CN106897826A - A kind of street accidents risks appraisal procedure and system - Google Patents

A kind of street accidents risks appraisal procedure and system Download PDF

Info

Publication number
CN106897826A
CN106897826A CN201710099426.0A CN201710099426A CN106897826A CN 106897826 A CN106897826 A CN 106897826A CN 201710099426 A CN201710099426 A CN 201710099426A CN 106897826 A CN106897826 A CN 106897826A
Authority
CN
China
Prior art keywords
risk
matrix
risk factors
sample set
evaluation model
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.)
Pending
Application number
CN201710099426.0A
Other languages
Chinese (zh)
Inventor
于德新
林赐云
张伟
胡馨月
龚勃文
杨庆芳
周户星
郑黎黎
王树兴
马晓刚
商强
杨帆
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.)
SHANDONG EXPRESSWAY CO Ltd
Jilin University
Original Assignee
SHANDONG EXPRESSWAY CO Ltd
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by SHANDONG EXPRESSWAY CO Ltd, Jilin University filed Critical SHANDONG EXPRESSWAY CO Ltd
Priority to CN201710099426.0A priority Critical patent/CN106897826A/en
Publication of CN106897826A publication Critical patent/CN106897826A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

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

Abstract

The invention discloses a kind of street accidents risks appraisal procedure and system, the method and system are by obtaining freeway tunnel street accidents risks factor and freeway tunnel street accidents risks grade, using spectral clustering, calculate and determine corresponding relation chain between risk factors and risk class；And according to relation chain and limit learning algorithm, build risk evaluation model；Any risk facior data is input to risk evaluation model, it is possible to obtain the ultimate risk grade corresponding to the risk factors.Therefore, the method or system provided using the present invention, freeway tunnel street accidents risks assessment efficiency is effectively improved, makes freeway tunnel street accidents risks evaluation process more efficient, convenient, accurately and effectively realize the safety evaluation that freeway tunnel runs the phase.

Description

A kind of street accidents risks appraisal procedure and system
Technical field
The present invention relates to Freeway Traffic Engineering field, the traffic thing that more particularly to a kind of freeway tunnel occurs Therefore methods of risk assessment and system.
Background technology
In recent years, with the Large scale construction of freeway tunnel and increasing substantially for the volume of traffic, to freeway tunnel Safe operation generates huge pressure and challenge.Due to freeway tunnel own characteristic, freeway tunnel not only turns into The multi-happening section of traffic accident, and increased the difficulty of accident rescue.As China is asked freeway tunnel traffic safety Topic pay attention to day by day, some take a turn for the better freeway tunnel traffic safety situation, and accident occurrence frequency and the number of casualties have declined, but Traffic accident still can result in very big life and property loss, have a strong impact on the operational efficiency of highway, and easily draw Hair second accident.Therefore, freeway tunnel traffic safety status still can not be ignored.
At present, safe evaluation method of the China on freeway tunnel is concentrated mainly on the construction period in tunnel, not Relate to later stage freeway tunnel run the safety evaluation of phase.Therefore, how later stage highway is accurately and effectively realized The safety evaluation of phase is runed in tunnel, is current Freeway Traffic Engineering research urgent problem.
The content of the invention
It is an object of the invention to provide a kind of street accidents risks appraisal procedure and system, accurately and effectively to realize at a high speed Vcehicular tunnel runs the safety evaluation of phase.
To achieve the above object, the invention provides a kind of street accidents risks appraisal procedure, methods described, including：
Obtain the risk factors of freeway tunnel traffic accident；
Obtain the risk class of freeway tunnel traffic accident；
According to the risk factors, the risk class corresponding to each described risk factors is determined；
Risk class according to corresponding to the risk factors and the risk factors, determines risk factors and described Relation chain between risk class；
According to the relation chain, risk evaluation model is set up；
According to the risk evaluation model, the street accidents risks coefficient of freeway tunnel is assessed.
Optionally, the relation chain determined between the risk factors and the risk class, specifically includes：
By spectral clustering, multigroup risk factors are clustered, obtain multiple clustering clusters；
Risk class according to corresponding to the risk factors, determine single described risk in each described clustering cluster because Risk class corresponding to element, and then determine the corresponding ultimate risk grade of each described clustering cluster；
According to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, risk factors and described are determined Relation chain between risk class.
Optionally, it is described to be clustered multigroup risk factors by spectral clustering, multiple clustering clusters are obtained, specifically Including：
According to the risk factors, it is determined that cluster sample set X；The cluster sample set X is expressed as X={ x1,…,xn, ε =1,2 ... n；
According to the risk class of the freeway tunnel traffic accident for obtaining, clusters number k is determined；
According to the cluster sample set X, it is the arest neighbors Affinity diagram W of k to build the clusters number；
Calculate Laplacian matrix Ls and standardize；The Laplacian matrix Ls are the arest neighbors Affinity diagram Laplacian matrixes, are defined as L=D-W；The standardized Laplacian matrix Ls are matrix Lsym, the matrix LsymTable Show Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Calculate the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…,ukAs row, structural matrix U；Institute State matrix U and represent U ∈ Rn×k；
Every a line of standardization matrix U, obtains matrix P；The matrix P represents P ∈ Rn×k；
By each row vector y of the matrix Pε∈RkAs a data point, using k-means algorithms to yεGathered Class, obtains multiple clustering clusters；Wherein, the ε row data y of the matrix PεClassification and raw data points xεClassification it is identical.
Optionally, it is described to set up risk evaluation model, specifically include：
According to the relation chain, training sample set is determined；
According to the training sample set, using extreme learning machine algorithm, risk evaluation model is set up；The risk assessment mould Type is：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent The ε output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent λ partially Put.
Optionally, it is described according to the training sample set, using extreme learning machine algorithm, risk evaluation model is set up, have Body includes：
Input weights α is generated using extreme learning machine algorithm at randomλWith biasing bλ, λ=1,2 ..., N；
According to the training sample set, the input weights αλAnd the biasing bλ, obtain hidden layer output matrix H；Institute State hidden layer output matrix H tablesWherein, g (x) is represented and swashed Function living, N represents node in hidden layer；
According to the hidden layer output matrix H, output weight vector β is obtained；The output weight vector β be expressed as β= H-1T；Wherein, t=[t1,t2,…,tn] be each training sample set expected risk grade, T represents
According to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, set up Risk evaluation model.
To achieve the above object, present invention also offers a kind of street accidents risks assessment system, the system includes：
Risk factors acquisition module, the risk factors for obtaining freeway tunnel traffic accident；
Risk class acquisition module, the risk class for obtaining freeway tunnel traffic accident；
Risk factors correspondence risk class determining module, for according to the risk factors, determine each described risk because Risk class corresponding to element；
Relation chain determining module, for the risk class according to corresponding to the risk factors and the risk factors, really Fixed relation chain between the risk factors and the risk class；
Risk evaluation model sets up module, for according to the relation chain, setting up risk evaluation model；
Evaluation module, for according to the risk evaluation model, assessing the street accidents risks coefficient of freeway tunnel.
Optionally, the relation chain determining module, specifically includes：
Clustering cluster obtains unit, for by spectral clustering, multigroup risk factors being clustered, obtains multiple clusters Cluster；
Clustering cluster correspondence ultimate risk level de-termination unit, for the risk class according to corresponding to the risk factors, Determine the single described risk factors correspondence risk class in each described clustering cluster, and then determine each described clustering cluster correspondence Ultimate risk grade；
Relation chain determining unit, for according to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, really Fixed relation chain between the risk factors and the risk class.
Optionally, the clustering cluster obtains unit, specifically includes：
Cluster sample collected works determining unit, for according to the risk factors, it is determined that cluster sample set X；The cluster sample This collection X is expressed as X={ x1,…,xn, ε=1,2 ... n；
The sub- determining unit of clusters number, for the risk class according to the freeway tunnel traffic accident for obtaining, Determine clusters number k；
Arest neighbors Affinity diagram builds subelement, is k's for according to the cluster sample set X, building the clusters number Arest neighbors Affinity diagram W；
Normalizer unit, for calculating Laplacian matrix Ls and standardizing；The Laplacian matrix Ls are described The Laplacian matrixes of arest neighbors Affinity diagram, are defined as L=D-W；The standardized Laplacian matrix Ls are matrix Lsym, The matrix LsymRepresent Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Matrix U constructs subelement, for calculating the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…, ukAs row, structural matrix U；The matrix U represents U ∈ Rn×k；
Matrix P obtains subelement, for every a line of standardization matrix U, obtains matrix P；The matrix P represents P ∈Rn×k；
Clustering cluster obtains subelement, by each row vector y of the matrix Pε∈RkAs a data point, using k- Means algorithms are to yεClustered, obtained multiple clustering clusters；Wherein, the ε row data y of the matrix PεClassification and original number Strong point xεClassification it is identical.
Optionally, the risk evaluation model sets up module, specifically includes：
Training sample set determining unit, for according to the relation chain, determining training sample set；
Risk evaluation model sets up unit, for according to the training sample set, using extreme learning machine algorithm, setting up wind Dangerous assessment models；The risk evaluation model is：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent The ε output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent λ partially Put.
Optionally, the risk evaluation model sets up unit, specifically includes：
Input weights αλWith biasing bλGeneration subelement, for generating input weights α at random using extreme learning machine algorithmλ With biasing bλ, λ=1,2 ..., N；
Hidden layer output matrix H obtains subelement, for according to the training sample set, the input power Value αλAnd the biasing bλ, obtain hidden layer output matrix H；The hidden layer output matrix H tablesWherein, g (x) represents activation primitive, and N represents hidden layer Nodes；
Output weight vector β obtains subelement, according to the hidden layer output matrix H, obtains output weight vector β；Institute State output weight vector β and be expressed as β=H-1T；Wherein, t=[t1,t2,…,tn] be each training sample set expectation wind Dangerous grade, T is represented
Risk evaluation model sets up subelement, for according to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, set up risk evaluation model.
According to the specific embodiment that the present invention is provided, the invention discloses following technique effect：
The invention discloses a kind of street accidents risks appraisal procedure and system, the method and system are public at a high speed by obtaining Road tunnel traffic accident risk factors and freeway tunnel street accidents risks grade, using spectral clustering, calculate and determine Relation chain between risk factors and risk class；And according to relation chain and limit learning algorithm, build risk evaluation model；To appoint Meaning risk facior data is input to risk evaluation model, it is possible to obtain the ultimate risk grade corresponding to the risk factors.Cause This, the method or system provided using the present invention effectively improve freeway tunnel street accidents risks assessment efficiency, make at a high speed Highway tunnel traffic accident risk evaluation process is more efficient, convenient, accurately and effectively realizes the freeway tunnel operation phase Safety evaluation.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to institute in embodiment The accompanying drawing for needing to use is briefly described, it should be apparent that, drawings in the following description are only some implementations of the invention Example, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these accompanying drawings Obtain other accompanying drawings.
Fig. 1 is the street accidents risks appraisal procedure flow chart of the embodiment of the present invention；
Fig. 2 is the street accidents risks assessment system structure chart of the embodiment of the present invention；
Fig. 3 is the confusion matrix of the risk evaluation model of the embodiment of the present invention one training；
Fig. 4 is the confusion matrix of the risk evaluation model of the embodiment of the present invention one checking；
Fig. 5 is the confusion matrix of the risk evaluation model of the embodiment of the present invention one test.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made Embodiment, belongs to the scope of protection of the invention.
Accuracy and high efficiency the invention aims to improve the assessment of freeway tunnel street accidents risks, it is accurate Truly have effect to realize that freeway tunnel runs the safety evaluation of phase, and propose a kind of spectral clustering and extreme learning machine algorithm The freeway tunnel street accidents risks appraisal procedure and system being combined.First, the limit is obtained by Spectral Clustering Necessary prior information needed for habit machine algorithm, i.e., the relation chain between accurate risk factors and risk class；Then, by pole Limit learning machine method carries out risk assessment, is prevented effectively from traditional methods of risk assessment, and qualitative analysis and subjective analysis are excessive etc. Problem, effectively improves freeway tunnel street accidents risks assessment efficiency, assesses freeway tunnel street accidents risks Process is more efficient, convenient, accurately and effectively realizes the safety evaluation that freeway tunnel runs the phase.
It is below in conjunction with the accompanying drawings and specific real to enable the above objects, features and advantages of the present invention more obvious understandable The present invention is further detailed explanation to apply mode.
Fig. 1 is the street accidents risks appraisal procedure flow chart of the embodiment of the present invention, as shown in figure 1, methods described includes：
Step 101：Obtain the risk factors of freeway tunnel traffic accident；
Wherein, the risk factors of freeway tunnel traffic accident include：Length of tunnel, the magnitude of traffic flow, cart ratio are handed over Logical saturation degree, travel speed, radius of horizontal curve.Present invention focuses on the method for risk assessment, influence factor is in different tunnels Can be appropriate in road environment increase or decrease.The present invention only provides some effects factor as an example, to illustrate that the present invention is carried The methods of risk assessment for going out.
Step 102：Obtain the risk class of freeway tunnel traffic accident；
Wherein, the loss of freeway tunnel street accidents risks is including casualties, economic loss, environmental loss etc..Root Lost according to street accidents risks, be four grades by street accidents risks grade classification：Very safe, safer, relatively hazardous, danger Danger, and assign risk assignment, respectively 1,2,3,4 by street accidents risks grade.
Step 103：According to the risk factors, the risk class corresponding to each described risk factors is determined；
The step 103 is specifically included：According to the characteristics of each risk factors, each risk factors and traffic thing are determined Therefore the corresponding relation of risk class.
Step 104：Risk class according to corresponding to the risk factors and the risk factors, determine the risk because Relation chain between plain and described risk class；
The step 104 includes：By spectral clustering, multigroup risk factors are clustered, obtain multiple clustering clusters； The spectral clustering is all data point sets to be divided into multiple set according to different demarcation criterion (intersection of sets collection is two-by-two It is empty), it is minimum by meeting the side right summation between set, and drawing for classification is realized in side right summation in gathering larger requirement The method divided.
Risk class according to corresponding to the risk factors, determine single described risk in each described clustering cluster because Risk class corresponding to element；And the grade of the single risk factors in each clustering cluster, give each clustering cluster most Whole risk class, and carry out assignment, and then determine the corresponding ultimate risk grade of each described clustering cluster；The assignment method with Step 102 is identical；
According to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, risk factors and described are determined Relation chain between risk class.
Wherein, by spectral clustering, multigroup risk factors are clustered, obtains multiple clustering clusters, specifically included：
Spectral clustering includes two different phases：1) Affinity diagram is constructed using sample data set；2) by Affinity diagram Optimum segmentation, cluster numbers strong point.
Affinity diagram is a non-directed graph G (V, E, W), wherein V=[v1,…,vn] summit is represented, E represents side, and W is corresponding Affine matrix.ei,jIt is vertex viWith vjBetween side, corresponding nonnegative curvature is wi,j, represent sample xiAnd xjBetween parent And degree.Therefore, Affinity diagram can use affine matrix W=[wi,j] represent.W is solved by Gauss similarity functioni,jFor：
Wherein, the σ in formula (2)2It is the variance of all samples.The Laplacian matrixes of Affinity diagram are defined as L=D-W, square The characteristic vector of battle array L is closely related with cluster.Wherein, D is diagonal matrix, diagonal entry
The present invention is calculated using NJW (Ng-Jordan-Weiss) spectral clustering of the arest neighbors Affinity diagram based on clusters number k Method, specially：According to the risk factors, it is determined that cluster sample set X；The cluster sample set X is expressed as X={ x1,…,xn, ε=1,2 ... n；
According to the risk class of the freeway tunnel traffic accident for obtaining, clusters number k is determined；
According to the cluster sample set X, it is the arest neighbors Affinity diagram W of k to build the clusters number；
Calculate Laplacian matrix Ls and standardize；The standardized Laplacian matrix Ls are matrix Lsym, the square Battle array LsymRepresent Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Calculate the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…,ukAs row, structural matrix U；Institute State matrix U and represent U ∈ Rn×k；
According toEvery a line of standardization matrix U, obtains matrix P；The matrix P Represent P ∈ Rn×k；
By each row vector y of the matrix Pε∈RkAs a data point, using k-means algorithms to yεGathered Class, obtains 4 clustering clusters；Wherein, the ε row data y of the matrix PεClassification and raw data points xεClassification it is identical.
Step 105：According to the relation chain, risk evaluation model is set up；
The step 105 includes：Understood for risk facior data to be divided into four classes according to step 104, according in each class The feature of single risk factors, determines the risk class and assignment of each class, is denoted as t.It can thus be concluded that risk factors and risk etc. The multi-group data of level relation chain, i.e. risk factors and ultimate risk grade corresponding relation.And according to risk factors and ultimate risk The multi-group data construction training sample set of grade corresponding relation, using extreme learning machine algorithm, sets up risk evaluation model.Specifically For：
According to the relation chain, training sample set is determined；The input of training sample set is risk facior data collection X= {x1,…,xn, training set is output as risk class t；
According to the training sample set, using extreme learning machine algorithm, risk evaluation model is set up；The risk assessment mould Type is：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent The ε output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent λ partially Put.
Wherein, according to the training sample set, using extreme learning machine algorithm, risk evaluation model is set up, is specifically included：
Input weights α is generated using extreme learning machine algorithm at randomλWith biasing bλ, λ=1,2 ..., N；
According to the training sample set, the input weights αλAnd the biasing bλ, obtain hidden layer output matrix H；Institute State hidden layer output matrix H tablesWherein, g (x) is represented and swashed Function living, N represents node in hidden layer；The i-th of the slice of matrix of the hidden layer output matrix H is classified as corresponding to W input sample The output vector of the implicit node of this i-th.Feedforward neural network for any one tight input sample collection, when input weights During with the random setting of biasing and network activation function non-zero continuously differentiable, then network can approach arbitrary continuous function.This The input weights and biasing for just illustrating feedforward neural network can set at random, and in the training process without or else breaking iteration more Newly.
According to the hidden layer output matrix H, output weight vector β is obtained；The output weight vector β be expressed as β= H-1T；Wherein, t=[t1,t2,…,tn] be each training sample set expected risk grade, T represents
According to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, set up Risk evaluation model.
Step 106：According to the risk evaluation model, the street accidents risks coefficient of freeway tunnel is assessed.
By method provided in an embodiment of the present invention, any risk facior data is input to risk evaluation model, so that it may To obtain the ultimate risk grade corresponding to the risk factors, and then determine the risk factor corresponding to the risk factors.Therefore, By method provided in an embodiment of the present invention, freeway tunnel street accidents risks assessment efficiency is effectively improved, made public at a high speed Road tunnel traffic accident risk assessment processes are more efficient, convenient, accurately and effectively realize the freeway tunnel operation phase Safety evaluation.
To achieve the above object, present invention also offers a kind of street accidents risks assessment system.
Fig. 2 is the street accidents risks assessment system structure chart of the embodiment of the present invention, as shown in Fig. 2 the system includes：
Risk factors acquisition module 201, the risk factors for obtaining freeway tunnel traffic accident；
Risk class acquisition module 202, the risk class for obtaining freeway tunnel traffic accident；
Risk factors correspondence risk class determining module 203, for according to the risk factors, determining each described risk Risk class corresponding to factor；
Relation chain determining module 204, for the risk class according to corresponding to the risk factors and the risk factors, Determine the relation chain between the risk factors and the risk class；
The relation chain determining module 204, specifically includes：
Clustering cluster obtains unit, for by spectral clustering, multigroup risk factors being clustered, obtains multiple clusters Cluster；
Clustering cluster correspondence ultimate risk level de-termination unit, for the risk class according to corresponding to the risk factors, Determine the single described risk factors correspondence risk class in each described clustering cluster, and then determine each described clustering cluster correspondence Ultimate risk grade；
Relation chain determining unit, for according to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, really Fixed relation chain between the risk factors and the risk class.
Wherein, the clustering cluster obtains unit, specifically includes：
Cluster sample collected works determining unit, for according to the risk factors, it is determined that cluster sample set X；The cluster sample This collection X is expressed as X={ x1,…,xn, ε=1,2 ... n；
The sub- determining unit of clusters number, for the risk class according to the freeway tunnel traffic accident for obtaining, Determine clusters number k；
Arest neighbors Affinity diagram builds subelement, is k's for according to the cluster sample set X, building the clusters number Arest neighbors Affinity diagram W；
Normalizer unit, for calculating Laplacian matrix Ls and standardizing；The Laplacian matrix Ls are described The Laplacian matrixes of arest neighbors Affinity diagram, are defined as L=D-W；The standardized Laplacian matrix Ls are matrix Lsym, The matrix LsymRepresent Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Matrix U constructs subelement, for calculating the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…, ukAs row, structural matrix U；The matrix U represents U ∈ Rn×k；
Matrix P obtains subelement, for every a line of standardization matrix U, obtains matrix P；The matrix P represents P ∈Rn×k；
Clustering cluster obtains subelement, by each row vector y of the matrix Pε∈RkAs a data point, using k- Means algorithms are to yεClustered, obtained multiple clustering clusters；Wherein, the ε row data y of the matrix PεClassification and original number Strong point xεClassification it is identical.
Risk evaluation model sets up module 205, for according to the relation chain, setting up risk evaluation model；
The risk evaluation model sets up module 205, specifically includes：
Training sample set determining unit, for according to the relation chain, determining training sample set；
Risk evaluation model sets up unit, for according to the training sample set, using extreme learning machine algorithm, setting up wind Dangerous assessment models；The risk evaluation model is：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent The ε output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent λ partially Put.
Wherein, the risk evaluation model sets up unit, specifically includes：
Input weights αλWith biasing bλGeneration subelement, for generating input weights α at random using extreme learning machine algorithmλ With biasing bλ, λ=1,2 ..., N；
Hidden layer output matrix H obtains subelement, for according to the training sample set, the input power Value αλAnd the biasing bλ, obtain hidden layer output matrix H；The hidden layer output matrix H tablesWherein, g (x) represents activation primitive, and N represents hidden layer Nodes；
Output weight vector β obtains subelement, according to the hidden layer output matrix H, obtains output weight vector β；Institute State output weight vector β and be expressed as β=H-1T；Wherein, t=[t1,t2,…,tn] be each training sample set expectation wind Dangerous grade, T is represented
Risk evaluation model sets up subelement, for according to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, set up risk evaluation model.
Evaluation module 206, for according to the risk evaluation model, assessing the street accidents risks system of freeway tunnel Number.
A kind of street accidents risks assessment system provided in an embodiment of the present invention, is prevented effectively from traditional methods of risk assessment In, the problems such as qualitative analysis and excessive subjective analysis, freeway tunnel street accidents risks assessment efficiency is effectively improved, make height Fast highway tunnel traffic accident risk evaluation process is more efficient, convenient, accurately and effectively realizes freeway tunnel operation The safety evaluation of phase.
In order to verify a kind of street accidents risks appraisal procedure and system of present invention offer, it is possible to increase highway tunnel The accuracy and high efficiency of road street accidents risks assessment, accurate and effective realize that freeway tunnel runs the safety evaluation of phase, Illustrated the invention provides a specific embodiment.
Embodiment one
The present embodiment carries out example explanation and checking mainly for appraisal procedure proposed by the invention.
Step one, chooses 6 kinds of factors of influence freeway tunnel street accidents risks, various risk factors and traffic thing Therefore the relation between risk class, it is as shown in table 1 below.
Relation between the risk factors of table 1 and street accidents risks grade
The present embodiment chooses national freeway tunnel and amounts to 100, carries out the collection of related data, obtains 100 groups of friendships Interpreter's event risk facior data.With these data as foundation, algorithm specific embodiment is illustrated.
Step 2, freeway tunnel risk class are divided
Risk class is divided into four hierarchically secures, safer, relatively hazardous, danger.Carry out being entered as 1 respectively, 2,3, 4。
Step 3, risk factors and risk class relation chain determine
By spectral clustering, 100 groups of street accidents risks factor datas are divided into four classes, four kinds of accident wind are corresponded to respectively Dangerous grade.According in each class data, the characteristics of single risk factors, it is determined that final risk class, has thus obtained wind The relation chain of dangerous factor and risk class.The corresponding relation data as shown in table 2 below of risk factors and risk class
The corresponding relation data of the risk factors of table 2 and risk class
Step 4, set up risk evaluation model
1st, node in hidden layer N is determined, random generation input weights α and biasing b；Wherein, the number of hidden nodes N takes 10
2nd, hidden layer output matrix H is calculated；
3rd, the least squares norm solution of output matrix β is solved, i.e. the least squares norm solution β of H β=T calculates output weights Vectorial β, β=H-1T；Wherein
Wherein, in step 4 the 1st, 2, the computational methods of 3 steps discussed in the above, do not explain in detail one by one herein.
4th, according to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, build Vertical risk evaluation model.
From the relation sample of 100 groups of risk factors and risk class, 70 groups are randomly selected as risk evaluation model Training sample set, 15 groups collect as checking.Risk evaluation model trains the confusion matrix with the result respectively such as Fig. 3 and Fig. 4 Shown, confusion matrix last cell grid represents the overall accuracy (above) and error rate (below) of risk assessment.By Fig. 3-4 It can be seen that, the training accuracy of risk evaluation model is 100%, and checking accuracy is 93.3%.
The street accidents risks coefficient of step 5, assessment freeway tunnel
Risk evaluation model is used to assess the street accidents risks coefficient of freeway tunnel, the risk evaluation model to be defeated It is various risk factors to enter, and risk evaluation model output is risk class, then according to final risk class, it is determined that should The street accidents risks coefficient of freeway tunnel.The present embodiment uses remaining 15 groups of risk factors and the relation of risk class Sample is used as test set, and the confusion matrix of test result is as shown in Figure 5.As seen from Figure 5, the accuracy of street accidents risks assessment It is 93.3%.
Can be drawn by the present embodiment, a kind of street accidents risks appraisal procedure and system that the present invention is provided, Neng Gouti The accuracy and high efficiency of freeway tunnel street accidents risks assessment high, accurate and effective realize that freeway tunnel runs the phase Safety evaluation.
Each embodiment is described by the way of progressive in this specification, and what each embodiment was stressed is and other The difference of embodiment, between each embodiment identical similar portion mutually referring to.For system disclosed in embodiment For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part It is bright.
Specific case used herein is set forth to principle of the invention and implementation method, and above example is said It is bright to be only intended to help and understand the method for the present invention and its core concept；Simultaneously for those of ordinary skill in the art, foundation Thought of the invention, will change in specific embodiments and applications.In sum, this specification content is not It is interpreted as limitation of the present invention.

Claims (10)

1. a kind of street accidents risks appraisal procedure, it is characterised in that methods described includes：
Obtain the risk factors of freeway tunnel traffic accident；
Obtain the risk class of freeway tunnel traffic accident；
According to the risk factors, the risk class corresponding to each described risk factors is determined；
Risk class according to corresponding to the risk factors and the risk factors, determines the risk factors and the risk Relation chain between grade；
According to the relation chain, risk evaluation model is set up；
According to the risk evaluation model, the street accidents risks coefficient of freeway tunnel is assessed.
2. method according to claim 1, it is characterised in that the determination risk factors and the risk class it Between relation chain, specifically include：
By spectral clustering, multigroup risk factors are clustered, obtain multiple clustering clusters；
Risk class according to corresponding to the risk factors, determines the single described risk factors institute in each described clustering cluster Corresponding risk class, and then determine the corresponding ultimate risk grade of each described clustering cluster；
According to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, the risk factors and the risk are determined Relation chain between grade.
3. method according to claim 2, it is characterised in that described by spectral clustering, by multigroup risk because Element cluster, obtains multiple clustering clusters, specifically includes：
According to the risk factors, it is determined that cluster sample set X；The cluster sample set X is expressed as X={ x1,…,xn, ε=1, 2,…n；
According to the risk class of the freeway tunnel traffic accident for obtaining, clusters number k is determined；
According to the cluster sample set X, it is the arest neighbors Affinity diagram W of k to build the clusters number；
Calculate Laplacian matrix Ls and standardize；The Laplacian matrix Ls are the arest neighbors Affinity diagram Laplacian matrixes, are defined as L=D-W；The standardized Laplacian matrix Ls are matrix Lsym, the matrix LsymTable Show Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Calculate the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…,ukAs row, structural matrix U；The square Battle array U represents U ∈ Rn×k；
Every a line of standardization matrix U, obtains matrix P；The matrix P represents P ∈ Rn×k；
By each row vector y of the matrix Pε∈RkAs a data point, using k-means algorithms to yεClustered, obtained To multiple clustering clusters；Wherein, the ε row data y of the matrix PεClassification and raw data points xεClassification it is identical.
4. method according to claim 1, it is characterised in that described to set up risk evaluation model, specifically includes：
According to the relation chain, training sample set is determined；
According to the training sample set, using extreme learning machine algorithm, risk evaluation model is set up；The risk evaluation model For：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent ε The output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent the λ biasing.
5. method according to claim 4, it is characterised in that described according to the training sample set, is learnt using the limit Machine algorithm, sets up risk evaluation model, specifically includes：
Input weights α is generated using extreme learning machine algorithm at randomλWith biasing bλ, λ=1,2 ..., N；
According to the training sample set, the input weights αλAnd the biasing bλ, obtain hidden layer output matrix H；It is described hidden The H tables of output matrix containing layerWherein, g (x) represents activation letter Number, N represents node in hidden layer；
According to the hidden layer output matrix H, output weight vector β is obtained；The output weight vector β is expressed as β=H-1T； Wherein, t=[t1,t2,…,tn] be each training sample set expected risk grade, T represents
According to the training sample set, the input weights αλ, the biasing bλAnd the output weight vector β, set up risk Assessment models.
6. a kind of street accidents risks assessment system, it is characterised in that the system includes：
Risk factors acquisition module, the risk factors for obtaining freeway tunnel traffic accident；
Risk class acquisition module, the risk class for obtaining freeway tunnel traffic accident；
Risk factors correspondence risk class determining module, for according to the risk factors, determining each described risk factors institute Corresponding risk class；
Relation chain determining module, for the risk class according to corresponding to the risk factors and the risk factors, determines institute State the relation chain between risk factors and the risk class；
Risk evaluation model sets up module, for according to the relation chain, setting up risk evaluation model；
Evaluation module, for according to the risk evaluation model, assessing the street accidents risks coefficient of freeway tunnel.
7. system according to claim 6, it is characterised in that the relation chain determining module, specifically includes：
Clustering cluster obtains unit, for by spectral clustering, multigroup risk factors being clustered, obtains multiple clustering clusters；
Clustering cluster correspondence ultimate risk level de-termination unit, for the risk class according to corresponding to the risk factors, it is determined that Single described risk factors correspondence risk class in each described clustering cluster, and then determine that each described clustering cluster is corresponding most Whole risk class；
Relation chain determining unit, for according to the clustering cluster and the corresponding ultimate risk grade of the clustering cluster, determining institute State the relation chain between risk factors and the risk class.
8. system according to claim 7, it is characterised in that the clustering cluster obtains unit, specifically includes：
Cluster sample collected works determining unit, for according to the risk factors, it is determined that cluster sample set X；The cluster sample set X It is expressed as X={ x1,…,xn, ε=1,2 ... n；
The sub- determining unit of clusters number, for the risk class according to the freeway tunnel traffic accident for obtaining, it is determined that Clusters number k；
Arest neighbors Affinity diagram builds subelement, is the nearest of k for according to the cluster sample set X, building the clusters number Adjacent Affinity diagram W；
Normalizer unit, for calculating Laplacian matrix Ls and standardizing；The Laplacian matrix Ls are described nearest The Laplacian matrixes of adjacent Affinity diagram, are defined as L=D-W；The standardized Laplacian matrix Ls are matrix Lsym, it is described Matrix LsymRepresent Lsym=D-1/2LD-1/2；Wherein, D is diagonal matrix；
Matrix U constructs subelement, for calculating the matrix LsymPreceding k characteristic vector u1,…,uk, and by u1,…,ukAs Row, structural matrix U；The matrix U represents U ∈ Rn×k；
Matrix P obtains subelement, for every a line of standardization matrix U, obtains matrix P；The matrix P represents P ∈ Rn ×k；
Clustering cluster obtains subelement, by each row vector y of the matrix Pε∈RkAs a data point, calculated using k-means Method is to yεClustered, obtained multiple clustering clusters；Wherein, the ε row data y of the matrix PεClassification and raw data points xε Classification it is identical.
9. system according to claim 6, it is characterised in that the risk evaluation model sets up module, specifically includes：
Training sample set determining unit, for according to the relation chain, determining training sample set；
Risk evaluation model sets up unit, for according to the training sample set, using extreme learning machine algorithm, sets up risk and comments Estimate model；The risk evaluation model is：
In formula (1), g (x) represents activation primitive；N is node in hidden layer；xεRepresent the ε input of sample；tεRepresent ε The output of sample；αλRepresent the λ input weight vector；βλRepresent the λ output weight vector；bλRepresent the λ biasing.
10. system according to claim 9, it is characterised in that the risk evaluation model sets up unit, specifically includes：
Input weights αλWith biasing bλGeneration subelement, for generating input weights α at random using extreme learning machine algorithmλWith it is inclined Put bλ, λ=1,2 ..., N；
Hidden layer output matrix H obtains subelement, for according to the training sample set, the input weights αλ And the biasing bλ, obtain hidden layer output matrix H；The hidden layer output matrix H tablesWherein, g (x) represents activation primitive, and N represents hidden layer Nodes；
Output weight vector β obtains subelement, according to the hidden layer output matrix H, obtains output weight vector β；It is described defeated Go out weight vector β and be expressed as β=H-1T；Wherein, t=[t1,t2,…,tn] it is expected risk of each training sample set etc. Level, T is represented
Risk evaluation model sets up subelement, for according to the training sample set, the input weights αλ, the biasing bλWith And the output weight vector β, set up risk evaluation model.
CN201710099426.0A 2017-02-23 2017-02-23 A kind of street accidents risks appraisal procedure and system Pending CN106897826A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710099426.0A CN106897826A (en) 2017-02-23 2017-02-23 A kind of street accidents risks appraisal procedure and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710099426.0A CN106897826A (en) 2017-02-23 2017-02-23 A kind of street accidents risks appraisal procedure and system

Publications (1)

Publication Number Publication Date
CN106897826A true CN106897826A (en) 2017-06-27

Family

ID=59185234

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710099426.0A Pending CN106897826A (en) 2017-02-23 2017-02-23 A kind of street accidents risks appraisal procedure and system

Country Status (1)

Country Link
CN (1) CN106897826A (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107526117A (en) * 2017-07-06 2017-12-29 天津科技大学 SVEL Forecasting Methodology based on autocoding and the study joint network that transfinites
CN108510185A (en) * 2018-03-29 2018-09-07 北京紫晶立方科技有限公司 The quick hazard analysis of road vehicle and methods of risk assessment
CN108629515A (en) * 2018-05-07 2018-10-09 招商局重庆交通科研设计院有限公司 A kind of roads and tunnels traffic risk real-time evaluation system
CN109118773A (en) * 2018-09-30 2019-01-01 中交第公路勘察设计研究院有限公司 A kind of traffic accidents methods of risk assessment
CN109636125A (en) * 2018-11-19 2019-04-16 北京交通大学 A kind of traffic accidents risk evaluating system
CN109671274A (en) * 2019-01-24 2019-04-23 交通运输部公路科学研究所 A kind of highway risk automatic evaluation method based on latent structure and fusion
CN110084291A (en) * 2019-04-12 2019-08-02 湖北工业大学 A kind of students ' behavior analysis method and device based on the study of the big data limit
CN110147874A (en) * 2019-04-28 2019-08-20 武汉理工大学 A kind of horizontal intelligent optimization method of environmental factor of major long tunnel speed distribution
CN110858334A (en) * 2018-08-13 2020-03-03 北京中科蓝图科技有限公司 Road safety assessment method and device and road safety early warning system
CN111126868A (en) * 2019-12-30 2020-05-08 中南大学 Road traffic accident occurrence risk determination method and system
CN112101759A (en) * 2020-09-03 2020-12-18 交通运输部科学研究院 Method and device for constructing risk assessment model and assessing risk of expressway tunnel
CN112308434A (en) * 2020-11-03 2021-02-02 中国交通通信信息中心 Traffic safety risk assessment method and system
CN112561175A (en) * 2020-12-18 2021-03-26 深圳赛安特技术服务有限公司 Traffic accident influence factor prediction method, device, equipment and storage medium
CN113345224A (en) * 2021-05-20 2021-09-03 河南硕翔信息技术有限公司 Highway data high efficiency storage system based on computer vision
CN113516819A (en) * 2021-05-25 2021-10-19 北京创源信诚管理体系认证有限公司 Intelligent electric fire early warning system and method for tobacco warehouse
CN114331019A (en) * 2021-12-01 2022-04-12 武汉长江通信智联技术有限公司 Urban traffic safety risk real-time assessment method and device based on risk factor
CN115985137A (en) * 2023-03-10 2023-04-18 深圳市城市交通规划设计研究中心股份有限公司 Risk assessment method for tunnel traffic operation, electronic equipment and storage medium

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107526117A (en) * 2017-07-06 2017-12-29 天津科技大学 SVEL Forecasting Methodology based on autocoding and the study joint network that transfinites
CN107526117B (en) * 2017-07-06 2019-08-13 天津科技大学 Based on autocoding and transfinites and learn the acoustic speed prediction technique of joint network
CN108510185A (en) * 2018-03-29 2018-09-07 北京紫晶立方科技有限公司 The quick hazard analysis of road vehicle and methods of risk assessment
CN108510185B (en) * 2018-03-29 2020-10-27 北京紫晶立方科技有限公司 Rapid hazard analysis and risk assessment method for road vehicles
CN108629515A (en) * 2018-05-07 2018-10-09 招商局重庆交通科研设计院有限公司 A kind of roads and tunnels traffic risk real-time evaluation system
CN110858334A (en) * 2018-08-13 2020-03-03 北京中科蓝图科技有限公司 Road safety assessment method and device and road safety early warning system
CN109118773A (en) * 2018-09-30 2019-01-01 中交第公路勘察设计研究院有限公司 A kind of traffic accidents methods of risk assessment
CN109118773B (en) * 2018-09-30 2019-10-01 中交第一公路勘察设计研究院有限公司 A kind of traffic accidents methods of risk assessment
CN109636125A (en) * 2018-11-19 2019-04-16 北京交通大学 A kind of traffic accidents risk evaluating system
CN109671274B (en) * 2019-01-24 2020-09-25 交通运输部公路科学研究所 Highway risk automatic evaluation method based on feature construction and fusion
CN109671274A (en) * 2019-01-24 2019-04-23 交通运输部公路科学研究所 A kind of highway risk automatic evaluation method based on latent structure and fusion
CN110084291A (en) * 2019-04-12 2019-08-02 湖北工业大学 A kind of students ' behavior analysis method and device based on the study of the big data limit
CN110147874A (en) * 2019-04-28 2019-08-20 武汉理工大学 A kind of horizontal intelligent optimization method of environmental factor of major long tunnel speed distribution
CN110147874B (en) * 2019-04-28 2022-12-16 武汉理工大学 Intelligent optimization method for environmental factor level of long and large tunnel vehicle speed distribution
CN111126868A (en) * 2019-12-30 2020-05-08 中南大学 Road traffic accident occurrence risk determination method and system
CN111126868B (en) * 2019-12-30 2023-07-04 中南大学 Road traffic accident occurrence risk determination method and system
CN112101759A (en) * 2020-09-03 2020-12-18 交通运输部科学研究院 Method and device for constructing risk assessment model and assessing risk of expressway tunnel
CN112308434A (en) * 2020-11-03 2021-02-02 中国交通通信信息中心 Traffic safety risk assessment method and system
CN112561175A (en) * 2020-12-18 2021-03-26 深圳赛安特技术服务有限公司 Traffic accident influence factor prediction method, device, equipment and storage medium
CN113345224A (en) * 2021-05-20 2021-09-03 河南硕翔信息技术有限公司 Highway data high efficiency storage system based on computer vision
CN113345224B (en) * 2021-05-20 2023-01-10 中咨数据有限公司 Highway data high efficiency storage system based on computer vision
CN113516819A (en) * 2021-05-25 2021-10-19 北京创源信诚管理体系认证有限公司 Intelligent electric fire early warning system and method for tobacco warehouse
CN114331019A (en) * 2021-12-01 2022-04-12 武汉长江通信智联技术有限公司 Urban traffic safety risk real-time assessment method and device based on risk factor
CN115985137A (en) * 2023-03-10 2023-04-18 深圳市城市交通规划设计研究中心股份有限公司 Risk assessment method for tunnel traffic operation, electronic equipment and storage medium

Similar Documents

Publication Publication Date Title
CN106897826A (en) A kind of street accidents risks appraisal procedure and system
CN102663264B (en) Semi-supervised synergistic evaluation method for static parameter of health monitoring of bridge structure
CN110348713A (en) A kind of platform area line loss calculation method based on association analysis and data mining
CN107248283A (en) A kind of urban area road network evaluation of running status method of consideration section criticality
CN112116160B (en) Important power transmission channel disaster monitoring method based on improved cellular automaton of optimized neural network
CN109063247A (en) Landslide disaster forecasting procedure based on deepness belief network
CN111582580B (en) Travel path planning method considering population pollutant exposure
CN102867217A (en) Projection pursuit-based risk evaluation method for meteorological disasters of facility agriculture
CN104809877A (en) Expressway site traffic state estimation method based on feature parameter weighted GEFCM algorithm
CN109948839B (en) Method and system for predicting and early warning galloping risk of overhead transmission line
CN104112062B (en) The acquisition methods of wind-resources distribution based on interpolation method
CN112966871A (en) Traffic jam prediction method and system based on convolution long-short term memory neural network
CN110110425A (en) A kind of slope sliding force prediction method based on reliability rule-based reasoning
CN114299742B (en) Speed limit information dynamic identification and update recommendation method for expressway
CN112446114A (en) Power transmission line engineering construction progress monitoring method based on three-dimensional model comparison
CN113689004A (en) Underground pipe network bearing capacity evaluation method and system based on machine learning
CN111126868A (en) Road traffic accident occurrence risk determination method and system
CN112149922A (en) Method for predicting severity of accident in exit and entrance area of down-link of highway tunnel
Xu et al. Traffic accident hotspots identification based on clustering ensemble model
CN106384507A (en) Travel time real-time estimation method based on sparse detector
CN111806443A (en) Pure electric heavy truck unmanned energy consumption optimization method
CN105956318B (en) Based on the wind power plant group of planes division methods for improving division H-K clustering method
CN107066501A (en) A kind of telemetering motor vehicle tail equipment points distributing method based on road similitude
CN112686404B (en) Power distribution network fault first-aid repair-based collaborative optimization method
CN112434887B (en) Water supply network risk prediction method combining network kernel density estimation and SVM

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20170627

RJ01 Rejection of invention patent application after publication