CN111784060A - Urban community social security chain coupling risk evolution scenario conjecture method - Google Patents

Urban community social security chain coupling risk evolution scenario conjecture method Download PDF

Info

Publication number
CN111784060A
CN111784060A CN202010644533.9A CN202010644533A CN111784060A CN 111784060 A CN111784060 A CN 111784060A CN 202010644533 A CN202010644533 A CN 202010644533A CN 111784060 A CN111784060 A CN 111784060A
Authority
CN
China
Prior art keywords
evolution
risk
scene
scenario
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.)
Granted
Application number
CN202010644533.9A
Other languages
Chinese (zh)
Other versions
CN111784060B (en
Inventor
胡啸峰
吴建松
李瑞雪
白一平
韩昕格
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.)
PEOPLE'S PUBLIC SECURITY UNIVERSITY OF CHINA
China University of Mining and Technology Beijing CUMTB
Original Assignee
PEOPLE'S PUBLIC SECURITY UNIVERSITY OF CHINA
China University of Mining and Technology Beijing CUMTB
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 PEOPLE'S PUBLIC SECURITY UNIVERSITY OF CHINA, China University of Mining and Technology Beijing CUMTB filed Critical PEOPLE'S PUBLIC SECURITY UNIVERSITY OF CHINA
Priority to CN202010644533.9A priority Critical patent/CN111784060B/en
Publication of CN111784060A publication Critical patent/CN111784060A/en
Application granted granted Critical
Publication of CN111784060B publication Critical patent/CN111784060B/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
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
    • 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
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/22—Social work
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G06Q50/265—Personal security, identity or safety

Abstract

The invention discloses a method for conjecturing the social security chain type coupling risk evolution situation of an urban community, belonging to the technical field of application information. The method comprises the steps of firstly, dynamically identifying various risk accidents of the social security events and SCJ model scene primitives of the elements of the occurrence and development of risk scene states in various stages through a risk scene identification model, and determining a worst accident scene of event evolution and scene states in the worst accident scene. And applying the worst accident scene and the scene state to a risk scene evolution model, and obtaining each social security event risk evolution scene library based on the scene conjecture roadmap through node definition and evolution rules of the evolution model. And finally, defining a scene conjecture road map risk expression mode based on a GERT network inference function as a principle, analyzing single evaluation units in the chain evolution scene of the social security coupling risk one by one, and determining the risk state value of each event evolution path, thereby providing a basis for emergency processing and decision making of the urban community social security events.

Description

Urban community social security chain coupling risk evolution scenario conjecture method
Technical Field
The invention particularly relates to a method for conjecturing the social security chain type coupling risk evolution situation of an urban community, belonging to the technical field of application information.
Background
With the rapid development of economic and scientific areas in China, the acceleration of urbanization process makes urban social security face a serious challenge. Along with the emergencies of various social safety events, huge economic loss and severe social influence are brought. The risk of urban social security is mainly reflected in public order of society and public security of people, such as public security, criminal affairs, terrorism and large-scale group affairs. By adopting a 'situation-response' emergency decision paradigm, the risk evolution process of the social security of the urban community can be represented, the existing information, knowledge and data are brought into the specific situation evolution of the emergency, and the possible situation and the dangerous consequences thereof are judged. Therefore, a scene library of the community social security chain type coupling risk evolution is established in a 'scene-response' mode, and the risk of the event chain is presumed, so that powerful support is provided for strengthening the urban public security management efficiency.
The conventional machine learning method, such as chinese invention patent CN2019108584189, discloses a training method, a prediction method and a device of a risk prediction model, and this conventional risk prediction and management method based on machine learning cannot be directly applied to urban community social security risk prediction and management, and because of the particularity of urban community social security risk, a new scheme must be developed to solve this problem.
Disclosure of Invention
Therefore, the invention aims to provide a method for conjecturing the social security chain type coupling risk evolution situation of the urban community, which is used for identifying and conjecturing the risk situation evolution process of each social security event in the urban community, calculating the risk degree of each evolution scene and providing a basis for emergency treatment and decision of the social security events of the urban community.
Specifically, the method for conjecturing the urban community social security chain type coupling risk evolution scene provided by the invention comprises the following steps:
s1, establishing a city community social security event risk situation recognition model, and recognizing the risk situation and the evolution scene of each social security event;
s2, establishing an urban community social security event risk situation evolution model according to the risk situation recognition model to form a social security event chain risk evolution situation library;
s3, evaluating and analyzing the risk state of the evolution of each social security event of the urban community according to the risk situation evolution model;
in step S2, among others:
SCJ model scene elements are minimum units of evolution, a risk evolution process is divided into state evolution among single scene elements, and for each element promoting the risk evolution to develop, the SCJ model scene elements are defined by three attributes, namely scene states, and are expressed by a disaster-bearing body and the corresponding states thereof; disaster-causing elements consisting of disaster-causing factors and a pregnant disaster environment; scene judgment, which consists of scene response and scene conditions;
the scene response is an emergency response or decision aiming at the scene, and the scene condition is an external environment condition which is in the scene and can influence the evolution; for SCJ model scene elements, each scene state has disaster-causing factors, a pregnant disaster environment and scene responses or scene conditions for determining the evolution direction of the scene state; if the scene judgment of the previous scene element is matched with the disaster-causing element of the next scene element, the two scene elements have a sequential evolution transfer relationship;
the social security event risk scenario evolution model is a risk evolution scenario conjecture diagram which is made on the basis of a worst risk scenario evolution path and an evolution scenario state by defining nodes and scenario evolution rules among the nodes, and a social security chain type coupling risk evolution scenario library based on the scenario conjecture diagram is established; evolution is the description of the situation transition direction of the situation of the occurrence and development of an emergency, wherein the state refers to the current state of the event, and the state refers to the future development trend of the event; the risk evolution process of the event can be described by the evolution process of the scenario, and the evolution process of the scenario is the state change process of the scenario; for the risk scenario conjecture road map, the risk scenario conjecture road map consists of nodes and directed branches, wherein the nodes represent 'states', and the branches represent 'potentials' of 'state' transfer;
defining nodes refers to that nodes of the risk scenario conjecture road map are divided into start nodes, state nodes and end nodes, and scenario evolution rules of the nodes are transfer rules among the state nodes;
forming a risk evolution scenario conjecture road map from left to right according to the sequence of event occurrence and development based on node definition and evolution transition rules, and forming a risk evolution scenario conjecture road map triggered by various social security events under the trend judgment of scenario response and scenario conditions through each state node; therefore, the evolving worst path formed by the first-step risk scenario recognition model appears at the lowest edge of the scenario conjecture diagram, namely, the path formed by each node under the worst trend of scenario judgment.
The invention has the beneficial effects that:
the invention provides a method for conjecturing a social security chain type coupling risk evolution situation of an urban community. And then applying the worst accident situation and SCJ model situation primitives to a risk situation evolution model to obtain a situation library of risk evolution of each social security event based on the risk evolution situation conjecture roadmap. And finally, evaluating and analyzing each evaluation unit of the scenario conjecture road map through risk expression based on a GERT network deduction transmission principle, and determining the risk state of each event evolution path, thereby providing support for emergency decision of the social security events under a 'scenario-response' paradigm.
Drawings
FIG. 1 is a flowchart of a method for realizing chain-type coupling risk evolution scenario inference of social security of urban communities according to the present invention.
FIG. 2 is a flowchart illustrating steps of a social security event risk context identification model according to the present invention.
FIG. 3 is a diagram of SCJ model scene primitives according to the present invention.
FIG. 4 is a content composition diagram of the social security event risk scenario evolution model of the present invention.
FIG. 5 is a schematic diagram of a social security event scenario inference path according to the present invention.
