CN106485327A - Under a kind of hazardous condition, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution - Google Patents

Abstract

The present invention relates to crowd tramples the Methods of Knowledge Discovering Based of behavior evolution under a kind of hazardous condition, comprise the following steps：1) extraction crowd tramples scenario factors feature, is emulated in colony intelligence crowd evacuation model, tramples evolved behavior feature during being evacuated, and builds the domain object set of domain space layer；2) trampling scenario factors feature to the crowd and trampling evolved behavior feature carries out rough set attribute sliding-model control；3) that sets up " conditional decision " two-dimensional signal pattern tramples evolution mechanism knowledge expression, and carries out Reduction of Knowledge to the evolution mechanism knowledge expression of trampling；4) meta-rule storehouse is generated according to the evolved behavior feature of trampling；5) load the meta-rule storehouse and example is evacuated, generate the abstraction rule storehouse for trampling evolution.Compared with prior art, by classical rough set theory, the crowd of realization tramples the automatic Knowledge Discovery of behavior evolution mechanism to the present invention, and the crowd of solution tramples the lack of knowledge of evolution mechanism.

Description

Under a kind of hazardous condition, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution

Technical field

The present invention relates to crowd tramples precautionary technology field, the crowd's behavior of trampling under a kind of hazardous condition that especially relates to is drilled The Methods of Knowledge Discovering Based of change.

Background technology

The analysis of accident is trampled currently for crowd and prevent also in the starting stage.Occur during crowd evacuation Trample what possibly " rumour-fear " pattern individually occurred, it is also possible to which (fire, earthquake, blast, poison gas are let out used as burst fire-disaster Leakage etc. accident) secondary disaster, with " disaster-fear " Mode Coupling generation.

Mostly existing crowd evacuation model is to focus on the research in terms of crowd movement is learned with dynamics, and main description crowd dredges Scattered behavior, and when analyzing building evacuation key element (stairs port and passageway for fire apparatus minimum widith etc.) and crowd's traffic capacity and evacuating Between relation, provide building some contingency management strategies for trampling of prevention crowd and suggestion.Most existing models not yet system Condition and the evolution mechanism of generation are trampled in research, are only examined as the typical unstable phenomenon in existing evacuation model using trampling Consider.

Existing crowd evacuation research conclusion is analyzed it is found that the elementary step is in for the research for trampling evolution mechanism, The problems such as still suffering from knowledge experience dependence and lack of knowledge.For Precautions are trampled, rest on the prediction scheme pipe based on experience more In the reason stage, for the real-time control problem that tramples, depend on the tissue experience of emergency evacuation floor manager personnel and crowd more Degree of cooperation.In the urgent need to adopting scientific method, analysis crowd tramples origin mechanism, and announcement crowd tramples Evolution.

Content of the invention

The purpose of the present invention is exactly to provide a kind of hazardous condition servant to overcome the defect of above-mentioned prior art presence Group tramples the Methods of Knowledge Discovering Based of behavior evolution, only relies only on actual domain record data itself, tight using rough set theory Combing logic, calculating and discovery crowd trample the knowledge of behavior evolution automatically.

The purpose of the present invention can be achieved through the following technical solutions：

Under a kind of hazardous condition, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, comprises the following steps：

1) extraction crowd tramples scenario factors feature, is emulated, obtain evacuating in colony intelligence crowd evacuation model Evolved behavior feature being trampled in journey, building the domain object set of domain space layer, the evolved behavior feature of trampling includes to step on Step on probability；

2) trampling scenario factors feature to the crowd and trampling evolved behavior feature is carried out at rough set attribute discretization Reason；

3) scenario factors feature is trampled as the conditional attribute of domain object using the crowd after the sliding-model control, with phase The evolved behavior feature of trampling that answers sets up " conditional decision " two-dimensional signal pattern as the decision attribute of domain object Evolution mechanism knowledge expression is trampled, and Reduction of Knowledge is carried out to the evolution mechanism knowledge expression of trampling；

4) meta-rule storehouse is generated according to the evolved behavior feature of trampling；

5) load the meta-rule storehouse and example is evacuated, generate the abstraction rule storehouse for trampling evolution.

The crowd tramples scenario factors feature to be included to evacuate individual physiologic factor, social factor, behavioural characteristic and environment Feature, the evacuation individuality physiologic factor include age, sex, disability degree, agility and body weight, and the social factor includes Strange degree, the behavioural characteristic include panic degree, and the environmental characteristic includes disaster factors and space constraint.

The fear degree is mapped as evacuating the individual of individuality according to Helbing, D. fear " Psychology and behavior " volatility model Body diameter, the solution procedure of the individuality diameter are specially：

f_iw={ A_iexp[(r_i-d_iw)/B_i]+kg(r_i-d_iw)}n_iw-γg(r_i-d_iw)v_it_iw(3)

In formula, m_iIt is the individual quality of i-th evacuation,It is the individual ideal velocity of i-th evacuation,It is i-th evacuation Individual direction initialization, v_iIt is the individual actual speed of i-th evacuation, τ_iIt it is i-th evacuation individual characteristic time, when t is Between, f_ijIt is to evacuate the interaction force between individuality i and evacuation individuality j, f_iwIt is to evacuate phase interaction of the individuality between i and border Firmly, A_i、B_iFor constant, d_cijIt is the individual mass centre's distance of two evacuations, d_ijDistance between individuality, n are evacuated for two_ijBe by Evacuate the standard vector that individuality j points to i, t_ijIt is n_ijTangential direction,It is the phasor difference of t speed, kg (d_ij-d_cij) represent mass force,Represent t force of sliding friction, k and γ is for determining to evacuate individuality i and j Between interaction blocking effect parameter, d_iwIt is to evacuate the distance between individuality i and border, n_iwRefer to vertical direction, t_iwRefer to tangential direction, r_iIt is the individual individual diameter of i-th evacuation, v_iIndividual individual speed is evacuated for i-th, g (x) is One function, if evacuate individuality collided, g (x)=0, otherwise g (x)=x.

The disaster factors are mapped as the individual individual speed of evacuating after panic propagation and are expressed, specially：

Defining rule θ is：

In formula, μ_DAFor the membership function of casualty loss degree DA,DA_maxFor maximum casualty loss degree, For risk assessment intensity I₀Membership function,I_maxIntensity is assessed for greateset risk, subscript t is h_disSequence Number, s is the sequence number of i value, and i is risk assessment intensity I₀On coordinate value, n for i maximum, r_stIt is in fuzzy relation matrix Element, using inference formula：

DA=I₀θR (5)

By risk assessment intensity I₀It is assigned on control point in the method for information distribution, finally obtains calamity source index h_disValue, R be fuzzy relation matrix；

Calculate each individual speed for individuality being evacuated after panic propagation：

h_dis=f (ρ) (6)

In formula, ρ is density of stream of people, and f () represents calamity source index h_disWith density of stream of people ρ institute linear relationship function, D_L =NA_P/W_AL_A=ρ A_P, D_LIt is the evacuation individual amount of unit area in horizontal plane, N is the total number of persons that walks in the stream of people, A_P For the horizontal projected area of single people, W_AFor the width of the stream of people, L_AFor the length of the stream of people, v_iIndividual individuality speed is evacuated for i-th Degree.

The crowd tramples scenario factors feature and is divided into qualitative features and quantitative characteristic, execution step 2) when, for qualitative Feature, is directly mapped as rough set Category Attributes；For quantitative characteristic, rough set connection attribute is first mapped as, so Adopt heuristic SOM self-organizing clustering model afterwards, automatic sliding-model control is carried out to rough set connection attribute, rough set is continuous Property value is converted into the distinguishable mathematic sign of rough set matrix.

The Reduction of Knowledge includes the domain space dimensionality reduction yojan carried out using full distance dimensionality reduction model and with coarse Attribute reduction and Value reduction that the theoretical yojan of collection and core computation model are carried out.

