive integer b as a variation position between the intervals (0, m), and negating the binary number of the position; for matrix coding, the mutation process is: randomly generating a positive integer b as a variation position between intervals (0, Q), setting all the b-th row elements of the individual matrix to be 0, randomly generating a positive integer d between intervals (0, c), and taking the d-th row element of the b-th row of the individual matrix as 1;
and (c) taking iteration times G as an algorithm termination condition, returning to the step (6) when G is less than 200, outputting an individual corresponding to the optimal fitness when G is 200, and obtaining an optimal configuration decision according to a coding mode.
CN201711433425.1A 2017-12-26 2017-12-26 Chemical industry park emergency resource allocation optimization method Active CN108171413B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711433425.1A CN108171413B (en) 2017-12-26 2017-12-26 Chemical industry park emergency resource allocation optimization method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711433425.1A CN108171413B (en) 2017-12-26 2017-12-26 Chemical industry park emergency resource allocation optimization method

Publications (2)

Publication Number Publication Date
CN108171413A CN108171413A (en) 2018-06-15
CN108171413B true CN108171413B (en) 2021-08-10

Family

ID=62521196

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711433425.1A Active CN108171413B (en) 2017-12-26 2017-12-26 Chemical industry park emergency resource allocation optimization method

Country Status (1)

Country Link
CN (1) CN108171413B (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109086914B (en) * 2018-07-12 2022-03-25 杭州电子科技大学 Hazardous chemical substance vehicle path planning modeling method based on dynamic domino risk
CN112183808B (en) * 2019-07-02 2023-06-23 湖南大学 Optimization method and system for multi-automobile-product-combined research and development resource allocation
CN111125848A (en) * 2019-11-25 2020-05-08 李林卿 Dangerous goods transportation network emergency rescue resource allocation method
CN112926814A (en) * 2019-12-06 2021-06-08 中国石油大学(华东) Chemical industry park emergency resource allocation method based on intelligent management platform
CN111242454B (en) * 2020-01-07 2022-05-31 杭州电子科技大学 Chemical accident multi-target two-stage emergency rescue material scheduling method
CN113821989A (en) * 2020-06-18 2021-12-21 中国石油化工股份有限公司 Emergency disposal simulation risk equipment modeling method and device and storage medium
CN113393108A (en) * 2021-06-07 2021-09-14 中国石油大学(北京) Risk evaluation method for dangerous chemical transport vehicle gathering area
CN114048966B (en) * 2021-10-26 2024-06-18 中国石油大学(华东) Multi-level chemical industry park emergency resource region collaborative scheduling optimization method
CN114742327B (en) * 2022-06-10 2022-09-23 湖南前行科创有限公司 Rapid emergency disposal method and device for smart park based on collaborative optimization
CN116663854B (en) * 2023-07-24 2023-10-17 匠人智慧(江苏)科技有限公司 Resource scheduling management method, system and storage medium based on intelligent park
CN116934126B (en) * 2023-09-15 2023-11-24 匠人智慧(江苏)科技有限公司 Intelligent planning method, system and storage medium for intelligent park emergency treatment system
CN117669893A (en) * 2023-11-04 2024-03-08 兰州理工大学 Emergency response decision method and system based on fire domino risk analysis
CN118014322A (en) * 2024-04-09 2024-05-10 中国科学院自动化研究所 Material distribution method and device, electronic equipment and storage medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1608241A (en) * 2001-11-07 2005-04-20 维特赛半导体公司 System and method for communicating between a number of elements and a method for configuring and testing the system
CN101118611A (en) * 2007-09-07 2008-02-06 北京航空航天大学 Business process model resource configuring optimizing method based on inheritance algorithm
CN101257424A (en) * 2008-04-08 2008-09-03 杭州电子科技大学 Underwater wireless sensor network cover control method based on surface even allocation
CN102323906A (en) * 2011-09-08 2012-01-18 哈尔滨工程大学 MC/DC test data automatic generation method based on genetic algorithm
CN103177306A (en) * 2011-12-20 2013-06-26 中工国际工程股份有限公司 Workflow control system of project implementation
CN104463394A (en) * 2013-09-18 2015-03-25 Sap欧洲公司 Production resource management
KR20160141457A (en) * 2015-06-01 2016-12-09 주식회사 에스씨엘 Risk assessment system for information security management system
CN107480813A (en) * 2017-07-27 2017-12-15 河海大学 Basin water resources Optimal Configuration Method based on multi-Objective Chaotic genetic algorithm

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10068054B2 (en) * 2013-01-17 2018-09-04 Edico Genome, Corp. Bioinformatics systems, apparatuses, and methods executed on an integrated circuit processing platform
EA201401064A1 (en) * 2014-10-28 2016-04-29 Общество с ограниченной ответственностью "Синезис" METHOD (OPTIONS) SYSTEMATIZATION OF VIDEO DATA PRODUCTION PROCESS AND SYSTEM (OPTIONS)

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1608241A (en) * 2001-11-07 2005-04-20 维特赛半导体公司 System and method for communicating between a number of elements and a method for configuring and testing the system
CN101118611A (en) * 2007-09-07 2008-02-06 北京航空航天大学 Business process model resource configuring optimizing method based on inheritance algorithm
CN101257424A (en) * 2008-04-08 2008-09-03 杭州电子科技大学 Underwater wireless sensor network cover control method based on surface even allocation
CN102323906A (en) * 2011-09-08 2012-01-18 哈尔滨工程大学 MC/DC test data automatic generation method based on genetic algorithm
CN103177306A (en) * 2011-12-20 2013-06-26 中工国际工程股份有限公司 Workflow control system of project implementation
CN104463394A (en) * 2013-09-18 2015-03-25 Sap欧洲公司 Production resource management
KR20160141457A (en) * 2015-06-01 2016-12-09 주식회사 에스씨엘 Risk assessment system for information security management system
CN107480813A (en) * 2017-07-27 2017-12-15 河海大学 Basin water resources Optimal Configuration Method based on multi-Objective Chaotic genetic algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
最小化流水时间的工作流资源优化模型和算法;衣杨;《系统工程与电子技术》;20080731;第30卷(第7期);第1264-1269页 *

Also Published As

Publication number Publication date
CN108171413A (en) 2018-06-15

Similar Documents

Publication Publication Date Title
CN108171413B (en) Chemical industry park emergency resource allocation optimization method
Tian et al. Dual-objective scheduling of rescue vehicles to distinguish forest fires via differential evolution and particle swarm optimization combined algorithm
Li et al. Genetic search for solving construction site-level unequal-area facility layout problems
Cova et al. Modelling community evacuation vulnerability using GIS
Li et al. A new model for road network repair after natural disasters: Integrating logistics support scheduling with repair crew scheduling and routing activities
Kandil et al. Optimization research: Enhancing the robustness of large-scale multiobjective optimization in construction
CN103839118A (en) Site selection method and device
Golabi et al. Multiple-server facility location problem with stochastic demands along the network edges
Wang et al. Multiobjective optimization on hierarchical refugee evacuation and resource allocation for disaster management
Bahrami et al. A maximal covering facility location model for emergency services within an M (t)/M/m/m queuing system
Ji et al. A robust optimization approach for decontamination planning of emergency planning zone: Facility location and assignment plan
Bolouri et al. Environmental sustainable development optimizing the location of urban facilities using vector assignment ordered median problem-integrated GIS
Octarina et al. Models and heuristic algorithms for solving discrete location problems of temporary disposal places in Palembang City
Zhong et al. A two-stage hierarchical model for spatial location and evacuation allocation problem of urban earthquake shelters: a case study in Central urban area of Yangbi county, Yunnan province, China
CN107155215B (en) Distribution method and device of application home service cluster
A Konstantinidou et al. A multi-objective network design model for post-disaster transportation network management
Fedorchenko et al. Modified genetic algorithm to determine the location of the distribution power supply networks in the city
Mohammadi et al. MCLP and SQM models for the emergency vehicle districting and location problem
Rayeni et al. Optimization of steel moment frame by a proposed evolutionary algorithm
Eelagh et al. A location-allocation optimization model for post-earthquake emergency shelters using network-based multi-criteria decision-making
Azimi et al. Developing a new bi-objective functions model for a hierarchical location-allocation problem using the queuing theory and mathematical programming
Kordjazi et al. Presenting a three-objective model in location-allocation problems using combinational interval full-ranking and maximal covering with backup model
Zhou et al. A multi-agent genetic algorithm for multi-period emergency resource scheduling problems in uncertain traffic network
Bisso et al. Efficient determination of heliports in the city of Rio de Janeiro for the olympic games and world cup: a fuzzy logic approach
CN116957303B (en) Emergency response scheduling decision method and system for flood disaster scene

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant