CN109242209A - Railway emergency event grading forewarning system method based on K-means cluster - Google Patents
Railway emergency event grading forewarning system method based on K-means cluster Download PDFInfo
- Publication number
- CN109242209A CN109242209A CN201811191236.2A CN201811191236A CN109242209A CN 109242209 A CN109242209 A CN 109242209A CN 201811191236 A CN201811191236 A CN 201811191236A CN 109242209 A CN109242209 A CN 109242209A
- Authority
- CN
- China
- Prior art keywords
- emergency event
- grade
- data
- feature
- cluster
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 21
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 238000013480 data collection Methods 0.000 claims description 4
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 abstract description 4
- 238000007418 data mining Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000034994 death Effects 0.000 description 1
- 231100000517 death Toxicity 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The present invention provides a kind of railway emergency event grading forewarning system methods based on K-means cluster.This method comprises: influencing data to history railway emergency event carries out signature analysis and data cleansing, training dataset is obtained;It determines cluster centre number, the training dataset is clustered based on K-means algorithm, the sample data feature of railway emergency events at different levels is obtained according to cluster result;The data characteristics of current rail emergency event is compared with the sample data feature of the railway emergency events at different levels, the emergency event grade of the current rail emergency event is determined according to comparison result.Method of the invention uses data mining machine Learning Theory, avoids subjective decision, influences dynamic quantitative assessment classification to railway emergency event.The experimental results showed that this method, which rationally can effectively solve emergency event, influences grading forewarning system, practicability is good.
Description
Technical field
The present invention relates to technical field more particularly to a kind of railway emergency event grading forewarning systems based on K-means cluster
Method.
Background technique
Safety of railway operation is increasingly paid close attention to, and railway emergency event, which influences grading forewarning system, has extremely safety of railway operation
Close important role.It is according to railroad embankment break period, personnel that it is mostly, which to influence grading forewarning system research, for railway emergency event at this stage
The features such as injures and deaths and property loss influence after-action review divided rank to railway emergency event using analytic hierarchy process (AHP), divide power
Repeated root according to expertise determine, have stronger subjectivity, and assess event influence granularity it is larger, mainly for railway operation
The major event seriously affected is caused, being not suitable for the public now, on schedule, comfortably service need to railway especially high-speed railway
It asks.
K-means clustering algorithm is a kind of unsupervised learning Data Clustering Algorithm, is used for the field of data mining.K-means
Similarity measurement is to data grouping between clustering algorithm is based on sample, so that sample similarity is maximum in organizing, between group, sample difference is most
Greatly, the purpose of similar sample clustering is realized.K-means clustering algorithm has many advantages, such as that structure is simple, speed is fast.
The railway emergency event grading forewarning system method that there are no a kind of effectively based on K-means cluster in the prior art.
Summary of the invention
The embodiment provides it is a kind of based on K-means cluster railway emergency event grading forewarning system method, with
The shortcomings that overcoming the prior art.
To achieve the goals above, this invention takes following technical solutions.
A kind of railway emergency event grading forewarning system method based on K-means cluster, comprising:
Data are influenced on history railway emergency event and carry out signature analysis and data cleansing, obtain training dataset;
It determines cluster centre number, the training dataset is clustered based on K-means algorithm, is obtained according to cluster result
Take the sample data feature of railway emergency events at different levels;
The data characteristics of current rail emergency event and the sample data feature of the railway emergency events at different levels are carried out
Compare, the emergency event grade of the current rail emergency event is determined according to comparison result.
Further, described that data progress signature analysis and data cleansing are influenced on history railway emergency event, it obtains
Training dataset, comprising:
By influencing data analysis to history railway emergency event, from railroad train operating status angle, it is prominent to extract railway
The influence feature X={ x of hair event1,x2,L,xd, the influence feature vector of real number is tieed up at d according to influence feature X-shaped, wherein xi
The value of feature is influenced for i-th dimension;
According to the influence feature vector, arranging the emergency event of history railway influences to form record set, carries out to the record set
Data cleansing handles the record sample that characteristic lacks in the record set, forms feature complete data collectionIts
Middle X 'i={ x 'i,1,x′i,2,L,x′i,dIt is the complete data sample of ith feature, x 'i,dFor the d Wei Te of i-th of data sample
Value indicative, N are characterized complete data number of samples;
One-hot coding is carried out to the classifying type feature that the feature complete data is concentrated, the feature complete data is concentrated
Numeric type feature carry out minimax normalized, the data feature values x after normalizationi,jAre as follows:
Wherein x 'i,jFor the j dimensional feature value of i-th of data sample before normalized,
Obtain training datasetXi={ xi,1,xi,2,L,xi,d}。
Further, the determination cluster centre number gathers the training dataset based on K-means algorithm
Class obtains the sample data feature of railway emergency events at different levels according to cluster result, comprising:
Step 2.1: use a certain range of trellis search method, calculate separately Calinski-Harabaz Index and
Iteration error determines cluster centre number k;
Step 2.2: k number is arbitrarily chosen from N number of data sample according to sample as initial cluster center
And initialize iterator m=0;
Step 2.3: calculating training data and concentrate the Euclidean distance between each data sample and cluster centre, to maximize
Distance between cluster, minimum intra-cluster distance is target, with φminObjective function divides the classification of each data sample, wherein
wi,jFor 0-1 variable, wi,j=1 indicates sample XiBelong to cluster centre CjSample in cluster, otherwise wi,j=0;
Step 2.4: according to sample data in cluster after update, calculating every cluster cluster centre, update m=m+1 and cluster centre
Step 2.5: calculating cluster variance, and judge to cluster whether variance meets minimum sandards, if it is satisfied, output k
Data sample in cluster centre and each cluster;Otherwise, step 2.3 is continued to execute.
Further, the sample data feature that railway emergency events at different levels are obtained according to cluster result, comprising:
I grade of early warning feature of emergency event is Train delay time 21min or more or train speed limit 250km/h or less;Burst
II grade of early warning feature of event is Train delay time 14~21min or 250~300km/h of train speed limit;III grade of emergency event pre-
Alert feature is Train delay time 9~14min or 300~310km/h of train speed limit;IV grade of early warning feature of emergency event is train
Late 5~9min of time or 310~325km/h of train speed limit;V grade of early warning feature of emergency event is Train delay time 5min
Below or 325~350km/h of train speed limit.
Further, the sample of the data characteristics by current rail emergency event and the railway emergency events at different levels
Notebook data feature is compared, and the emergency event grade of the current rail emergency event is determined according to comparison result, comprising:
The sample data feature vector that I grade, II grade, III grade, IV grade and V grade of the emergency event includes: train number grade,
Position and attribute, event, which occur, for route, event where vehicle, event type, event occur influences section and station quantity, train
Whether late type the late time, train speed limit, late train quantity, stoppage in transit train quantity, influences the duration and enables standby
Use EMU;
By in the railway emergency event currently occurred the Train delay time and train speed limit quantization respectively with above-mentioned burst
The late time in sample data feature vector that I grade, II grade, III grade, IV grade and V grade of event, train speed limit are compared,
According to I grade, II grade, III grade, IV grade and V grade event class criteria for classifying of comparison result combination emergency event, current burst is positioned
Event class realizes railway emergency event grading forewarning system.
As can be seen from the technical scheme provided by the above-mentioned embodiment of the present invention, the method for the embodiment of the present invention is based on data
Machine Learning Theory is excavated, data are influenced according to history railway emergency event and are quantified by data cleansing and characteristic processing
Evaluation event influences feature and standardization sample data;It is trained using K-means cluster, finally obtains railway emergency event
Grade and related corresponding quantitative characteristic are influenced, more accurate grading forewarning system, verification result table can be influenced on emergency event
Bright the method has very high practical value.
The additional aspect of the present invention and advantage will be set forth in part in the description, these will become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill of field, without creative efforts, it can also be obtained according to these attached drawings others
Attached drawing.
Fig. 1 is a kind of railway emergency event grading forewarning system method based on K-means cluster provided in an embodiment of the present invention
Flow chart;
Fig. 2 is the hyper parameter K value schematic diagram in a kind of optimization k-means algorithm provided in an embodiment of the present invention;
Fig. 3 is that a kind of cluster centre sample size provided in an embodiment of the present invention counts schematic diagram;
Fig. 4 is that a kind of training set sample clustering two-dimensional space provided in an embodiment of the present invention visualizes schematic diagram;
Fig. 5 be a kind of training set sample provided in an embodiment of the present invention away from cluster centre apart from schematic diagram.
Specific embodiment
Embodiments of the present invention are described below in detail, the example of the embodiment is shown in the accompanying drawings, wherein from beginning
Same or similar element or element with the same or similar functions are indicated to same or similar label eventually.Below by ginseng
The embodiment for examining attached drawing description is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or coupling.Wording used herein
"and/or" includes one or more associated any cells for listing item and all combinations.
Those skilled in the art of the present technique are appreciated that unless otherwise defined, all terms used herein (including technology art
Language and scientific term) there is meaning identical with the general understanding of those of ordinary skill in fields of the present invention.Should also
Understand, those terms such as defined in the general dictionary, which should be understood that, to be had and the meaning in the context of the prior art
The consistent meaning of justice, and unless defined as here, it will not be explained in an idealized or overly formal meaning.
In order to facilitate understanding of embodiments of the present invention, it is done by taking several specific embodiments as an example below in conjunction with attached drawing further
Explanation, and each embodiment does not constitute the restriction to the embodiment of the present invention.
Compared with railway emergency event traditional at present influences grading forewarning system Hierarchy Analysis Method, clustered based on K-means
Railway emergency event grading forewarning system method, from history railway emergency event influence data, from railroad train operating status
Analysis assessment emergency event influences quantitative characteristic, excavates machine Learning Theory using data information, to railway emergency event dynamic
Quantitative classification early warning, avoids subjective decision, especially for no one was injured but as caused by equipment fault or adverse weather
Train delay event has high susceptibility, meets the needs of public is to railway high quality transportation service.Therefore this patent proposes
Railway emergency event influence grading forewarning system method have great practical value and dissemination.
Embodiment one
A kind of processing of railway emergency event grading forewarning system method based on K-means cluster provided in an embodiment of the present invention
Process is as shown in Figure 1, include the following steps:
Step 1 influences data to history railway emergency event and carries out signature analysis and data cleansing, obtains training data
Collection
1.1, by influencing data analysis to history railway emergency event, from railroad train operating status angle, extract railway
The influence feature X={ x of emergency event1,x2,L,xd, according to influence feature X-shaped at the influence feature vector of d dimension real number, wherein
xiThe value of feature is influenced for i-th dimension;
1.2 according to the influence feature vector of the railway emergency event in 1.1, and arranging the emergency event of history railway influences to be formed
Digitized record collection.Data are influenced based on Railway Site history railway emergency event, the influence to every emergency event is according to quarter
The feature vector that emergency event influences is drawn, successively quantifying railway emergency event influences record, and forming railway emergency event influences number
Word record set.Such as: certain railway emergency event influences to include Train delay time 30min, then corresponds to the railway emergency event
Influencing the digitlization Train delay temporal characteristics value in data record is 30min.