Reduction of Knowledge is carried out with rough set theory to be specially：

For knowledge system S=(U, A), U is domain object set, U={ x₁,x₂,…,x_n, element x therein_iIt is opinion Object in domain, n are object sum, and A is the community set of non-NULL, and A=C ∪ D, C are conditional attribute set, and D is decision attribute Collection, and C ∩ D=φ, calculate the differentiation matrix M of the knowledge system_DS(C)：

Element m in discrimination matrix_ijIt is to discriminate between object x_iAnd x_jAll conditions attribute set；

Define Boolean function f_DSAs follows：

Wherein, Boolean variableCorresponding to m conditional attributeSymbol ∨ Represent computing of extracting, symbol ∧ represents conjunction computing；

The decision matrix M of calculation knowledge system S_DS(C)：

Element in decision matrixIt is to discriminate between object x_iAnd x_jAll conditions attribute set；

Define decision function f_DRDSAs follows：

Here

Compared with prior art, the present invention has advantages below：

Structure crowd tramples the rough set Knowledge Discovery Model of evolution mechanism, and crowd is trampled the Knowledge Discovery of evolution It is converted into rough set decision rule and automatically generates problem.

Crowd is trampled to the feature extraction of scenario factors and the research of discretization, based on Self-organizing Maps (SOM) network Heuristic automatic cluster model, extraction crowd tramples key scene key element and evolved behavior feature, and discrete turns to rough set The conditional attribute of domain object and decision attribute, build and contain the two-dimensional signal table that crowd tramples evolution knowledge, with succinct and complete Standby mode proposes the knowledge-representation system that crowd tramples evolution mechanism.Known by rough set matrix computations auto-building model Know expression meta-rule, load crowd evacuation during trample situational examples, reduction tight for meta-rule can be become tool There is the abstraction rule of realistic background meaning, form Explicit Knowledge.The uniformity and completeness of above-mentioned each method, be system-computed simultaneously It was found that crowd tramples the valuable knowledge contained by evolved behavior, and scientific basis is provided, become the innovative point of the present invention.

Description of the drawings

Fig. 1 is the principle schematic of the present invention.

Specific embodiment

The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.The present embodiment is with technical solution of the present invention Premised on implemented, give detailed embodiment and specific operating process, but protection scope of the present invention be not limited to Following embodiments.

As shown in figure 1, the present embodiment provides the Methods of Knowledge Discovering Based that crowd under a kind of hazardous condition tramples behavior evolution, bag Include following steps：

Step one, extraction crowd trample scenario factors feature, emulated, dredged in colony intelligence crowd evacuation model Evolved behavior feature being trampled during dissipating, building the domain object set of domain space layer, trampling evolved behavior feature includes to step on Step on probability.

The crowd tramples scenario factors feature to be included to evacuate individual physiologic factor, social factor, behavioural characteristic and environment Feature, the evacuation individuality physiologic factor include age, sex, disability degree, agility and body weight etc., the social factor bag Strange degree is included, the behavioural characteristic includes panic degree, the environmental characteristic includes disaster factors and space constraint.

In the present invention, the fear degree is mapped as evacuating according to Helbing, D. fear " Psychology and behavior " volatility model Individual individual diameter, measures its panic degree from the individual tensity (Psychological phobia) of evacuation and velocity variations, by panic journey Degree is mapped as evacuating individual diameter, sets up panic quantitative model.

The solution procedure of the individuality diameter is specially：

f_iw={ A_iexp[(r_i-d_iw)/B_i]+kg(r_i-d_iw)}n_iw-γg(r_i-d_iw)v_it_iw(3)

In formula, m_iIt is the individual quality of i-th evacuation,It is the individual ideal velocity of i-th evacuation,It is i-th evacuation Individual direction initialization, v_iIt is the individual actual speed of i-th evacuation, τ_iIt it is i-th evacuation individual characteristic time, when t is Between, f_ijIt is to evacuate the interaction force between individuality i and evacuation individuality j, f_iwIt is to evacuate phase interaction of the individuality between i and border Firmly, A_i、B_iFor constant, d_cijIt is the individual mass centre's distance of two evacuations, d_ijDistance between individuality, n are evacuated for two_ijBe by Evacuate the standard vector that individuality j points to i, t_ijIt is n_ijTangential direction,It is the phasor difference of t speed, kg (d_ij-d_cij) represent mass force,Represent t force of sliding friction, k and γ is for determining to evacuate individuality i and j Between interaction blocking effect parameter, d_iwIt is to evacuate the distance between individuality i and border, n_iwRefer to vertical direction, t_iwRefer to tangential direction, r_iIt is the individual individual diameter of i-th evacuation, v_iIndividual individual speed is evacuated for i-th, g (x) is One function, if evacuate individuality collided, g (x)=0, otherwise g (x)=x.

The model data of the typical disaster such as fire and toxic gas leakage is derived from hazard model storehouse, extracts disaster to evacuation Key influence factor；Derive from building model storehouse and related building space structure data is evacuated, extract space about Key influence factor of the bundle (such as exit width and wall locations etc.) to evacuation.Will be special as environment to disaster factors and space constraint Levy, and classified according to qualitative and quantitative.Disaster is the trigger condition that crowd panic is propagated, and disaster factors are mapped to fear In propagation model, individual speed is evacuated in disaster factors impact.

In the present invention, the disaster factors are mapped as the individual individual speed of evacuating after panic propagation and are expressed, Specially：

Defining rule θ is：

In formula, μ_DAFor the membership function of casualty loss degree DA,DA_maxFor maximum casualty loss degree, For risk assessment intensity I₀Membership function,I_maxIntensity is assessed for greateset risk, subscript t is h_disSequence Number, s is the sequence number of i value, and i is risk assessment intensity I₀On coordinate value, n for i maximum, r_stIt is in fuzzy relation matrix Element, using inference formula：

DA=I₀θR (5)

By risk assessment intensity I₀It is assigned on control point in the method for information distribution, finally obtains calamity source index h_disValue, R be fuzzy relation matrix；

Calculate each individual speed for individuality being evacuated after panic propagation：

h_dis=f (ρ) (6)

In formula, ρ is density of stream of people, and f () represents calamity source index h_disWith density of stream of people ρ institute linear relationship function, D_L =NA_P/W_AL_A=ρ A_P, D_LIt is the evacuation individual amount of unit area in horizontal plane, N is the total number of persons that walks in the stream of people, A_P For the horizontal projected area of single people, W_AFor the width of the stream of people, L_AFor the length of the stream of people, v_iIndividual individuality speed is evacuated for i-th Degree.

Step 2, scenario factors feature is trampled to crowd and tramples evolved behavior feature and carry out at rough set attribute discretization Reason, carries out discretization using the characteristic attribute discretization algorithm based on SOM self-organized mapping network to the evacuation feature for extracting, and Carry out quantifying and dependent linearity mapping.

The crowd tramples scenario factors feature and is divided into qualitative features and quantitative characteristic, execution step 2) when, for qualitative Feature (as sex), is directly mapped as rough set Category Attributes (such as 1 and 0 etc.)；For quantitative characteristic (as age, fear Spend and trample probability etc.), rough set connection attribute is first mapped as, heuristic SOM self-organizing clustering model is then adopted, right Rough set connection attribute carries out automatic sliding-model control, and rough set continuous property is converted into the distinguishable number of rough set matrix Symbol (such as 1,2 ... etc.).

Step 3, trampled using the crowd after the sliding-model control scenario factors feature as domain object condition belong to Property, using the corresponding evolved behavior feature of trampling as the decision attribute of domain object, set up " conditional decision " two dimension letter Breath pattern trample evolution mechanism knowledge expression, and Reduction of Knowledge is carried out to the evolution mechanism knowledge expression of trampling.

In terms of evolution knowledge representation and reduction method research is trampled, the attribute discretization of scenario factors is trampled based on crowd Expression, using rough set theory Knowledge Representation Model (domain, attribute, property value, information function), analyzes scenario factors (as people Group's quantity, translational speed and panic degree etc.) relation between attribute and variation characteristic, formed knowledge-representation system attribute and Property value.

U={ x₁,x₂,x₃···x_n} (8)

Wherein, U is domain set, each element x in set, and x represents that crowd evacuation result (is dredged containing crowd in this patent Dissipate each characteristic value) wall scroll record.

With the physiology in scenario factors, society, psychology and environmental characteristic as the conditional attribute a of rough set domain object_i, with The decision attribute D that probability is rough set domain object of trampling in colony intelligence evacuation model simulation result, " condition is determined for proposition The information function expression formula of plan " two-dimensional signal table schema, forms the knowledge representation for trampling evolution mechanism, can be expressed as：

C={ a₁,a₂,a₃···a_n} (9)

Wherein, C is conditional attribute, can arrange a₁For age, a₂For sex, a₃For disability degree, a₄For agility, a₅For body Weight, a₆It is the individual diameter mapped by panic degree, a₇It is the individual speed mapped by disaster factors.D is decision kind set.

In terms of Reduction of Knowledge, under conditions of keeping domain object set classification capacity constant, using full distance dimensionality reduction mould Type merges domain object, carries out dimensionality reduction yojan to domain space, simplifies follow-up domain space and calculates.With rough set theory yojan (Reduct) redundant attributes of domain object and core (Core) computation model, are deleted, attribute reduction and Value reduction is carried out, with Simplify and trample evolution knowledge.If Q is independent, and Q ∈ C, if having

IND (Q)=IND (C) (10)

Then Q is a yojan of equivalence relation family C, and in C, all collection that can not save relation are combined into the core of equivalence relation family C, Note Core (C).There are multiple yojan, represent the set of all yojan of C with Red (C).

In the present invention, Reduction of Knowledge includes the domain space dimensionality reduction yojan carried out using full distance dimensionality reduction model and utilization Attribute reduction and Value reduction that rough set theory yojan and core computation model are carried out.

Step 4, trample the rough set Knowledge Discovery of evolution mechanism.Unit's rule are generated according to the evolved behavior feature of trampling Then storehouse, loads the meta-rule storehouse and evacuates example, generate the abstraction rule storehouse for trampling evolution.

Using classical Skowron matrix computational approach, rough set matrix computations model is built；For knowledge system S=(U, A, V, f), U is domain, x_iFor the object in domain, U={ x₁,x₂,…,x_n}.A is the community set of non-NULL, and A=C ∪ D, C are Conditional attribute, D are decision attributes, and C ∩ D=φ.V represents property value, and f is information function.For system S, condition is only considered Attribute, forms the information system with regard to conditional attribute C.Number of objects in the differentiation matrix DM of the system, its exponent number and domain Relevant, as n × n rank, is designated as M_DS(C)：

Element m in discrimination matrix_ijIt is to discriminate between object x_iAnd x_jAll conditions attribute set.For variable j, i Scope is defined as 1≤j≤i≤n, obtains lower triangular matrix, but as variable j=i, i.e. M_DS(C), diagonal entry is obtained, And object oneself result of the comparison is empty set φ certainly.So comparing to reduce n time, the scope to j, i is defined as here：1 ≤ j ＜ i≤n, reduces the computation complexity of original definition.

Introduce a Boolean function f_DS, referred to as resolution function, as follows：

Wherein, Boolean variableCorresponding to m conditional attributeSymbol ∨ Represent computing of extracting, symbol ∧ represents conjunction computing；

The decision matrix of calculation knowledge system S, its exponent number are relevant with the number of objects in domain, and as n × n rank, is designated as M_DS(C)：

With finding to the calculating of above-mentioned matrix, this definition only reflect decision attribute values identical beyond object compare Situation, without the description for conflict object (inconsistent).Conditional attribute value is identical and conflict that decision attribute values are different is right The matrix value of elephant remains as empty set, and this cannot just reflect the presence of conflict phenomenon, cannot more reflect the degree problem of conflict.For This, has carried out improving definition to the matrix, has eliminated limitation.

Element in decision matrixIt is to discriminate between object x_iAnd x_jAll conditions attribute set.

Define decision function f_DRDSAs follows：

Here

The purpose of rough set theory research is to simplify calculating, reduces computation complexity, optimized algorithm structure, more effectively Solve engineering problem.Summary achievement in research, in data preprocessing phase, carries out MAWD dimensionality reduction meter to domain space Calculate；For connection attribute, cluster calculation is carried out using the SOM network method based on modified IMDV；For quantitative attributes or The qualitative attribute of change little (i.e. value is relatively fixed), adopts direct-discrete method here.

So, the optimization achievement of links above is integrated into organic whole, just obtains the MS- of overall performance improvement VPRS Knowledge Discovery Model.Chapters and sections afterwards, if how research is applied to IMSRS model in engineering and medical domain In dry typical diagnostic case, it is enterprise customer and medical researchers' solving practical problems, also from Knowledge Discovery and decision support Aspect, the validity to the model and versatility carry out engineering verification.

Rough set matrix computations model can generate the meta-rule of attribute nucleus collection, and be input to meta-rule storehouse；Meta-rule can table It is shown as：

Rule 1：

if a₃=0and a₄=0and a₇=1.then d=0

Rule 2：

if a₃=1and a₄=0and a₇=1.then d=1

Rule 3：

if a₃=0and a₄=1and a₇=0.then d=1

Rule 4：

if a₃=1and a₄=0and a₇=1.then d=2

Meta-rule is loaded, using the reversible process analytic method of attribute discretization, by domain object properties with trample feelings Scape key element forms invertible mapping；Domain characteristics of objects value is reloaded, the mathematic sign (if gender attribute value is 0) of discretization is also Original becomes evacuation scenario factors (if sex is man)；The regular generating portion of evolution is being trampled, and example is being evacuated by loading, will be coarse Collection meta-rule is extensive to be become towards the concrete abstraction rule for evacuating example, forms the abstraction rule storehouse with practical guided significance, Form Explicit Knowledge.As rule 1, under disaster serious conditions, non-disability, overweight women old man, when not producing Psychological phobia The probability that tramples is 0.

A case verification said method with Hongqiao in Shanghai transport hub.

The first step, the extraction Shanghai Hongqiao Integrative Transport Hub construction example data that pass through, analysis and extraction environmental characteristic, mould The real conditions at the scene of drawing up, are emulated.(calculating according to State Statistics Bureau's population ratio data can for total number of persons N=2109 people Obtain 723 people of man, 686 people of woman, 348 people of children, 221 people of old man and 131 people of physical disabilities).

Second step, 4 record composition domain set U, and U={ x is extracted from whole emulation₁,x₂,x₃,x₄}.a₁For year Age, a₂For sex, a₃For disability degree, a₄For agility, a₅For body weight, a₆It is the individual diameter mapped by panic degree, a₇For The individual speed mapped by disaster factors, C={ a₁,a₂,a₃,a₄,a₅,a₆,a₇}.Wherein, old man is 2, and children are 1, in youth Year is 0；Man is 1, and female is 0；Disability is 1, and non-disability is 0；Agile is 2, and action is generally 1, and handicapped is 0；Overweight For 2, it is 1 less than normal type, normal type is 0；Panic is 2, and nervous is 1, is normally 0；Disaster scenarios it is seriously 1, disaster Situation is not seriously 0, as shown in table 1.

Table 1

U	a1	a2	a3	a4	a5	a6	a7	D
									x1	2	0	0	0	2	0	1	0
x2	0	0	1	0	0	1	1	1
									x3	0	1	0	1	0	2	0	1
x4	1	1	1	0	1	2	1	2

3rd step, obtain differentiation matrix, resolution function and decision function.

According to formula (11), differentiation matrix is obtained：

Resolution function is：

f_DS=(a₁∨a₂∨a₃∨a₅∨a₆)(a₁∨a₂∨a₄∨a₅∨a₆∨a₇)

(a₂∨a₃∨a₄∨a₆∨a₇)(a₁∨a₂∨a₃∨a₅∨a₆)

(a₁∨a₂∨a₅∨a₆)(a₁∨a₃∨a₄∨a₅∨a₇)

=a₁a₂a₄a₅a₇

According to formula (14), discrimination matrix is obtained：

Obtain decision function：

f_DRDS=(a₁∨a₂∨a₃∨a₅∨a₆)(a₁∨a₂∨a₄∨a₅∨a₆∨a₇)

(a₂∨a₃∨a₄∨a₆∨a₇)(a₁∨a₂∨a₅∨a₆)

(a₁∨a₃∨a₄∨a₅∨a₇)

=a₃a₄a₇

After yojan as shown in table 2.

Table 2

U	a3	a4	a7	D
					x1	0	0	1	0
x2	1	0	1	1
					x3	0	1	0	1
x4	1	0	1	2

According to meta-rule, the element value in decision kind set D (D is the probability that tramples, in the range from 0-1) is obtained For；Pass through abstraction rule again, decision attribute is translated as Explicit Knowledge, d₁=0 is 0, d for the probability trampled₂=1 for sending out The probability that life is trampled is 0.1, d₃=1 is 0.1, d for the probability trampled₄=2 is 0.2 for the probability trampled.

The present invention proposes the Methods of Knowledge Discovering Based that crowd tramples evolution mechanism, evacuates scenario factors and people under hazardous condition Group tramples extraction correlated characteristic key element in evolved behavior, is calculated with existing rule base, obtains decision attribute values, prevents Pernicious accident of trampling provides scientific basis and theories integration, with important theory value and social effect.

Claims (7)

1. under a kind of hazardous condition, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that comprise the following steps：

1) extraction crowd tramples scenario factors feature, is emulated, during being evacuated in colony intelligence crowd evacuation model Evolved behavior feature is trampled, the domain object set of domain space layer is built, the evolved behavior feature of trampling includes to trample generally Rate；

2) trampling scenario factors feature to the crowd and trampling evolved behavior feature carries out rough set attribute sliding-model control；

3) scenario factors feature is trampled as the conditional attribute of domain object using the crowd after the sliding-model control, with corresponding The evolved behavior feature of trampling sets up trampling for " conditional decision " two-dimensional signal pattern as the decision attribute of domain object Evolution mechanism knowledge expression, and Reduction of Knowledge is carried out to the evolution mechanism knowledge expression of trampling；

4) meta-rule storehouse is generated according to the evolved behavior feature of trampling；

5) load the meta-rule storehouse and example is evacuated, generate the abstraction rule storehouse for trampling evolution.

2. under hazardous condition according to claim 1, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that The crowd tramples scenario factors feature to be included to evacuate individual physiologic factor, social factor, behavioural characteristic and environmental characteristic, described Evacuating individual physiologic factor includes age, sex, disability degree, agility and body weight, and the social factor includes strange degree, The behavioural characteristic includes panic degree, and the environmental characteristic includes disaster factors and space constraint.

3. under hazardous condition according to claim 2, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that The fear degree is mapped as evacuating individual individual diameter, institute according to Helbing, D. fear " Psychology and behavior " volatility model The solution procedure for stating individual diameter is specially：

$m_i=\frac{{\mathrm{dv}}_i}{dt}=m_i\frac{v_i^0\left(t\right)e_i^0\left(t\right)-v_i\left(t\right)}{{\mathrm{\τ}}_i}+\underset{j\left(\≠i\right)}{\Σ}{f}_{ij}+\underset{w}{\Σ}{f}_{iw}---\left(1\right)$

$f_{ij}=\{A_i\exp \[(d_{ij}-d_{cij})/B_i\]+kg(d_{ij}-d_{cij})\}{n}_{ij}+\mathrm{\γ}g(d_{ij}-d_{cij}){\mathrm{\Δv}}_{ji}^t{t}_{ij}---\left(2\right)$

f_iw={ A_iexp[(r_i-d_iw)/B_i]+kg(r_i-d_iw)}n_iw-γg(r_i-d_iw)v_it_iw(3)

In formula, m_iIt is the individual quality of i-th evacuation,It is the individual ideal velocity of i-th evacuation,It is that i-th evacuation is individual Direction initialization, v_iIt is the individual actual speed of i-th evacuation, τ_iIt it is i-th evacuation individual characteristic time, t is the time, f_ij It is to evacuate the interaction force between individuality i and evacuation individuality j, f_iwIt is to evacuate interaction force of the individuality between i and border, A_i、B_iFor constant, d_cijIt is the individual mass centre's distance of two evacuations, d_ijDistance between individuality, n are evacuated for two_ijIt is individual by evacuating Body j points to the standard vector of i, t_ijIt is n_ijTangential direction,It is the phasor difference of t speed, kg (d_ij-d_cij) Represent mass force,Represent t force of sliding friction, k and γ is mutual between decision evacuation individuality i and j The parameter of the blocking effect of effect, d_iwIt is to evacuate the distance between individuality i and border, n_iwRefer to vertical direction, t_iwRefer to tangential Direction, r_iIt is the individual individual diameter of i-th evacuation, v_iIndividual individual speed is evacuated for i-th, g (x) is a function, such as Fruit is evacuated individuality and collides, g (x)=0, otherwise g (x)=x.

4. under hazardous condition according to claim 2, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that The disaster factors are mapped as the individual individual speed of evacuating after panic propagation and are expressed, specially：

Defining rule θ is：

${\mathrm{\μ}}_{DA}\left({\left(h_{dis}\right)}_t\right)=min\{1,\underset{s=1}{\overset{n}{\Σ}}{\mathrm{\μ}}_{t_0}\left(i_s\right)r_{st}\}---\left(4\right)$

In formula, μ_DAFor the membership function of casualty loss degree DA,DA_maxFor maximum casualty loss degree,For wind Danger assessment intensity I₀Membership function,I_maxIntensity is assessed for greateset risk, subscript t is h_disSequence number, s It is the sequence number of i value, i is risk assessment intensity I₀On coordinate value, n for i maximum, r_stIt is the unit in fuzzy relation matrix Element, using inference formula：

DA=I₀θR (5)

By risk assessment intensity I₀It is assigned on control point in the method for information distribution, finally obtains calamity source index h_dis's Value, R are fuzzy relation matrix；

Calculate each individual speed for individuality being evacuated after panic propagation：

h_dis=f (ρ) (6)

$v_i=1.867D_L^4-6.333D_L^3+7.233D_L^2-3.617D_L+0.95---\left(7\right)$

In formula, ρ is density of stream of people, and f () represents calamity source index h_disWith density of stream of people ρ institute linear relationship function, D_L= NA_P/W_AL_A=ρ A_P, D_LIt is the evacuation individual amount of unit area in horizontal plane, N is the total number of persons that walks in the stream of people, A_PFor The horizontal projected area of single people, W_AFor the width of the stream of people, L_AFor the length of the stream of people, v_iIndividual individuality speed is evacuated for i-th Degree.

5. under hazardous condition according to claim 1, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that The crowd tramples scenario factors feature and is divided into qualitative features and quantitative characteristic, execution step 2) when, for qualitative features, directly It is mapped as rough set Category Attributes；For quantitative characteristic, rough set connection attribute is first mapped as, then using inspiration Formula SOM self-organizing clustering model, carries out automatic sliding-model control to rough set connection attribute, and rough set continuous property is converted For the distinguishable mathematic sign of rough set matrix.

6. under hazardous condition according to claim 1, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that The Reduction of Knowledge is included the domain space dimensionality reduction yojan carried out using full distance dimensionality reduction model and uses rough set theory about Attribute reduction and Value reduction that letter and core computation model are carried out.

7. under hazardous condition according to claim 6, crowd tramples the Methods of Knowledge Discovering Based of behavior evolution, it is characterised in that Reduction of Knowledge is carried out with rough set theory to be specially：

For knowledge system S=(U, A), U is domain object set, U={ x₁,x₂,…,x_n, element x therein_iFor in domain Object, n be object sum, A for non-NULL community set, A=C ∪ D, C are conditional attribute set, and D is decision kind set, and C ∩ D=φ, calculates the differentiation matrix M of the knowledge system_DS(C)：

Element m in discrimination matrix_ijIt is to discriminate between object x_iAnd x_jAll conditions attribute set；

Define Boolean function f_DSAs follows：

Wherein, Boolean variableCorresponding to m conditional attribute a₁,...,a_m,Symbol ∨ represents Extract computing, symbol ∧ represents conjunction computing；

The decision matrix M of calculation knowledge system S_DS(C)：

$M_{DRDS}\left(C\right)=\[m_{{\mathrm{DS}}_{ij}}\left(x_i,x_j\right)\]=\left\{\begin{array}{c}\left\{a\∈C|a\left(x_i\right)\≠a\left(x_j\right)andd\left(x_i\right)\≠d\left(x_j\right)\right\}\\ -1,a\left(x_i\right)=a\left(x_j\right)andd\left(x_i\right)\≠d\left(x_j\right)\\ \mathrm{\φ},d\left(x_i\right)=d\left(x_j\right)\end{array}\right.---\left(10\right)$

Element in decision matrixIt is to discriminate between object x_iAnd x_jAll conditions attribute set；

Define decision function f_DRDSAs follows：

Here