Fig. 5a, 5b, 5c are exploded views of fig. 5.
Fig. 6 is a road view of the criminal event fire scenario conjecture of the present invention.
Fig. 6a, 6b and 6c are exploded views of fig. 6.
Fig. 7 is a road view of the criminal event explosion scenario conjecture of the present invention.
Fig. 7a, 7b, 7c, and 7d are exploded views of fig. 7.
Fig. 8 is a road view of the criminal event mechanical injury case scenario inference of the present invention.
Fig. 8a, 8b and 8c are exploded views of fig. 8.
Fig. 9 is a road diagram for estimating the criminal incident poison case scenario of the present invention.
FIG. 10 is a road diagram illustrating a group event scenario inference method according to the present invention.
Fig. 10a, 10b and 10c are exploded views of fig. 10.
FIG. 11 is a schematic diagram of a terrorist attack scenario according to the present invention.
Fig. 11a, 11b, 11c, and 11d are exploded views of fig. 11.
FIG. 12 is a functional deduction schematic diagram of the risk evolution scenario inference road diagram of the present invention.
Fig. 13 is a schematic diagram of feature parameter determination based on an evaluation unit in the scenario inference path of the present invention.
FIG. 14 is a simplified computational diagram of one embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided with reference to the accompanying drawings:
example 1
As shown in FIG. 1, the method for urban community social security chain coupling risk evolution scenario conjecture is composed of a risk scenario recognition model, a risk scenario evolution model and a risk evolution evaluation analysis. Under the dimension of urban communities, social security events are considered as major criminal cases, group events and terrorist attack events, and explosion cases, fire cases, mechanical injury cases and poison cases are considered in the major criminal cases. Therefore, the risk scene identification model of the first part obtains the worst accident scene path and the scene state of risk evolution by identifying the accidents of various events and the SCJ model scene primitives of the accident scene occurrence and development elements. And in the risk scenario evolution model of the second part, a risk scenario conjecture road map consisting of the three social security events is formed by defining the meaning and the evolution rule of the risk evolution node. And the risk evolution evaluation analysis of the third part is to realize the risk state calculation of each event scenario conjecture road map by defining a risk expression mode based on a self transfer function of the GERT network and analyzing each evaluation unit one by one.
FIG. 2 is a flowchart illustrating steps of a social security event risk context identification model according to the present invention. And finally, determining the worst risk evolution scene and each scene state in the worst risk evolution scene by comparing the disaster consequences of each risk scene.
Referring to fig. 3, the SCJ (State-Cause-projection) model scene primitive, i.e. the minimum unit of evolution, is composed of State-scene State, Cause-disaster element, and projection-scene judgment. The model divides the risk evolution process into state evolutions among single scene elements. For each element promoting risk evolution to occur and develop, SCJ model scene primitives are defined by three attributes, and a scene State State is expressed by a disaster-bearing body and a corresponding State thereof; disaster causing element Cause, which consists of disaster causing factors and a pregnant disaster environment; the context judgment Judge consists of a context response and a context condition, wherein the context response is an emergency response or decision aiming at the context, and the context condition is an external environment condition which is in the context and can influence the evolution. For SCJ model scene primitives, each scene state has disaster-causing factors, a pregnant disaster environment and a scene response or scene condition for determining the evolution direction of the scene state. If the scene judgment of the previous scene element is matched with the disaster-causing element of the next scene element, the two scene elements have a sequential evolution transfer relationship, so that the recognition of the SCJ scene element is beneficial to the analysis and recognition of the evolution path and the development trend of the scene.
The disaster-bearing body consists of people, a lifeline system, a building, a social environment, an economic environment and a natural environment.
The life line system consists of life lines of electric power, communication, network, gas, water supply and drainage, heat and the like, a traffic system and a fire-fighting system.
FIG. 4 is a content composition diagram of the social security event risk scenario evolution model of the present invention. The evolution model is used for making a risk evolution situation presumption road map aiming at criminal cases, group events and terrorist attack events on the basis of a worst risk situation evolution path by defining nodes and scene evolution rules among the nodes, and establishing a social security chain type coupling risk evolution scene library based on the situation presumption road map. Evolution is the description of the situation transition direction of the situation of the development of an emergency, wherein "state" refers to the current state of the event, and "potential" refers to the future development trend of the event. The risk evolution process of an event can be described by the evolution process of a scenario, which is a state change process of a scenario, i.e. a process in which a scenario changes from one state to another. For the risk scenario conjecture road map, the risk scenario conjecture road map is composed of nodes and directed branches, wherein the nodes represent 'states', and the branches represent 'potentials' of 'state' transition.
The above node definition refers to that the nodes of the risk scenario conjecture road map are divided into a start node, a state node and an end node, and the characteristics are shown in the following table 1:
table 1 context-inferred roadmap node definition
Figure BDA0002572528030000051
As can be seen from the scenario evolution rule schematic in fig. 3, for the scenario inference roadmap, each state node represents one state in the evolution, and the transition of "potential" specifies only three directions, i.e., "best trend", "worst trend", and "remain unchanged", the horizontal branch line represents the change of "state" to the best trend, the vertical branch line represents the worst trend, and the return branch line represents the remain unchanged. And based on SCJ model scene primitive, "potential" is also scene judgment, and is composed of scene response and scene conditions. The meaning of the scenario judgment transition trend is extended to be the best, worst and invariable scenario response and scenario condition or the existence, nonexistence and invariable scenario response and scenario condition.
Based on the node definition and the evolution transition rule, a risk evolution scenario conjecture road map is formed from left to right according to the sequence of occurrence and development of events, and the risk evolution scenario conjecture road map triggered by various social security events is formed through each state node under the trend judgment of scenario response and scenario conditions. Therefore, the evolving worst path formed by the first-step risk scenario recognition model appears at the lowest edge of the scenario inference path diagram, that is, the path formed by each node under the worst trend of scenario judgment.
Fig. 5 (fig. 5 a-5 c) are schematic diagrams of the risk evolution scenario of the social security incident in the present invention. Criminal cases, group events and terrorist attack events in social security events have different specific contents of evolution due to different respective event characteristics, wherein decision-making subjects of the criminal cases and the terrorist attack events are unified with personnel, and the decision-making subjects of the group events are opposite to the personnel. The overall evolution analysis of the social security event can be generally considered as two paths for the influence of the current event ending or the invalid disaster range of emergency disposal on the expansion of the emergency disposal.
Fig. 6 (fig. 6 a-6 c) is a road view for the criminal event fire scenario conjecture of the present invention. The fire-fighting case in criminal cases is considered as a risk evolution scenario library for personnel on purpose firing in buildings.
Fig. 7 (fig. 7 a-7 d) is a road view for the situation of criminal event explosion scenario of the present invention. The explosion case in criminal cases is considered as a library of risk evolution scenarios where personnel are detonated in public transport or buildings.
Fig. 8 (fig. 8 a-8 c) is a road view for estimating the criminal event holding injury scenario of the present invention. A case of holding a mechanical injury in a criminal case is considered as a risk evolution scenario library of persons injuring persons in public places.
Fig. 9 is a road diagram for estimating the criminal incident poison case scenario of the present invention. The poison throwing case in the criminal case is considered as a risk evolution scene library of continuous poison throwing behaviors when people are not found and identified.
Fig. 10 (fig. 10 a-10 c) are road charts of the group event scenario inference of the present invention, i.e. a risk evolution scenario library of group events. The group event is basically driven by the emotion of a person in the evolution, the emotion of the person has the reciprocity, and the reciprocity is realized in the evolution direction, so that the situation state of the node which is evolved in the early stage is recovered under the situation response when the network is established.
Fig. 11 (fig. 11 a-11 d) are schematic diagrams of the terrorist attack event scenario inference path, namely, a risk evolution scenario library of the terrorist attack event. Terrorist attack events primarily consider risk evolution scenarios in the form of explosive fire and armed attacks.
Fig. 12 is a functional deduction schematic diagram of the risk evolution scenario inference road diagram of the evaluation analysis of the present invention. The network operation analysis principle of the scenario conjecture roadmap is realized based on a Graphic Evaluation and review technology (graphical Evaluation and review technique), namely a characteristic transfer function algorithm of a GERT network. GEThe RT network is a network analysis method developed by a generalized random network technology. And obtaining a risk deduction result of a certain designated social security event chain risk evolution path according to the scenario conjecture path diagram based on the GERT network deduction principle. The method has two flows of input and output for a scene node based on the GERT network, and both have the meaning of scene judgment, wherein the flow of an input branch line is U1Representing the scene judgment of the previous node; flows with three transfer directions on the output branch line are judged according to the situation of the node, including the best trend U of state transfer2Worst trend U3And maintain a constant trend U0The stream U is represented on each branch by three parameters:
u (P, T, C) formula one
P represents the probability that the output branch is to be realized when the node of the arrow output end of the branch is realized; t represents casualties brought by the branch line under the state condition of two adjacent nodes in front and back, and is a random variable obeying certain probability distribution; c represents the property loss brought by the branch line under the state condition of two adjacent nodes in front and back, the branch line is a random variable obeying certain probability distribution, and for the random variable t and any real number s, the moment mother function of the random variable t is as follows:
Figure BDA0002572528030000061
where f (t) and p (t) are the probability density function for a continuous variable t and the probability distribution function for a discrete variable t, respectively.
According to the signal flow diagram theory, the equivalent transfer coefficient W between two arbitrary nodes i and j can be expressed as follows by applying the Meisen formula:
Figure BDA0002572528030000071
in the formula, xi，xjRespectively the variable values, W, of any two nodes i, j in the signal flow graphijIs the equivalent transfer coefficient from node i to node j, m is the order of the ring in the signal flow graph, PkTo be a slave nodeThe value of the transmission coefficient of the kth line from the i to the j is equal to the product of the transmission coefficients of the branches on the line, delta is a signal flow diagram characteristic expression, delta is the transmission coefficient of an odd ring from 1 to ∑ + the transmission coefficient of an even ring from ∑, and delta is equal tokFor the signatures of the remaining sub-graphs in the signal flow graph not in contact with the k-th line,
the characteristic transfer functions of two nodes in the scenario conjecture road map, namely the equivalent transfer coefficient W, are the product of the moment-mother function and the branch probability:
WE(s)＝ME(s)PEformula four
The equivalent transition probability P from the starting node to the ending node is obtainedEAnd the expected number of casualties E (t) is:
PE＝WE(s)|S＝0＝WE(0) formula five
Figure BDA0002572528030000072
The operation rule of the random variable c and any real number s also satisfies the characteristic transfer function, and the expected property loss number E (c) can be obtained through network operation and is as follows:
Figure BDA0002572528030000073
the risk deduction result is expressed by the risk degree, namely the degree value R of the accident risk is expressed by the product of the occurrence probability P of the accident and the severity S of the accident consequence, and the calculation formula is as follows:
formula eight where R is P × S
The expression of the consequence severity S is measured by two dimensions of casualty severity T and property loss severity C, and the risk expression of casualty and property loss is as follows:
RT＝PT× T formula nine
RC＝PC× c formula ten
Then, according to the deduction principle of the scenario conjecture road chart, the risk expression is as follows:
RT＝PEe (t) formula eleven
RC＝PEE (c) formula twelve
By conjecturing the urban community social security chain type coupling risk evolution scenes, the risk degree of each scene evolution chain type passage of the social security incident can be obtained. Based on quantitative analysis of the casualty severity and the property loss severity of the people, the method can obtain that after a social security event occurs, under different scene judgment, the direction and the consequence of event risk evolution have obvious difference. By introducing real-time monitoring data, the chain coupling risk evolution scenario conjecture method has reference value for making, managing and correcting emergency plans of social security incidents when the incidents do not occur, and has support value for emergency decisions based on a scenario-response mode after the incidents occur.
The order of the loop in the signal flow diagram is divided into a 1-order loop, a 2-order loop and an n-order loop according to the loop condition of the node in the signal flow diagram. The 1-order loop is a loop starting from one node and returning to the node; the 2-step ring is composed of two 1-step rings which are not contacted with each other and have no common node, and the value of the 2-step ring is equal to the product of the values of the two 1-step rings which are not contacted with each other and form the 2-step ring; the n-order ring is composed of n 1-order rings which are not in contact with each other and have no common node, and the value of the n-order rings is equal to the product of the values of the n 1-order rings.
Fig. 13 is a schematic diagram of feature parameter determination based on an evaluation unit in the scenario inference path of the present invention. An evaluation unit consists of a front node and a rear node i and j and a branch line which is communicated with the middle, and the evaluation principle is that after the state of the node i occurs, the node j is realized under the condition of judging the situation represented by the communicated branch line. Therefore, the value of the characteristic parameter P is the probability that the node i is transferred to the node j under the situation judgment represented by the branch line, i.e. the probability that j is realized under the situation judgment of the branch line after i occurs; the expected values of the characteristic parameters T and C and the satisfied probability distribution represent the conditions of casualties and property loss caused by the process of transferring the activity from the whole node i to the node j. And analyzing the single evaluation units on a certain risk evolution chain in the scenario conjecture road map one by one to obtain the risk state value of the whole risk evolution chain.
The values of the characteristic parameters P, T and C of the output stream U are obtained by historical data learning, expert experience methods or field monitoring data. Wherein the probability of occurrence in each evaluation unit, the expected values of casualties and property loss and the satisfied probability distribution can be obtained from the case base based on historical data learning. The monitoring number based on the site can be taken as information update to be carried into a scene presumption road map for calculation, and specific constant values for determining the occurrence probability of events and casualties and property loss are input. The opinion value based on the expert experience enables the opinions of the expert group to be consistent through the Delphi method, namely a group of experts with relatively familiar related knowledge is selected, a risk evolution chain path of a social security event is selected for parameters P, T and C of the branch flow, a questionnaire based on each evaluation unit is established, then the experts score the parameters to judge the values, and then the expert opinions are arranged for statistical analysis. And judging the deviation degree of the expert opinions through the variation coefficient, and judging whether the opinions of the experts tend to be consistent by utilizing consistency test. The expert opinion tends to be consistent when the kronebara coefficient alpha is greater than 0.8. The formula for coefficient of variation and consistency tests is shown below:
Figure BDA0002572528030000081
Figure BDA0002572528030000091
in the formula, ViThe coefficient of variation representing the ith question represents the degree of fluctuation of the expert's opinion with respect to this question,ithe standard deviation of the i-th order is shown,
Figure BDA0002572528030000092
then mean value is represented, α represents the Kranbaha coefficient value, σ2X represents the variance, σ, of the expert's scoring results for all questions2Y represents the variance of all the scoring results for a particular question, and K represents the number of questions.
Example 2
FIG. 14 is a simplified computational diagram of one embodiment of the present invention. The figure is a simplified diagram of an event chain selected from a population event scenario library. When the community aggregation event occurs due to interest disputes, the dispersion effect is poor after the dispersion education of the crowd personnel in the initial stage, and the crowd on the site never leaves and continues to aggregate on the site. Then, through latest effective information distribution, rumors generated in the gathered people are controlled, discontent emotions of the gathered people are resolved and calmed, and then the gathered people leave under the current persuasion and emergency evacuation of police force. According to the occurrence, development and evolution process of the emergency, an event chain from the occurrence of the group event at the starting point to the end point of the number 2 of the dangerous situation presumption road map in the group event scene library is selected. According to the numbers of fig. 14, the specific contents of the intermediate nodes and the scenario judgment of the path are shown in table 2; according to the connotation of a single evaluation unit, the event risk scenario chain is divided into 6 evaluation units in total, wherein the 6 evaluation units are 1-2, 2-3, 3-2, 3-4, 4-4 and 4-5, and the specific contents are shown in table 2:
table 2 embodiment path intermediate node and scenario judgment contents
Figure BDA0002572528030000093
According to the actual evolution conditions, and finally through comprehensive arrangement and correction of historical data and expert experience, the probability and distribution parameters of the GERT network related to the given risk evolution scenario are shown in the following table 3:
TABLE 3 Risk evolution scenarios GERT network parameters
Figure BDA0002572528030000094
Figure BDA0002572528030000101
In this network, the 1 st order ring is (W)2，W3) And W52-ring is (W)2，W3，W5) Then the network is characterized by:
Δ＝1-W2W3-W5+W2W3W5
the equivalent transmission probability of the network is P1The network corresponding characteristic is Δ1The equivalent characteristic transfer function of the network is WE(s)：
P1＝W1W2W4W6
Δ1＝1-0＝1
Figure BDA0002572528030000102
Therefore, the equivalent occurrence probability of the group accident event chain is that casualties are:
PE＝WE(0)＝0.706
Figure BDA0002572528030000111
Figure BDA0002572528030000112
therefore, after people are gathered in the urban community, the initial direct people evacuation effect is poor, but after the police release the latest message to calm down the curiosity or dissatisfaction of the gathered people, the gathered people leave the scene after being evacuated again by the police, and the casualty risk of the group event is RTThe risk of property loss is Rc：
RT＝PEE(t)＝0.66
RC＝PEE(c)＝0.35
Through network calculation, 0.66 persons are casualty and 0.35 ten thousand yuan is lost in the event chain of the urban community group events.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (7)

1. A method for conjecturing the urban community social security chain coupling risk evolution scene is characterized by comprising the following steps:
s1, establishing a city community social security event risk situation recognition model, and recognizing the risk situation and the evolution scene of each social security event;
s2, establishing an urban community social security event risk situation evolution model according to the risk situation recognition model to form a social security event chain risk evolution situation library;
s3, evaluating and analyzing the risk state of the evolution of each social security event of the urban community according to the risk situation evolution model;
in step S2, among others:
SCJ model scene elements are minimum units of evolution, a risk evolution process is divided into state evolution among single scene elements, and for each element promoting the risk evolution to develop, the SCJ model scene elements are defined by three attributes, namely scene states, and are expressed by a disaster-bearing body and the corresponding states thereof; disaster-causing elements consisting of disaster-causing factors and a pregnant disaster environment; scene judgment, which consists of scene response and scene conditions;
the scene response is an emergency response or decision aiming at the scene, and the scene condition is an external environment condition which is in the scene and can influence the evolution; for SCJ model scene elements, each scene state has disaster-causing factors, a pregnant disaster environment and scene responses or scene conditions for determining the evolution direction of the scene state; if the scene judgment of the previous scene element is matched with the disaster-causing element of the next scene element, the two scene elements have a sequential evolution transfer relationship;
the social security event risk scenario evolution model is a risk evolution scenario conjecture roadmap which is made on the basis of a worst risk scenario evolution path and an evolution scenario state by defining nodes and scenario evolution rules among the nodes, and a social security chain type coupling risk evolution scenario library based on the scenario conjecture roadmap is established; evolution is the description of the situation transition direction of the situation of the occurrence and development of an emergency, wherein the state refers to the current state of the event, and the state refers to the future development trend of the event; the risk evolution process of the event can be described by the evolution process of the scenario, and the evolution process of the scenario is the state change process of the scenario; for the risk scenario conjecture road map, the risk scenario conjecture road map consists of nodes and directed branches, wherein the nodes represent 'states', and the branches represent 'potentials' of 'state' transfer;
defining nodes refers to that nodes of the risk scenario conjecture road map are divided into start nodes, state nodes and end nodes, and scenario evolution rules of the nodes are transfer rules among the state nodes;
forming a risk evolution scenario conjecture road map from left to right according to the sequence of event occurrence and development based on node definition and evolution transition rules, and forming a risk evolution scenario conjecture road map triggered by various social security events under the trend judgment of scenario response and scenario conditions through each state node; therefore, the evolving worst path formed by the first-step risk scenario recognition model appears at the lowest edge of the scenario conjecture diagram, namely, the path formed by each node under the worst trend of scenario judgment.
2. The method for conjecturing the evolution scenario of the safe chain type coupled risks in urban community society of claim 1, wherein in step S2:
the meaning of the starting node is that a specific social security event occurs, which represents that initial influence and loss are generated;
the state node includes:
the accident type node represents the accident type possibly generated in the event evolution process;
the scene State type node is a State State in the SCJ model scene elements and is a set of disaster-bearing bodies and states;
a situation description type node, namely, the degree or development state of the situation evolution situation is described;
the significance of the end node is representing casualties and property loss, namely the final influence of accidents of each event, and the end nodes of different paths represent different degrees of consequences.
3. The method for conjecturing the evolution scenario of the safe chain type coupled risks in urban community society of claim 2, wherein the method comprises the following steps in step S2:
for the scenario conjecture, each state node represents one state in the evolution, the transition of the potential is only provided with three directions, namely, a best trend, a worst trend and a constant maintenance, the horizontal branch represents the change of the state from the best trend to the longitudinal branch, and the return branch represents the constant maintenance; based on SCJ model scene primitives, the situation is composed of scene responses and scene conditions, and the meaning of the scene judgment transition trend is extended to be the scene responses and scene conditions of 'best, worst and unchanged' or the scene responses and scene conditions of 'existence, nonexistence and unchanged'.
4. The method for conjecturing the evolution situation of the urban community social security chain coupling risk as claimed in claim 1, wherein the method comprises step S3, in the step of evaluating and analyzing the risk status of the evolution of each social security event in the urban community:
the network operation analysis principle of the scenario conjecture road map is realized by an algorithm based on a GERT network characteristic transfer function, a risk deduction result of a certain designated social security event chain type risk evolution channel can be obtained according to the scenario conjecture road map based on the GERT network deduction principle, and an input flow and an output flow both have the meaning of scenario judgment for a scenario node based on the GERT network, wherein the flow of an input branch line is U1Representing a scenario judgment for the last node; flows with three transfer directions on the output branch line are judged according to the situation of the node and comprise the best trend U of state transfer2Worst trend U3And maintain a constant trend U0The stream U is represented on each branch by three parameters:
u (P, T, C) formula one
P represents the probability that the output branch is to be realized when the node of the arrow output end of the branch is realized; t represents casualties brought by the branch line under the state condition of two adjacent nodes in front and back, and is a random variable obeying certain probability distribution; c represents the property loss brought by the branch line under the state condition of two adjacent nodes in front and back, the branch line is a random variable obeying certain probability distribution, and for the random variable t and any real number s, the moment mother function of the random variable t is as follows:
Figure FDA0002572528020000021
wherein f (t) and p (t) are the probability density function when t is a continuous variable and the probability distribution function when t is a discrete variable,
according to the signal flow diagram theory, the equivalent transfer coefficient W between two arbitrary nodes i and j can be expressed as follows by applying the Meisen formula:
Figure FDA0002572528020000031
in the formula, xi，xjRespectively the variable values, W, of any two nodes i, j in the signal flow graphijIs the equivalent transfer coefficient from node i to node j, m is the order of the ring in the signal flow graph, PkThe transfer coefficient of the kth line from the node i to the node j is equal to the product of the transfer coefficients of the branches on the line, delta is a characteristic expression of the signal flow diagram, the transfer coefficient of the odd ring is 1- ∑, the transfer coefficient of the even ring is ∑, and the transfer coefficient of the k line from the node i to the node j is deltakFor the signatures of the remaining sub-graphs in the signal flow graph not in contact with the k-th line,
the characteristic transfer functions of two nodes in the scenario conjecture road map, namely the equivalent transfer coefficient W, are the product of the moment-mother function and the branch probability:
WE(s)＝ME(s)PEformula four
The equivalent transition probability P from the starting node to the ending node is obtainedEAnd the expected number of casualties E (t) is:
PE＝WE(s)|S＝0＝WE(0) formula five
Figure FDA0002572528020000032
The operation rule of the random variable c and any real number s also satisfies the characteristic transfer function, and the expected property loss number E (c) can be obtained through network operation and is as follows:
Figure FDA0002572528020000033
the risk deduction result is expressed by the risk degree, namely the degree value R of the accident risk is expressed by the product of the occurrence probability P of the accident and the severity S of the accident consequence, and the calculation formula is as follows:
formula eight where R is P × S
The expression of the consequence severity S is measured by two dimensions of casualty severity T and property loss severity C, and the risk expression of casualty and property loss is as follows:
RT＝PT× T formula nine
RC＝RC× C formula ten
Then, according to the deduction principle of the scenario conjecture road chart, the risk expression is as follows:
RT＝PEe (t) formula eleven
RC＝PEE (c) formula twelve
Through the conjecture of the urban community social security chain type coupling risk evolution scenes, the risk degree of each scene evolution chain type passage of the social security incident can be obtained, and after the social security incident occurs, under the judgment of different scenes, the obvious difference exists between the direction and the consequence of the incident risk evolution;
the order of a loop in the signal flow diagram is divided into a 1-order loop, a 2-order loop and an n-order loop according to the loop condition of a node in the signal flow diagram, wherein the 1-order loop is a loop starting from one node and returning to the node; the 2-step ring is composed of two 1-step rings which are not contacted with each other and have no common node, and the value of the 2-step ring is equal to the product of the values of the two 1-step rings which are not contacted with each other and form the 2-step ring; the n-order ring is composed of n 1-order rings which are not contacted with each other and have no common node, and the value of the n-order ring is equal to the product of the values of the n 1-order rings;
the principle of determining the characteristic parameters based on one evaluation unit in the scenario conjecture road map is as follows: an evaluation unit consists of a front node and a rear node i, j and a branch line which is communicated with the middle, the evaluation principle is that after the state of the node i occurs, the node j is realized under the condition of the situation represented by the communicated branch line, and the value of the characteristic parameter P is the probability of transferring the node i to the node j under the condition of the situation represented by the branch line, namely the probability of realizing the node j under the condition of the situation judgment of the branch line after the occurrence of i; the expected values of the characteristic parameters T and C and the satisfied probability distribution are the conditions of casualties and property loss caused by the whole activity process of transferring the node i to the node j, and the risk state value of the whole risk evolution chain is obtained by analyzing single evaluation units on a certain risk evolution chain in the scenario conjecture road map one by one;
the values of the characteristic parameters P, T and C of the output stream U are obtained by historical data learning, expert experience methods or field monitoring data.
5. The method for conjecturing the urban community social security chain type coupling risk evolution scenario as claimed in claim 4, wherein the method obtains the occurrence probability in each evaluation unit, the expected values of casualties and property losses and the satisfied probability distribution from the case base based on historical data learning; updating and substituting the monitoring number as information into a scene presumption road map for calculation based on the field, and inputting specific constant values for determining the occurrence probability of an event and casualties and property loss; the opinion value based on expert experience enables opinions of expert groups to be consistent through a Delphi method, namely a group of experts with relatively familiar related knowledge is selected, a risk evolution chain path of social security events is selected for parameters P, T and C of branch flow to establish a questionnaire based on each evaluation unit, then the experts score and judge parameter values and then arrange the expert opinions, statistical analysis is carried out, the deviation degree of the expert opinions is judged through a variation coefficient, whether the opinions of the experts tend to be consistent or not is judged through consistency test, the experts tend to be consistent when the Kranbaha coefficient alpha is larger than 0.8, and the variation coefficient and the formula of the consistency test are as follows:
Figure FDA0002572528020000041
Figure FDA0002572528020000051
in the formula, ViThe coefficient of variation representing the ith question represents the degree of fluctuation of the expert's opinion with respect to this question,ithe standard deviation of the i-th order is shown,
Figure FDA0002572528020000052
then mean value is represented, α represents the Kranbaha coefficient value, σ2X represents the variance, σ, of the expert's scoring results for all questions2Y represents the variance of all the scoring results for a particular question, and K represents the number of questions.
6. The method for conjecturing the evolution scenario of the safe chain type coupled risks in urban community society of claim 1, wherein the method comprises the following steps of S1:
under the dimensionality of urban communities, social security events are considered as major criminal cases, group events and terrorist attack events, and explosion cases, fire cases, mechanical injury cases and poison throwing cases are considered in the major criminal cases; the risk scenario identification model obtains the worst accident scenario path and scenario state of risk evolution by identifying accidents of various events and SCJ model scenario primitives of accident scenario occurrence and development elements.
7. The method for conjecturing the evolution scenario of the safe chain type coupled risks in urban community society of claim 6, wherein the method comprises the following steps in step S1:
the method comprises the steps of dynamically identifying accidents of three social security events and SCJ model scene primitives by an expert experience method, updating knowledge information and monitoring information to obtain more risk scenes, and determining a worst risk evolution scene and each scene state by comparing disaster consequences of each risk scene.
CN202010644533.9A 2020-07-07 2020-07-07 Urban community social security chain coupling risk evolution scenario conjecture method Active CN111784060B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010644533.9A CN111784060B (en) 2020-07-07 2020-07-07 Urban community social security chain coupling risk evolution scenario conjecture method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010644533.9A CN111784060B (en) 2020-07-07 2020-07-07 Urban community social security chain coupling risk evolution scenario conjecture method

Publications (2)

Publication Number Publication Date
CN111784060A true CN111784060A (en) 2020-10-16
CN111784060B CN111784060B (en) 2021-04-23

Family

ID=72757985

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010644533.9A Active CN111784060B (en) 2020-07-07 2020-07-07 Urban community social security chain coupling risk evolution scenario conjecture method

Country Status (1)

Country Link
CN (1) CN111784060B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116485587A (en) * 2023-04-21 2023-07-25 深圳润高智慧产业有限公司 Community service acquisition method, community service providing method, electronic device and storage medium
WO2023141913A1 (en) * 2022-01-28 2023-08-03 华为技术有限公司 Risk treatment method and related device

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109377437A (en) * 2018-11-21 2019-02-22 南京莱斯信息技术股份有限公司 Urban Emergency emergency inference pattern based on event chain
CN109993508A (en) * 2019-04-23 2019-07-09 中国科学院遥感与数字地球研究所 A kind of earthquake disaster scene evolutionary model
CN109999311A (en) * 2018-01-05 2019-07-12 陈锦焜 The method of scenarios generation system and scenarios generation
CN110348578A (en) * 2019-07-04 2019-10-18 北京仿真中心 A kind of security incident scene deduces construction method, system, equipment and medium
CN111080139A (en) * 2019-12-19 2020-04-28 石家庄铁道大学 Hypergraph-based public security event scenario representation method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109999311A (en) * 2018-01-05 2019-07-12 陈锦焜 The method of scenarios generation system and scenarios generation
CN109377437A (en) * 2018-11-21 2019-02-22 南京莱斯信息技术股份有限公司 Urban Emergency emergency inference pattern based on event chain
CN109993508A (en) * 2019-04-23 2019-07-09 中国科学院遥感与数字地球研究所 A kind of earthquake disaster scene evolutionary model
CN110348578A (en) * 2019-07-04 2019-10-18 北京仿真中心 A kind of security incident scene deduces construction method, system, equipment and medium
CN111080139A (en) * 2019-12-19 2020-04-28 石家庄铁道大学 Hypergraph-based public security event scenario representation method

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
Y ZHANG等: "A scenario-based model for earthquake emergency management effectiveness evaluation", 《TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE》 *
付娉娉: "基于情景推演的非常规突发事件应急决策研究", 《中国博士学位论文全文数据库 社会科学Ⅰ辑》 *
刘浩然 等: "基于情景推演的机场冲/偏出跑道应急能力评估", 《航空计算技术》 *
周愉峰: "基于情景推演的地震灾害演化动态 GERT 网络模型", 《自然灾害学报》 *
朱晓寒等: "自然灾害链情景态势组合推演方法", 《管理评论》 *
陈英达: "突发事件情景间演化关系建模及推演方法研究", 《中国优秀硕士学位论文全文数据库 社会科学Ⅰ辑》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023141913A1 (en) * 2022-01-28 2023-08-03 华为技术有限公司 Risk treatment method and related device
CN116485587A (en) * 2023-04-21 2023-07-25 深圳润高智慧产业有限公司 Community service acquisition method, community service providing method, electronic device and storage medium
CN116485587B (en) * 2023-04-21 2024-04-09 深圳润高智慧产业有限公司 Community service acquisition method, community service providing method, electronic device and storage medium

Also Published As

Publication number Publication date
CN111784060B (en) 2021-04-23

Similar Documents

Publication Publication Date Title
Wilmot et al. Comparison of alternative trip generation models for hurricane evacuation
CN109948000B (en) Abnormal target detection method, device, equipment and storage medium for heterogeneous network
CN111784060B (en) Urban community social security chain coupling risk evolution scenario conjecture method
CN106161209B (en) A kind of method for filtering spam short messages and system based on depth self study
CN117591944B (en) Learning early warning method and system for big data analysis
KR20210125702A (en) Disaster prediction system and method using major keyword analysis technology of artificial intelligence
CN112200263B (en) Self-organizing federal clustering method applied to power distribution internet of things
CN111738044A (en) Campus violence assessment method based on deep learning behavior recognition
CN111126437A (en) Abnormal group detection method based on weighted dynamic network representation learning
CN111967729A (en) Industrialized personnel portrait evaluation method based on data mining
CN111967011B (en) Interpretable internal threat assessment method
CN113628059A (en) Associated user identification method and device based on multilayer graph attention network
Naili et al. Stability-based model for evacuation system using agent-based social simulation and Monte Carlo method
CN103838806A (en) Analysis method for subject participation behaviors of user in social network
Ageyev et al. Traffic Abnormality Detection ML-based Method for IoT
KR102519918B1 (en) Artificial intelligence-based data labeling method and system for recognizing objects at construction sites and determining risk situations
CN116884159A (en) Intelligent fire-fighting early warning method and system based on AI (advanced technology attachment) identification
Xu et al. Fire Safety Assessment of High-Rise Buildings Based on Fuzzy Theory and Radial Basis Function Neural Network.
CN116644954A (en) Intelligent decision-making method for construction safety risk of pumping and accumulating engineering
Su Design of intelligent classroom teaching scheme using artificial intelligence
Manocha et al. Digital Twin-assisted Fuzzy Logic-inspired Intelligent Approach for Flood Prediction
Li et al. Key influence factors on the evacuation route selection for fire emergencies in urban underground complexes
Asgary et al. Intelligent security systems engineering for modeling fire critical incidents: Towards sustainable security
CN114067436B (en) Fall detection method and system based on wearable sensor and video monitoring
CN117573809B (en) Event map-based public opinion deduction method and related device

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