The record sample of processing feature shortage of data carries out data cleansing to record set, and forming railway emergency event influences
The complete data set of featureWherein X 'i={ x 'i,1,x′i,2,L,x′i,dIt is the complete data sample of ith feature
This, x 'i,dFor the d dimensional feature value of i-th of data sample, N is characterized complete data number of samples;
1.3 carry out characteristic processing for feature complete data collection, carry out one-hot coding (One- to classifying type feature respectively
Hot Encoding), logarithm type feature carries out minimax normalized, then the data feature values x after normalizingi,jAre as follows:
Wherein x 'i,jFor the j dimensional feature value of i-th of data sample before normalized, training dataset
Xi={ xi,1,xi,2,L,xi,d};
Step 2 determined cluster centre number k, clustered based on K-means algorithm to training dataset, cluster here
Process is to influence data towards railway emergency event, determines that suitable railway emergency event influences classification number based on clustering processing
Amount cuts the method for customizing out and railway emergency event being suitble to influence grading forewarning system.Specific step is as follows:
2.1 use a certain range of trellis search method, calculate separately CH index (Calinski-Harabaz index)
Cluster centre number k is determined with iteration error (inertia-values);
2.2 arbitrarily choose k number according to sample as initial cluster center from N number of data sampleAnd just
Beginningization iterator m=0;
2.3 calculate the Euclidean distance in data set between each sample and cluster centre, minimum to maximize distance between cluster
Change intra-cluster distance is target, divides the classification of each sample, objective function are as follows:
Wherein wi,jFor 0-1 variable, wi,j=1 indicates sample XiBelong to cluster centre CjSample in cluster, otherwise wi,j=0;
2.4, according to sample data in cluster after update, calculate every cluster cluster centre, update m=m+1 and cluster centre
2.5 calculate cluster variance, and judge to cluster whether variance meets minimum sandards, if it is satisfied, in k cluster of output
Data sample in the heart and each cluster;Otherwise, step 2.3 is continued to execute.
Step 3, according to sample data in the cluster centre and each cluster of cluster each in step 2 cluster result, statistical analysis is each
The sample data feature of grade railway emergency event;
Above-mentioned railway emergency event is divided into following grade:
I.e. I grade of early warning feature of emergency event is Train delay time 21min or more or train speed limit 250km/h or less;It is prominent
II grade of early warning feature of hair event is Train delay time 14~21min or 250~300km/h of train speed limit;III grade of emergency event
Early warning feature is Train delay time 9~14min or 300~310km/h of train speed limit;IV grade of early warning feature of emergency event is column
Vehicle late 5~9min of time or 310~325km/h of train speed limit;V grade of early warning feature of emergency event is the Train delay time
325~350km/h of 5min or less or train speed limit.
The sample data feature vector that I grade, II grade, III grade, IV grade and V grade of above-mentioned emergency event includes: train number grade,
Position and attribute, event, which occur, for route, event where vehicle, event type, event occur influences section and station quantity, train
Late type, the late time, train speed limit, late train quantity, stoppage in transit train quantity, influence the duration, whether enable it is standby
With EMU totally 14 features.
Step 4 occurs emergency event for current, extracts the data characteristics of current rail emergency event.Then, will work as
The data characteristics of preceding railway emergency event is compared with the sample data feature of the railway emergency events at different levels, according to comparing
As a result the emergency event grade of the current rail emergency event is determined.
By in the railway emergency event currently occurred the Train delay time and train speed limit quantization respectively with above-mentioned burst
The late time in sample data feature vector that I grade, II grade, III grade, IV grade and V grade of event, train speed limit are compared,
According to I grade, II grade, III grade, IV grade and V grade event class criteria for classifying of comparison result combination emergency event, current burst is positioned
Event class realizes railway emergency event grading forewarning system.Such as: certain event influences: Train delay time 3min, train speed limit
330km/h, then the event is in V grade.
Embodiment two:
The emergency event of analysis of history railway influences data first, and from railroad train operating status angle, Extraction and determination is portrayed
The influence feature of railway emergency event, the influence feature include train number grade, vehicle, event type, event occur where route,
Position and attribute occur for event, event influences section and station quantity, Train delay type, the late time, train speed limit, late
Train quantity, influences the duration, whether enables spare EMU totally 14 feature (steps 1.1) stoppage in transit train quantity, to upper
Data are influenced in November, 2016 history railway emergency event extra large Railway Bureau in January, 2015 carry out data cleansing and characteristic processing,
Acquisition standard emergency event influences data set 167 (step 1.2,1.3).
Fig. 2 is the hyper parameter K value schematic diagram in a kind of optimization k-means algorithm provided in an embodiment of the present invention, and Fig. 3 is poly-
Class central sample quantity statistics schematic diagram, Fig. 4 are that training set sample clustering two-dimensional space visualizes schematic diagram, and Fig. 5 is training set
Sample is away from cluster centre apart from schematic diagram.Data set is influenced based on standard emergency event, using K-means algorithm to data set
Cluster.CH (Calinski- is calculated separately using a certain range of trellis search method for determination preferable k value
Harabaz) wherein CH refers to that target value is bigger for index and iteration error (inertia-values) (step 2.1), indicates cluster effect
Fruit is better;Inertia-values value is smaller, indicates that Clustering Effect is better.
Based on K-means algorithm to cluster data, sample in cluster centre cluster is obtained, sample characteristics in cluster is counted, obtains
Grade and individual features are influenced to emergency event.For example, influencing data based on railway emergency event, determine k=5 (such as Fig. 2 institute
Show).It is clustered using k-means algorithm, obtains in each cluster centre cluster sample size (as shown in Figure 3), sample data in two dimension
The distance (as shown in Figure 5) of spatial visualization (as shown in Figure 4) and sample away from cluster centre.Railway emergency event influence is clustered
Divide 5 grades, i.e. I grade of early warning feature of emergency event be Train delay time 21min or more or train speed limit 250km/h with
Under;II grade of early warning feature of emergency event is Train delay time 14~21min or 250~300km/h of train speed limit;Emergency event
III grade of early warning feature is Train delay time 9~14min or 300~310km/h of train speed limit;IV grade of early warning feature of emergency event
For 310~325km/h of 5~9min of Train delay time or train speed limit;When V grade of early warning feature of emergency event is Train delay
Between 325~350km/h of 5min or less or train speed limit.
In conclusion the method for the embodiment of the present invention is based on data mining machine Learning Theory, happened suddenly according to history railway
Event influences data, and by data cleansing and characteristic processing, obtaining quantitative assessment event influences feature and standardization sample data;
It is trained using K-means cluster, finally obtaining railway emergency event influences grade and related corresponding quantitative characteristic, can
More accurate grading forewarning system is influenced on emergency event, verification result shows that the method has very high practical value.
The method of the embodiment of the present invention avoids subjective decision, influences dynamic quantitative assessment classification to railway emergency event.It is real
It tests the result shows that this method, which rationally can effectively solve emergency event, influences grading forewarning system, practicability is good.
Those of ordinary skill in the art will appreciate that: attached drawing is the schematic diagram of one embodiment, module in attached drawing or
Process is not necessarily implemented necessary to the present invention.
As seen through the above description of the embodiments, those skilled in the art can be understood that the present invention can
It realizes by means of software and necessary general hardware platform.Based on this understanding, technical solution of the present invention essence
On in other words the part that contributes to existing technology can be embodied in the form of software products, the computer software product
It can store in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) executes the certain of each embodiment or embodiment of the invention
Method described in part.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device or
For system embodiment, since it is substantially similar to the method embodiment, so describing fairly simple, related place is referring to method
The part of embodiment illustrates.Apparatus and system embodiment described above is only schematical, wherein the conduct
The unit of separate part description may or may not be physically separated, component shown as a unit can be or
Person may not be physical unit, it can and it is in one place, or may be distributed over multiple network units.It can root
According to actual need that some or all of the modules therein is selected to achieve the purpose of the solution of this embodiment.Ordinary skill
Personnel can understand and implement without creative efforts.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by anyone skilled in the art,
It should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims
Subject to.
Claims (5)
1. a kind of railway emergency event grading forewarning system method based on K-means cluster characterized by comprising
Data are influenced on history railway emergency event and carry out signature analysis and data cleansing, obtain training dataset;
It determines cluster centre number, the training dataset is clustered based on K-means algorithm, obtained according to cluster result each
The sample data feature of grade railway emergency event;
The data characteristics of current rail emergency event is compared with the sample data feature of the railway emergency events at different levels,
The emergency event grade of the current rail emergency event is determined according to comparison result.
2. the method according to claim 1, wherein described influence data progress to history railway emergency event
Signature analysis and data cleansing obtain training dataset, comprising:
By influencing data analysis to history railway emergency event, from railroad train operating status angle, railway burst thing is extracted
The influence feature X={ x of part1,x2,L,xd, the influence feature vector of real number is tieed up at d according to influence feature X-shaped, wherein xiIt is i-th
Dimension influences the value of feature;
According to the influence feature vector, arranging the emergency event of history railway influences to form record set, carries out to the record set
Data cleansing handles the record sample that characteristic lacks in the record set, forms feature complete data collection
Wherein X 'i={ x 'i,1,x′i,2,L,x′i,dIt is the complete data sample of ith feature, x 'i,dFor the d dimension of i-th of data sample
Characteristic value, N are characterized complete data number of samples;
One-hot coding is carried out to the classifying type feature that the feature complete data is concentrated, the number concentrated to the feature complete data
Value type feature carries out minimax normalized, the data feature values x after normalizationi,jAre as follows:
Wherein x 'i,jFor the j dimensional feature value of i-th of data sample before normalized,
Obtain training datasetXi={ xi,1,xi,2,L,xi,d}。
3. according to the method described in claim 2, it is characterized in that, the determination cluster centre number, is based on K-means algorithm
The training dataset is clustered, the sample data feature of railway emergency events at different levels is obtained according to cluster result, comprising:
Step 2.1: using a certain range of trellis search method, calculate separately Calinski-Harabaz Index and iteration
Error determines cluster centre number k;
Step 2.2: k number is arbitrarily chosen from N number of data sample according to sample as initial cluster centerAnd just
Beginningization iterator m=0;
Step 2.3: calculating training data and concentrate the Euclidean distance between each data sample and cluster centre, to maximize between cluster
Distance, minimum intra-cluster distance is target, with φminObjective function divides the classification of each data sample, wherein
wi,jFor 0-1 variable, wi,j=1 indicates sample XiBelong to cluster centre CjSample in cluster, otherwise wi,j=0;
Step 2.4: according to sample data in cluster after update, calculating every cluster cluster centre, update m=m+1 and cluster centre
Step 2.5: calculating cluster variance, and judge to cluster whether variance meets minimum sandards, if it is satisfied, k cluster of output
Data sample in center and each cluster;Otherwise, step 2.3 is continued to execute.
4. according to the method described in claim 3, it is characterized in that, described obtain railways burst things at different levels according to cluster result
The sample data feature of part, comprising:
I grade of early warning feature of emergency event is Train delay time 21min or more or train speed limit 250km/h or less;Emergency event
II grade of early warning feature is Train delay time 14~21min or 250~300km/h of train speed limit;III grade of early warning spy of emergency event
Sign is Train delay time 9~14min or 300~310km/h of train speed limit;IV grade of early warning feature of emergency event is Train delay
310~325km/h of 5~9min of time or train speed limit;V grade of early warning feature of emergency event is Train delay time 5min or less
Or 325~350km/h of train speed limit.
5. according to the method described in claim 4, it is characterized in that, the data characteristics by current rail emergency event with
The sample data feature of the railway emergency event at different levels is compared, and determines the current rail burst thing according to comparison result
The emergency event grade of part, comprising:
The sample data feature vector that I grade, II grade, III grade, IV grade and V grade of the emergency event includes: train number grade, vehicle,
Position and attribute, event, which occur, for route, event where event type, event occur influences section and station quantity, Train delay
Whether type the late time, train speed limit, late train quantity, stoppage in transit train quantity, influences the duration and enables spare dynamic
Vehicle group;
By in the railway emergency event currently occurred the Train delay time and train speed limit quantization respectively with above-mentioned emergency event I
Grade, II grade, III grade, the late time in the sample data feature vector of IV grade and V grade, train speed limit be compared, according to than
Relatively I grade, II grade, III grade, IV grade and V grade event class criteria for classifying of result combination emergency event, positions current emergency event etc.
Grade realizes railway emergency event grading forewarning system.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811191236.2A CN109242209B (en) | 2018-10-12 | 2018-10-12 | Railway emergency grading early warning method based on K-means clustering |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811191236.2A CN109242209B (en) | 2018-10-12 | 2018-10-12 | Railway emergency grading early warning method based on K-means clustering |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109242209A true CN109242209A (en) | 2019-01-18 |
CN109242209B CN109242209B (en) | 2022-03-15 |
Family
ID=65053434
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811191236.2A Active CN109242209B (en) | 2018-10-12 | 2018-10-12 | Railway emergency grading early warning method based on K-means clustering |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109242209B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110569981A (en) * | 2019-07-22 | 2019-12-13 | 国网冀北电力有限公司电力科学研究院 | Method and device for determining fuel consumption of scattered coal |
CN111047167A (en) * | 2019-11-29 | 2020-04-21 | 阿里巴巴集团控股有限公司 | Line data processing method, device, equipment and storage medium |
CN111667392A (en) * | 2020-06-12 | 2020-09-15 | 成都国铁电气设备有限公司 | Railway contact net defect hot spot area early warning method based on space-time clustering |
CN111907562A (en) * | 2020-06-24 | 2020-11-10 | 中铁第一勘察设计院集团有限公司 | Real-time monitoring and control system for railway communication state |
CN113421000A (en) * | 2021-06-30 | 2021-09-21 | 中国人民解放军国防科技大学 | Autonomous and controllable evaluation method for communication equipment |
CN113420722A (en) * | 2021-07-21 | 2021-09-21 | 上海塞嘉电子科技有限公司 | Emergency linkage method and system for airport security management platform |
CN115374855A (en) * | 2022-08-23 | 2022-11-22 | 北京交通大学 | Method for grading railway emergency based on clustering |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101833858A (en) * | 2009-12-17 | 2010-09-15 | 南京城际在线信息技术有限公司 | Method for judging road traffic state based on annular coil of signal lamp system |
CN103150900A (en) * | 2013-02-04 | 2013-06-12 | 合肥革绿信息科技有限公司 | Traffic jam event automatic detecting method based on videos |
US20140095425A1 (en) * | 2012-09-28 | 2014-04-03 | Sphere Of Influence, Inc. | System and method for predicting events |
CN106202335A (en) * | 2016-06-28 | 2016-12-07 | 银江股份有限公司 | A kind of big Data Cleaning Method of traffic based on cloud computing framework |
CN106650784A (en) * | 2016-11-04 | 2017-05-10 | 许继集团有限公司 | Feature clustering comparison-based power prediction method and device for photovoltaic power station |
CN108492557A (en) * | 2018-03-23 | 2018-09-04 | 四川高路交通信息工程有限公司 | Highway jam level judgment method based on multi-model fusion |
-
2018
- 2018-10-12 CN CN201811191236.2A patent/CN109242209B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101833858A (en) * | 2009-12-17 | 2010-09-15 | 南京城际在线信息技术有限公司 | Method for judging road traffic state based on annular coil of signal lamp system |
US20140095425A1 (en) * | 2012-09-28 | 2014-04-03 | Sphere Of Influence, Inc. | System and method for predicting events |
CN103150900A (en) * | 2013-02-04 | 2013-06-12 | 合肥革绿信息科技有限公司 | Traffic jam event automatic detecting method based on videos |
CN106202335A (en) * | 2016-06-28 | 2016-12-07 | 银江股份有限公司 | A kind of big Data Cleaning Method of traffic based on cloud computing framework |
CN106650784A (en) * | 2016-11-04 | 2017-05-10 | 许继集团有限公司 | Feature clustering comparison-based power prediction method and device for photovoltaic power station |
CN108492557A (en) * | 2018-03-23 | 2018-09-04 | 四川高路交通信息工程有限公司 | Highway jam level judgment method based on multi-model fusion |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110569981A (en) * | 2019-07-22 | 2019-12-13 | 国网冀北电力有限公司电力科学研究院 | Method and device for determining fuel consumption of scattered coal |
CN111047167A (en) * | 2019-11-29 | 2020-04-21 | 阿里巴巴集团控股有限公司 | Line data processing method, device, equipment and storage medium |
CN111047167B (en) * | 2019-11-29 | 2023-04-28 | 阿里巴巴集团控股有限公司 | Line data processing method, device, equipment and storage medium |
CN111667392A (en) * | 2020-06-12 | 2020-09-15 | 成都国铁电气设备有限公司 | Railway contact net defect hot spot area early warning method based on space-time clustering |
CN111907562A (en) * | 2020-06-24 | 2020-11-10 | 中铁第一勘察设计院集团有限公司 | Real-time monitoring and control system for railway communication state |
CN113421000A (en) * | 2021-06-30 | 2021-09-21 | 中国人民解放军国防科技大学 | Autonomous and controllable evaluation method for communication equipment |
CN113421000B (en) * | 2021-06-30 | 2023-07-11 | 中国人民解放军国防科技大学 | Autonomous controllable evaluation method for communication equipment |
CN113420722A (en) * | 2021-07-21 | 2021-09-21 | 上海塞嘉电子科技有限公司 | Emergency linkage method and system for airport security management platform |
CN113420722B (en) * | 2021-07-21 | 2023-02-17 | 上海塞嘉电子科技有限公司 | Emergency linkage method and system for airport security management platform |
CN115374855A (en) * | 2022-08-23 | 2022-11-22 | 北京交通大学 | Method for grading railway emergency based on clustering |
CN115374855B (en) * | 2022-08-23 | 2024-05-14 | 北京交通大学 | Clustering-based method for grading railway emergency |
Also Published As
Publication number | Publication date |
---|---|
CN109242209B (en) | 2022-03-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242209A (en) | Railway emergency event grading forewarning system method based on K-means cluster | |
Xia et al. | A clustering approach to online freeway traffic state identification using ITS data | |
CN105893208A (en) | Cloud computing platform system fault prediction method based on hidden semi-Markov models | |
CN110084151A (en) | Video abnormal behaviour method of discrimination based on non-local network's deep learning | |
CN104156403B (en) | A kind of big data normal mode extracting method and system based on cluster | |
CN112949715A (en) | SVM (support vector machine) -based rail transit fault diagnosis method | |
CN110033135A (en) | The train braking system failure prediction method that Multivariate Time Series feature is reinforced | |
Hua et al. | Extraction and analysis of risk factors from Chinese railway accident reports | |
CN109472462A (en) | A kind of project risk ranking method and device based on the fusion of multi-model storehouse | |
CN105574489A (en) | Layered stack based violent group behavior detection method | |
CN110532298A (en) | More attribute railway accident reason weight analysis methods | |
CN112884179A (en) | Urban rail turn-back fault diagnosis method based on machine fault and text topic analysis | |
CN106611016B (en) | A kind of image search method based on decomposable word packet model | |
CN116090819A (en) | Power distribution network risk situation prediction method based on association rule | |
CN109033351A (en) | The merging method and device of merit data | |
Xinqin et al. | Application of bagging ensemble classifier based on genetic algorithm in the text classification of railway fault hazards | |
CN112101779A (en) | Comprehensive evaluation system for airport scene operation based on subjective and objective combination structure entropy weight method | |
CN103761449A (en) | Criminal propensity and risk degree quantifying method and system based on AHP | |
CN113295413B (en) | Traction motor bearing fault diagnosis method based on indirect signals | |
Zhang et al. | Vibration signal-based defect detection method for railway signal relay using parameter-optimized VMD and ensemble feature selection | |
CN109582743A (en) | A kind of data digging method for the attack of terrorism | |
CN113379318A (en) | Method and device for evaluating operation service quality of public transport system and computer equipment | |
CN114880923A (en) | Enterprise credit prediction method fusing power data | |
Yang et al. | Design and Application of Big Data Technology Management for the Analysis System of High Speed Railway Operation Safety Rules | |
Yaghoubi et al. | Model-based clustering (MBC) for road data via multivariate mixture of normal distributions and factor analysis (FA) |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |