Community fire-fighting toughness evaluation method and system based on association rule and complex network

Abstract

The invention discloses a community fire-fighting toughness evaluation method and system based on an association rule and a complex network, and the method comprises the steps: analyzing the disaster cause of an accident case based on a community fire accident case, and determining a community fire-fighting toughness evaluation index system from four dimensions: defending, rescuing, recovering and adapting; an Apriori algorithm is adopted to mine a coupling relationship between indexes, an association rule is generated, and a community fire-fighting toughness index complex network is constructed; key factors influencing the community fire-fighting toughness are determined by calculating the degree of a community fire-fighting toughness node, the intermediary centrality, the clustering coefficient and the like; and finally, making a fire-fighting toughness improvement scheme according to the key factors. According to the method, the invisible relation between the toughness indexes can be mined, the key factors influencing the community fire-fighting toughness can be determined, the community fire-fighting toughness improvement scheme can be formulated, the problem that community fire-fighting toughness evaluation mainly depends on subjective experience at present is effectively solved, and improvement of the overall fire-fighting toughness of the community is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of public safety technology, and in particular to a community fire resilience assessment method and system based on association rules and complex networks. Background Art

[0002] Community fire resilience refers to the ability of a community to respond quickly and effectively when disasters such as fire occur, and to recover quickly after the disaster, reduce disaster losses, and protect the lives and property of residents. This ability depends not only on perfect firefighting facilities and equipment, but also on efficient fire management, residents' self-rescue and mutual rescue capabilities, and the community's overall response mechanism. Community fire resilience assessment refers to the process of evaluating and measuring the comprehensive performance of the community in terms of residents' response capabilities, building fire protection, firefighting facilities, community environment, and management when disasters such as fire occur. This process aims to evaluate the community's capabilities and levels in fire prevention and control, emergency rescue, and post-disaster recovery through scientific methods and standards, so as to provide decision-making basis and improvement direction for improving community fire resilience.

[0003] There are two main types of community fire resilience assessments: one is based on the traditional indicator system, which only builds a community indicator system based on the building's own conditions, and rarely considers the core elements of the community's overall fire prevention and control from the perspective of resilience; the other is based on the consequences of an accident, which is mainly based on a certain scenario, but the factors considered are relatively single and cannot measure the fire resilience level of the community before the accident. Both resilience assessment methods have certain problems: the assessment based on the traditional indicator system is very subjective, making it difficult to effectively promote fire risk assessments in many areas; the assessment method based on the consequences of an accident is more suitable for emergency rescue and has certain limitations in fire prevention. Summary of the invention

[0004] The purpose of the present invention is to provide a community fire resilience assessment method and system based on association rules and complex networks. Based on the community fire accident case investigation report, the causes of the accident cases are analyzed, and a community fire resilience assessment index system is formulated from four aspects: personnel factors, building and equipment factors, environmental factors and management factors, including four dimensions of resistance, rescue, recovery and adaptation. The Apriori algorithm is used to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators. By calculating the degree, betweenness centrality, clustering coefficient, etc. of the community fire resilience nodes, the key factors affecting the community fire resilience are determined. Finally, a fire resilience improvement plan is formulated for the key factors to solve the problem that the community fire resilience factors involve many factors and the control direction is unclear.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In one aspect, the present invention provides a community fire resilience assessment method based on association rules and complex networks, the method comprising the following steps:

[0007] Step S1, based on community fire accident cases, analyzing the causes of the accident cases, and determining a community fire resilience assessment index system from four dimensions: resistance, rescue, recovery, and adaptation;

[0008] Furthermore, the cause of the accident case is to analyze the cause and consequences of the accident based on the community fire accident case, clarify the fire development process and accident disaster factors, etc.; the community fire resilience assessment index system includes four resilience factors: personnel factors, building and equipment factors, environmental factors, and management factors;

[0009] Step S2, using the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators;

[0010] Furthermore, according to the association rule algorithm, the support, confidence and gain between the indicators are calculated, and the frequent item sets and association rules between the indicators are screened out using the Apriori algorithm to form the nodes and edges of the complex network, thereby constructing the complex network of community fire resilience indicators;

[0011] Step S3, determining the key factors affecting community fire resilience by calculating the degree, betweenness centrality, clustering coefficient, etc. of the community fire resilience nodes;

[0012] Further, the complex network topology structure analysis is performed based on the nodes and edges of the complex network, and the key factors affecting the community fire resilience are determined based on the degree, clustering coefficient and betweenness centrality index of the complex network;

[0013] Furthermore, the community fire resilience assessment index system is determined from the four dimensions of resistance, rescue, recovery and adaptation, specifically including:

[0014] Resilience factors are determined from four aspects: personnel factors, building and equipment factors, environmental factors, and management factors. The personnel factors include permanent population density, personnel misoperation, personnel illegal operation, a large proportion of vulnerable population, and poor self-rescue and emergency response capabilities of residents; the building and equipment factors include building fire prevention factors and firefighting facilities and equipment factors; the environmental factors include physical environment factors and community space environment factors; the management factors include fire safety management work mechanism factors, fire brigade construction factors, failure to implement grid management, monitoring and early warning factors, hidden danger investigation and control factors, fire safety education and training factors, emergency plan system factors, and insufficient investment in fire safety funds;

[0015] Furthermore, the Apriori algorithm is used to mine the coupling relationship between indicators to obtain the frequent item sets and strong association rules, which specifically includes:

[0016] Based on the community fire resilience assessment index system and the resilience factors determined in step S1, the support, confidence, gain, etc. are calculated to clarify the strong correlation.

[0017] Based on the community fire resilience association rule, the support of the single item set is calculated; the support is an important indicator to measure the frequency of an item set in the data set. The higher the support of the community fire resilience item set, the more times the item set appears in the data set, and the more dangerous the item set is. The calculation method is as follows:

[0018]

[0019] Where Transactions containing X is the number of transactions containing item set X, and Totaltransactions is the total number of transactions.

[0020] Furthermore, based on the community fire resilience association rule, the confidence of the single item set is calculated; the confidence is an indicator to measure the reliability of an item set under given conditions. The higher the confidence of the community resilience item set X and item set Y, the higher the probability of item set Y appearing when item set X is observed in the data set. The calculation method is as follows:

[0021]

[0022] In the formula, Support(X∪Y) represents the ratio of transactions containing both X and Y to the total number of transactions, that is, the support of the itemset {X,Y}; Support(X) represents the ratio of transactions containing X to the total number of transactions, that is, the support of the itemset X.

[0023] Furthermore, based on the community fire resilience association rule, the gain of the single item set is calculated; the gain is an important indicator for measuring the strength and directionality of the association rule; under the condition of including the antecedent, the degree of increase in the probability of occurrence of the latter item relative to its unconditional probability of occurrence is greater than 1, then the association rule is called a strong association rule, and the calculation method is as follows:

[0024]

[0025] In the formula, Confidence(X→Y) represents the confidence of rule X→Y, that is, the conditional probability p(Y|X); Support(Y) represents the support of item set Y, that is, the unconditional probability p(Y); p(XY) represents the probability that item sets X and Y appear at the same time.

[0026] Furthermore, according to the support, confidence and gain of the association rules, the thresholds are continuously adjusted until the emerging rules can fully cover the relationship between the community fire resilience factors. The thresholds include: minimum condition support ≥ 10%, minimum rule support ≥ 50%, and minimum gain > 100%.

[0027] Furthermore, the process of determining the key factors of community fire resilience specifically includes:

[0028] According to the frequent item sets as nodes of a complex network;

[0029] According to the connection between the nodes of the complex network, the connection is an association rule, that is, an edge of the complex network;

[0030] Based on the fire resilience complex network, the degree of the resilience factor is determined, that is, the out-degree, in-degree and total degree of the resilience factor; the degree refers to the number of connections between a certain resilience factor and other resilience factors. The more connections a resilience factor has with other resilience factors, the more important the resilience factor is. The total degree is the sum of the out-degree and the in-degree.

[0031] Furthermore, based on the fire resilience complex network, the clustering coefficient of the resilience factors in community fire protection is calculated; the clustering coefficient indicates the degree to which a certain resilience factor is affected by other resilience factors. The larger the clustering coefficient of the resilience factors in community fire protection, the more critical the resilience factors are. The calculation method is as follows:

[0032]

[0033] Where si represents the degree of the node i, and Bi is the actual number of edges connected to the node i;

[0034] Based on the community fire complex network, the betweenness centrality of different resilience factors is calculated; the betweenness centrality is used to evaluate the importance of the community fire resilience factor, and the calculation method is as follows:

[0035]

[0036] In the formula, represents the number of shortest paths from c to d passing through the node i, Represents the number of shortest paths from c to d, where c and d are other nodes that are not equal to i.

[0037] Furthermore, based on the degree, clustering coefficient and betweenness centrality index of the complex network, the key factors involved in the community fire resilience are determined, and when the resilience factor meets a preset threshold, the resilience factor is determined to be a key factor; the thresholds include: the total degree of the complex network ≥ 2, and the betweenness centrality > 0.

[0038] Step S4, formulating a fire resilience improvement plan targeting the key factors based on the association rule algorithm and the complex network analysis results.

[0039] The community fire protection full chain resilience improvement plan is based on the key factors determined by the complex network, and is formulated from four aspects: personnel factors, building and equipment factors, environmental factors, and management factors, including the four dimensions of resistance, rescue, recovery, and adaptation;

[0040] The personnel factors mentioned above refer to the improvement of personnel factors in formulating community fire resilience plans for community residents in terms of personnel misoperation, personnel illegal operation behaviors, proportion of vulnerable population, and residents' self-rescue and emergency response capabilities;

[0041] The building and equipment factors refer to the improvement of building fire resistance performance and fire protection facilities and equipment from the aspects of fire protection distance, fire load, configuration and setting of fire protection facilities and equipment, laying of electrical lines, fire separation measures, evacuation passages and safety exits to determine the building and equipment factors of the community fire resilience plan;

[0042] The environmental factors mentioned above refer to the improvement of the community fire resilience plan environmental factors in terms of the number of beds in medical and health institutions, the illegal addition of facilities, and the installation of charging piles for the community space environment;

[0043] The management factors mentioned above refer to the improvement of community fire resilience plan management factors in terms of fire safety management work mechanism, hidden danger investigation and control, fire safety education and training, and emergency plan formulation for community management.

[0044] In another aspect, the present invention provides a community fire resilience assessment system based on association rules and complex networks, the system comprising:

[0045] The accident case analysis module analyzes the causes of the accident cases based on the community fire accident case investigation report, and determines the community fire resilience assessment indicator system from the four dimensions of resistance, rescue, recovery, and adaptation;

[0046] The association rule formation module uses the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators;

[0047] The complex network analysis module determines the key factors affecting community fire resilience by calculating the degree, betweenness centrality, clustering coefficient, etc. of community fire resilience nodes;

[0048] The fire resilience improvement plan preparation module formulates a fire resilience improvement plan based on the key factors according to the association rule algorithm and the complex network analysis results.

[0049] Based on the above technical solution, the embodiments of the present invention can at least produce the following technical effects:

[0050] The method of the present invention has the advantages of comprehensive analysis, systematic optimization, predictive management, decision support, dynamic and real-time performance, etc., which can effectively improve the accuracy of community fire resilience assessment. The association rule model can deeply explore the potential connections between various factors in the community fire system, thereby more comprehensively reflecting all aspects of community fire resilience. At the same time, the complex network model can consider the interaction and mutual influence between various factors in community fire resilience, identify the key nodes and weak links in the community fire system, and provide direction for improving community fire resilience. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0052] Figure 1 A flow chart of a community fire resilience assessment method based on association rules and complex networks according to an embodiment of the present invention;

[0053] Figure 2 is a schematic diagram of a complex network model according to an embodiment of the present invention;

[0054] Figure 3 is a node degree calculation value graph of an embodiment of the present invention;

[0055] Figure 4 is a graph of calculated values ​​of clustering coefficients of nodes in an embodiment of the present invention;

[0056] Figure 5 It is a schematic diagram of the structure of a community fire resilience assessment system based on association rules and complex networks according to an embodiment of the present invention;

[0057] Figure 6 This is a diagram of a community fire resilience improvement solution according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Example 1

[0060] like Figure 1 As shown, the community fire resilience assessment method based on association rules and complex networks in this embodiment includes the following steps:

[0061] Step S1, based on community fire accident cases, analyzing the causes of the accident cases, and determining a community fire resilience assessment index system from four dimensions: resistance, rescue, recovery, and adaptation;

[0062] The cause of the accident case is to analyze the cause and consequences of the accident based on the community fire accident case, clarify the fire development process and accident disaster factors, etc.; the community fire resilience assessment index system includes four resilience factors: personnel factors, building and equipment factors, environmental factors, and management factors;

[0063] Step S2, using the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators;

[0064] According to the association rule algorithm, the support, confidence, and gain between the indicators are calculated, and the frequent item sets and association rules between the indicators are screened out using the Apriori algorithm to form the nodes and edges of the complex network, thereby constructing the complex network of community fire resilience indicators;

[0065] Step S3, determining the key factors affecting community fire resilience by calculating the degree, betweenness centrality, clustering coefficient, etc. of the community fire resilience nodes;

[0066] According to the nodes and edges of the complex network, the complex network topology structure analysis is performed, and according to the degree, the clustering coefficient and the betweenness centrality index of the complex network, the key factors affecting the community fire resilience are determined;

[0067] It should be noted that in terms of case sources, a total of 150 community fire accident investigation reports from 2010 to 2024 were collected through official channels such as the Ministry of Emergency Management and provincial and municipal emergency management bureaus. In terms of analyzing the causes of the accident cases, based on the community fire accident cases, the causes and consequences of the accidents were analyzed, the fire development process and accident disaster factors were clarified, and the community fire resilience assessment index system was determined from the four dimensions of resistance, rescue, recovery, and adaptation, including four resilience factors: personnel factors, building and equipment factors, environmental factors, and management factors.

[0068] The personnel factors include permanent population density, personnel errors, personnel violations, a large proportion of vulnerable population, and poor self-rescue and emergency response capabilities of residents; the building and equipment factors include building fire prevention factors and fire-fighting facilities and equipment factors; the environmental factors include physical environment factors and community space environment factors; the management factors include fire safety management work mechanism factors, fire brigade construction factors, failure to implement grid management, monitoring and early warning factors, hidden danger investigation and control factors, fire safety education and training factors, emergency plan system factors, and insufficient investment in fire safety funds.

[0069] In establishing the community fire resilience data set, according to the community fire resilience evaluation index system, the accident causes and control factors in the community fire accident case investigation report are converted from text form to numerical form through 0-1 coding. If a certain accident cause or control factor appears in the text report, it is recorded as 1, otherwise it is recorded as 0, and the community fire resilience data set is established. In obtaining frequent item sets and strong association rules, the community fire resilience data set is input, and the items of all transactions that appear are counted. At the same time, the support, confidence, gain and other indicators of a single item set are calculated, and the calculation results are analyzed to obtain frequent item sets and strong association rules. In determining the key factors, according to the resilience factors, a complex network of community fire resilience is constructed, and the degree, clustering coefficient, betweenness centrality and other indicators of the complex network are calculated. The calculation results are analyzed to determine the key factors of community fire resilience.

[0070] The specific process of obtaining frequent item sets and strong association rules is as follows: first, by inputting the community fire resilience data set, the items of all transactions that appear are counted, and the support, confidence, gain and other indicators of a single item set are calculated; then, based on the community fire resilience association rule, a frequent 1-item set L1 is generated, and it is determined whether the support, confidence, and gain meet the minimum support, minimum rule confidence, and minimum gain preset by the algorithm; if so, the frequent 2-item set L2 is obtained from L1 according to the Apriori operation, and then L3 is found from L2, and finally, this is repeated until the algorithm ends, and frequent item sets and strong association rules are obtained.

[0071] Based on the community fire resilience association rule, the support of the single item set is calculated; the support is an important indicator to measure the frequency of an item set in the data set. The higher the support of the community fire resilience item set, the more times the item set appears in the data set, and the more dangerous the item set is. The calculation method is as follows:

[0072]

[0073] Where Transactions containing X is the number of transactions containing item set X, and Totaltransactions is the total number of transactions.

[0074] Based on the community fire resilience association rule, the confidence of the single item set is calculated; the confidence is an indicator to measure the reliability of an item set under given conditions. The higher the confidence of the community resilience item set X and item set Y, the higher the probability of item set Y appearing when item set X is observed in the data set. The calculation method is as follows:

[0075]

[0076] In the formula, Support(X∪Y) represents the ratio of transactions containing both X and Y to the total number of transactions, that is, the support of the itemset {X,Y}; Support(X) represents the ratio of transactions containing X to the total number of transactions, that is, the support of the itemset X.

[0077] Based on the community fire resilience dataset, the Apriori algorithm is used to obtain the association rules between resilience factors. After continuously adjusting the threshold, association rule mining is performed to obtain the association rules between resilience factors.

[0078] Table 1 shows the mining results of association rules with high support. It should be noted that in this embodiment, considering that many association rules meet the support threshold, only the top 5 association rules are displayed. As can be seen from Table 1, the poor self-rescue and emergency response capabilities of residents A5 and the large proportion of vulnerable population A4 and the large fire load B17 have a high-frequency cause co-occurrence relationship; the inadequate implementation of fire safety responsibilities D12 and the inadequate fire prevention and investigation have a high-frequency cause co-occurrence relationship with the large fire load B17; the failure to conduct emergency plan drills D72 on a regular basis has a high-frequency cause co-occurrence relationship with the poor self-rescue and emergency response capabilities of residents A5, the inadequate fire prevention and investigation D51 and the large proportion of vulnerable population A4. These factors that lead to the frequent occurrence of reduced community fire resilience levels put the community in a risky state and ultimately lead to fire accidents.

[0079] Table 1 Results of mining high-support association rules

[0080]

[0081] Furthermore, Table 2 shows the mining results of high-confidence association rules. In this embodiment, considering that many association rules meet the confidence threshold, only the top 5 association rules are displayed. As can be seen from Table 2, when there is a large proportion of vulnerable population A4, no regular fire safety training for residents, no fire publicity and education D62, no regular emergency plan drills D72, inadequate fire prevention inspections D51, and inadequate implementation of fire responsibilities D12, it is easy to lead to poor self-rescue and emergency response capabilities of residents A5. There is a high degree of correlation between the above factors. The occurrence of the former factor is likely to lead to the occurrence of the latter factor, thereby triggering a chain reaction.

[0082] Table 2 High confidence association rule mining results

[0083]

[0084] According to the support, confidence and gain of the association rules, the thresholds are continuously adjusted until the emerging rules can fully cover the relationship between the community fire resilience factors. The thresholds include: minimum condition support ≥ 10%, minimum rule support ≥ 50%, and minimum gain > 100%.

[0085] The specific process of constructing the complex network is as follows: first, based on the frequent item sets of community fire resilience factors and the experience of frontline experts on the scene, the resilience factors involved in community fire fighting are determined, and the nodes of the complex network are determined; then, based on the relationship between the resilience factors, the edges of the complex network are determined; finally, by calculating the degree, clustering coefficient and betweenness centrality of each node in the complex network, the key resilience factors are identified.

[0086] The process of determining the key resilience factors of community firefighting specifically includes: establishing nodes of the community firefighting resilience index complex network based on the community firefighting resilience evaluation index system and the resilience factors determined in step S1. According to the correlation between the resilience factors, determining the connection between the nodes of the community firefighting resilience complex network, i.e., the edge of the network.

[0087] like Figure 2 As shown, it is a schematic diagram of the complex network model of this embodiment. According to the complex network model diagram of the community fire resilience index, the relationship between the resilience factors is input into the Gephi software to establish the complex network model of the community fire resilience index.

[0088] It should be noted that, in this embodiment, the fire resilience of the community involves 42 resilience factors, among which the personnel factors include the density of permanent population A1, personnel misoperation A2, personnel illegal operation A3, large proportion of vulnerable population A4, and poor self-rescue and emergency response capabilities of residents A5; the building and equipment factors include the age of the house B11, unreasonable fire separation distance B12, no fire separation measures, or the measures fail B13, low fire resistance of building materials B14, irregular laying of electrical lines B15, internal equipment does not meet the requirements B16, a lot of piled up debris, large fire load B17, occupation / blocking / closure of evacuation passages and safety exits B18, emergency evacuation lighting and evacuation signs cannot be used normally B19, the configuration and setting of fire-fighting facilities and equipment do not meet the standards B21, failure of fire-fighting facilities and equipment B22, and occupation of fire truck passages B23; the environmental factors include high temperature and dry environment C11, far from the fire station C13, number of beds in medical and health institutions C14, road network accessibility C15, illegal installation of facilities C21, charging piles not installed or installed in an irregular manner C22; the management factors include imperfect fire safety system D11, inadequate implementation of fire safety responsibilities D12, irregular management of fire and electricity use D13, imperfect post-disaster resident resettlement system D14, imperfect post-disaster equipment maintenance and replenishment system D15, failure to establish micro fire stations D21, insufficient number of firefighters D22, unqualified firefighters D23, failure to implement grid management D3, imperfect emergency warning and response mechanisms D41, failure to establish a smart fire warning information system D42, inadequate fire prevention inspections D51, inadequate hidden danger rectification D52, inadequate maintenance and management of firefighting facilities and equipment D53, failure to regularly organize employees to carry out fire safety training D61, failure to regularly carry out fire safety training for residents, failure to carry out fire publicity and education D62, no emergency plan, or the emergency plan is not updated in a timely manner D71, failure to regularly carry out emergency plan drills D72, and insufficient investment in fire safety funds D8.

[0089] Based on the complex network of community fire resilience indicators, the degree of resilience factors, i.e., the out-degree, in-degree, and total degree of the elements, is determined; the degree refers to the number of connections between a resilience factor and other resilience factors, and the total degree is the sum of the out-degree and in-degree. The calculation results are as follows: Figure 3 The figure shows the node degree calculation value diagram of this embodiment. According to the complex network model of community fire resilience index, the degree, clustering coefficient and betweenness centrality of the complex network are calculated, and the Origin software is used to draw a graph to obtain the calculated value diagram of the degree and clustering coefficient of the complex network of community fire resilience index.

[0090] Based on the complex network of community fire resilience indicators, the clustering coefficient O of resilience factors was calculated. i The clustering coefficient indicates the degree to which a certain resilience factor is affected by other resilience factors, and is calculated as follows:

[0091]

[0092] In the formula, si represents the degree of the node i, and Bi is the actual number of edges connected to the node i. The calculation result is as follows: Figure 4 shown.

[0093] Based on the complex network of community fire resilience indicators, the betweenness centrality of different resilience factors is calculated; the betweenness centrality is mainly used to evaluate the importance of the resilience factors, and the calculation method is as follows:

[0094]

[0095] In the formula, represents the number of shortest paths from c to d passing through the node i, Represents the number of shortest paths from c to d, where c and d are other nodes that are not equal to i.

[0096] Among them, the degree of some nodes is 1, and the value of the clustering coefficient is set to 999999998. The node is deleted when the value of the clustering coefficient is plotted.

[0097] According to the calculation of the degree, clustering coefficient and betweenness centrality of the complex network of community fire resilience indicators, it can be seen that: the total degree of establishing micro fire stations D21, implementing grid management D3 and stacking a lot of debris and large fire load B17 is relatively high, indicating that when strengthening the level of community fire resilience, it is necessary to focus on management factors and building fire prevention factors; there is no emergency plan, or the emergency plan is not updated in time D71, the number of firefighters is insufficient D22, and the configuration and setting of firefighting facilities and equipment do not meet the standards B21. The clustering coefficient value is high, indicating that these resilience factors are crucial to the occurrence of community fire accidents; the fire safety responsibility is not implemented in place D12, the maintenance and management of firefighting facilities and equipment are not in place D53, and the charging pile is not installed or installed in an irregular manner C22. The value of betweenness centrality is high, indicating that it is necessary to focus on community management and environmental conditions in community fire resilience.

[0098] According to the degree, average shortest distance, clustering coefficient and betweenness centrality index of the complex network, the key resilience factors involved in the community fire resilience index are determined. When the resilience factor meets the preset threshold, the resilience factor is determined to be the key resilience factor; the threshold includes: the total degree of the complex network ≥2, and the betweenness centrality>0.

[0099] It should be noted that, in this embodiment, a total of 24 key resilience factors are screened out according to the set thresholds, among which the personnel resilience factors include the permanent population density A1, illegal operation of personnel A3, large proportion of vulnerable population A4, and poor self-rescue and emergency response capabilities of residents A5; the building and equipment resilience factors include no fire separation measures or failure of measures B13, irregular laying of electrical lines B15, large amount of piled up debris and large fire load B17, occupied / blocked / closed evacuation passages and emergency exits B18, and the configuration and setting of fire-fighting facilities and equipment do not meet the standards B21; the environmental resilience factors include the number of beds in medical and health institutions C14, and no charging piles installed. Or the installation is not standardized C22; the management resilience factors include imperfect fire safety system D11, failure to fully implement fire safety responsibilities D12, irregular management of fire and electricity use D13, failure to establish micro fire stations D21, insufficient number of firefighters D22, failure to implement grid management D3, imperfect emergency warning and response mechanisms D41, inadequate fire prevention inspections D51, inadequate rectification of hidden dangers D52, inadequate maintenance and management of fire-fighting facilities and equipment D53, failure to regularly organize fire safety training for employees D61, failure to regularly conduct fire safety training for residents, failure to conduct fire publicity and education D62, and no emergency plan or failure to update the emergency plan in a timely manner D71.

[0100] like Figure 1 As shown, determination of the resilience assessment method also includes step S4, formulating a fire resilience improvement plan for the key factors based on the association rule algorithm and complex network analysis results.

[0101] The community fire protection full chain resilience improvement plan is based on the key factors determined by the complex network, and is formulated from four aspects: personnel factors, building and equipment factors, environmental factors, and management factors, including the four dimensions of resistance, rescue, recovery, and adaptation;

[0102] In terms of resisting community fires, we can improve our ability to resist fires by strengthening the construction and maintenance of fire-fighting facilities, improving the intelligence level of fire-fighting facilities, and strengthening fire safety education for community residents.

[0103] In terms of rescuing community fires, it is necessary to establish an efficient emergency command system and a rapid response mechanism. This includes regular emergency plan drills and the establishment of a professional fire rescue team. In addition, through multi-department collaboration and joint law enforcement, fire safety management can be strengthened to ensure that rescue operations can be carried out quickly and effectively when a fire occurs.

[0104] In terms of restoring community firefighting capabilities, it involves post-disaster reconstruction and rapid repair of infrastructure. Formulating a detailed post-disaster reconstruction plan, including improving the post-disaster resident resettlement system and the post-disaster equipment maintenance and replenishment system, will help the community quickly resume normal operations after a fire.

[0105] In the dimension of adaptive community firefighting capacity, it refers to the community's long-term ability to adapt and improve when facing disasters such as fires. This includes learning and summarizing lessons learned from fire accidents, updating fire safety management systems and technologies, and improving community firefighting resilience through innovation and learning indicators.

[0106] Example 2

[0107] This embodiment provides a community fire resilience assessment system based on the association rules and complex networks of Embodiment 1, such as Figure 5 FIG. 1 is a schematic diagram of the composition of the system disclosed in this embodiment, and the system includes:

[0108] The accident case analysis module analyzes the causes of the accident cases based on the community fire accident case investigation report, and determines the community fire resilience assessment indicator system from the four dimensions of resistance, rescue, recovery, and adaptation;

[0109] The association rule formation module uses the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators;

[0110] The complex network analysis module determines the key factors affecting community fire resilience by calculating the degree, betweenness centrality, clustering coefficient, etc. of community fire resilience nodes;

[0111] The fire resilience improvement plan preparation module formulates a fire resilience improvement plan based on the key factors according to the association rule algorithm and the complex network analysis results.

[0112] It should be noted that, based on the association rule algorithm and the complex network analysis results, a community fire resilience improvement plan is formulated, including four aspects: personnel resilience, building and equipment resilience, environmental resilience, and management resilience; the output results are as follows Figure 6 shown.

[0113] In response to resilience factors such as personnel illegal operations and poor self-rescue and emergency response capabilities of residents in the key factor analysis of complex networks, the personnel resilience aspect of the community fire resilience improvement plan stipulates: regular fire safety education and training for community residents and staff, strengthening safety awareness, clarifying operating procedures and safety standards; establishing a strict supervision mechanism, promptly correcting illegal operations, and holding them accountable to serve as a warning; organizing regular fire and emergency evacuation drills to improve residents' self-rescue and mutual rescue capabilities in real situations.

[0114] In response to the resilience factors such as the absence of fire separation measures or ineffective measures, improper installation of electrical lines, large amounts of debris, heavy fire loads, occupied / blocked / closed evacuation passages, emergency exits, and non-compliant configuration and settings of fire-fighting facilities and equipment in the key factor analysis of complex networks, comprehensive requirements have been formulated in the Community Fire Resilience Improvement Plan, which stipulates: ensuring that there are sufficient fire separation measures inside the building and checking that the fire separations are intact; regularly checking electrical lines to ensure that they are installed in accordance with regulations; regularly cleaning up debris in the building; ensuring that all evacuation passages and emergency exits meet standards; and configuring a sufficient number of fire-fighting facilities, and conducting regular inspections and maintenance.

[0115] In view of the two resilience factors of the number of beds in medical and health institutions and the lack of installation or improper installation of charging piles in the key factor analysis of complex networks, the community fire resilience improvement plan stipulates: design convertible wards and treatment areas to quickly increase the number of beds when needed, and improve the flexibility and efficiency of space use; when constructing charging piles, strictly abide by the latest building and electrical safety standards to ensure the safety and reliability of the charging piles; conduct safety assessments on existing illegally added facilities, and renovate facilities that do not meet safety regulations to ensure that all facilities meet building and fire safety standards.

[0116] In response to the resilience factors found in the analysis of key factors of complex networks, such as imperfect fire safety system, inadequate implementation of fire safety responsibilities, irregular management of fire and electricity use, failure to establish micro fire stations, insufficient number of firefighters, failure to implement grid management, inadequate emergency warning and response mechanisms, inadequate fire prevention inspections, inadequate rectification of hidden dangers, inadequate maintenance and management of fire-fighting facilities and equipment, failure to regularly organize fire safety training for employees, failure to regularly conduct fire safety training for residents, failure to conduct fire publicity and education, absence of emergency plans, or failure to update emergency plans in a timely manner, the Community Fire Resilience Improvement Plan stipulates: establishing and improving the responsibility system, optimizing the operating mechanism, improving the comprehensive emergency rescue capability, building a variety of fire organization systems, establishing and improving the fire safety grid management mechanism, promoting the construction of smart fire protection, establishing a unified fire emergency command and dispatch system, and strengthening the system and financial support.

[0117] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A community fire resilience assessment method based on association rules and complex networks, characterized by: The method comprises the following steps: Step S1, based on community fire accident cases, analyzing the causes of the accident cases, and determining a community fire resilience assessment index system from four dimensions: resistance, rescue, recovery, and adaptation; The cause of the accident case is to analyze the cause and consequences of the accident based on the community fire accident case, clarify the fire development process and accident disaster factors; the community fire resilience assessment index system includes four resilience factors: personnel factors, building and equipment factors, environmental factors, and management factors; Step S2, using the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators; According to the association rule algorithm, the support, confidence, and gain between the indicators are calculated, and the frequent item sets and association rules between the indicators are screened out using the Apriori algorithm to form the nodes and edges of the complex network, thereby constructing the complex network of community fire resilience indicators; Step S3, determining the key factors affecting community fire resilience by calculating the degree, betweenness centrality, and clustering coefficient of the community fire resilience nodes; According to the nodes and edges of the complex network, the topological structure analysis of the complex network is performed, and according to the degree, clustering coefficient and betweenness centrality index of the complex network, the key factors affecting the community fire resilience are determined; Step S4, formulating a fire resilience improvement plan targeting the key factors based on the association rule algorithm and the complex network analysis results.

2. The community fire resilience assessment method based on association rules and complex networks according to claim 1 is characterized in that: In step S1, the community fire resilience assessment index system is determined from the four dimensions of resistance, rescue, recovery, and adaptation, specifically including: Resilience factors are determined from four aspects: personnel factors, building and equipment factors, environmental factors, and management factors. The personnel factors include permanent population density, personnel errors, personnel violations, a large proportion of vulnerable population, and poor residents' self-rescue and emergency response capabilities; the building and equipment factors include building fire prevention factors and fire-fighting facilities and equipment factors; the environmental factors include physical environment factors and community space environment factors; the management factors include fire safety management work mechanism factors, fire brigade construction factors, failure to implement grid management, monitoring and early warning factors, hidden danger investigation and control factors, fire safety education and training factors, emergency plan system factors, and insufficient investment in fire safety funds.

3. The community fire resilience assessment method based on association rules and complex networks according to claim 1 is characterized in that: In step S2, the Apriori algorithm is used to mine the coupling relationship between indicators to obtain the frequent item sets and strong association rules, which specifically includes: Based on the community fire resilience assessment index system and the resilience factors determined in step S1, the support, confidence and gain are calculated to clarify the strong correlation relationship; Based on the community fire resilience association rule, the support of a single item set is calculated; the support is an important indicator to measure the frequency of an item set in a data set. The higher the support of the community fire resilience item set, the more times the item set appears in the data set, and the more dangerous the item set is. The calculation method is as follows: Where Transactions containing X is the number of transactions containing item set X, and Total transactions is the total number of transactions; Based on the community fire resilience association rule, the confidence of the single item set is calculated; the confidence is an indicator to measure the reliability of an item set under given conditions; the higher the confidence of the community resilience item set X and item set Y, the higher the possibility of item set Y appearing when item set X is observed in the data set, and the calculation method is as follows: In the formula, Support(X∪Y) represents the ratio of transactions containing both X and Y to the total number of transactions, that is, the support of the itemset {X,Y}; Support(X) represents the ratio of transactions containing X to the total number of transactions, that is, the support of the itemset X; Based on the community fire resilience association rule, the gain of the single item set is calculated; the gain is an important indicator for measuring the strength and directionality of the association rule; under the condition of including the antecedent, if the probability of occurrence of the latter item is greater than 1 relative to its unconditional probability of occurrence, then the association rule is called a strong association rule, and the calculation method is as follows: In the formula, Confidence(X→Y) represents the confidence of rule X→Y, that is, the conditional probability p(Y|X); Support(Y) represents the support of item set Y, that is, the unconditional probability p(Y); p(XY) represents the probability that item sets X and Y appear at the same time.

4. The community fire resilience assessment method based on association rules and complex networks according to claim 3 is characterized in that: According to the support, confidence and gain of the association rules, the thresholds are continuously adjusted until the emerging rules can fully cover the relationships between the community fire resilience factors. The thresholds include: minimum condition support ≥ 10%, minimum rule support ≥ 50%, and minimum gain > 100%.

5. The community fire resilience assessment method based on association rules and complex networks according to claim 4 is characterized in that: According to the degree, clustering coefficient and betweenness centrality index of the complex network, the key factors involved in the community fire resilience are determined. When the resilience factor meets the preset threshold, the resilience factor is determined to be a key factor; the threshold includes: the total degree of the complex network ≥ 2, and the betweenness centrality > 0.

6. The community fire resilience assessment method based on association rules and complex networks according to claim 1 is characterized in that: The process of determining the key factors of community fire resilience specifically includes: According to the frequent item sets as nodes of a complex network; According to the connection between the nodes of the complex network, the connection is an association rule, that is, an edge of the complex network; Based on the fire resilience complex network, determine the degree of the resilience factor, that is, the out-degree, in-degree and total degree of the resilience factor; the degree refers to the number of connections between a certain resilience factor and other resilience factors. The more connections a resilience factor has with other resilience factors, the more important the resilience factor is. The total degree is the sum of the out-degree and the in-degree. Based on the fire resilience complex network, the clustering coefficient of the resilience factors of community fire protection is calculated; the clustering coefficient indicates the degree to which a certain resilience factor is affected by other resilience factors. The larger the clustering coefficient of the resilience factors in the community fire protection, the more critical the resilience factors are. The calculation method is as follows: Where si represents the degree of the node i, and Bi is the actual number of edges connected to the node i; Based on the community fire complex network, the betweenness centrality of different resilience factors is calculated; the betweenness centrality is used to evaluate the importance of the community fire resilience factor, and the calculation method is as follows: In the formula, represents the number of shortest paths from c to d passing through the node i, Represents the number of shortest paths from c to d, where c and d are other nodes that are not equal to i.

7. The community fire resilience assessment method based on association rules and complex networks according to any one of claims 1 to 6, characterized in that: The community fire protection full chain resilience improvement plan is based on the key factors determined by the complex network, and is formulated from four aspects: personnel factors, building and equipment factors, environmental factors, and management factors, including the four dimensions of resistance, rescue, recovery, and adaptation; The human factor refers to the improvement of human resilience in community fire resilience from three aspects: safety education, emergency drills and supervision and regulation for community residents; The building and equipment factors refer to the improvement of building and equipment resilience in community fire resilience from three aspects: building structure layout, combustible management, and fire facilities and equipment, with regard to building fire resistance performance and firefighting facilities and equipment; The environmental factors refer to the improvement of environmental resilience in community fire resilience from both public and special areas in the community space environment; The management factors refer to the improvement of management resilience in community fire resilience from five aspects: system and responsibility system, facilities and personnel allocation, management and maintenance, early warning response mechanism and emergency plan.

8. A community fire resilience assessment system based on association rules and complex networks, used to implement the assessment method according to any one of claims 1 to 7, characterized in that: The system comprises: The accident case analysis module analyzes the causes of the accident cases based on the community fire accident case investigation report, and determines the community fire resilience assessment indicator system from the four dimensions of resistance, rescue, recovery, and adaptation; The association rule formation module uses the Apriori algorithm to mine the coupling relationship between indicators, generate association rules, and construct a complex network of community fire resilience indicators; The complex network analysis module determines the key factors affecting community fire resilience by calculating the degree, betweenness centrality, clustering coefficient, etc. of community fire resilience nodes; The fire resilience improvement plan preparation module formulates a fire resilience improvement plan based on the key factors according to the association rule algorithm and the complex network analysis results.

Images