CN104077552B - A kind of track traffic signal based on cloud computing integrates O&M method and system - Google Patents

A kind of track traffic signal based on cloud computing integrates O&M method and system Download PDF

Info

Publication number
CN104077552B
CN104077552B CN201410321145.1A CN201410321145A CN104077552B CN 104077552 B CN104077552 B CN 104077552B CN 201410321145 A CN201410321145 A CN 201410321145A CN 104077552 B CN104077552 B CN 104077552B
Authority
CN
China
Prior art keywords
data
station
index
railroad
electricity business
Prior art date
Application number
CN201410321145.1A
Other languages
Chinese (zh)
Other versions
CN104077552A (en
Inventor
鲍侠
Original Assignee
北京泰乐德信息技术有限公司
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 北京泰乐德信息技术有限公司 filed Critical 北京泰乐德信息技术有限公司
Priority to CN201410321145.1A priority Critical patent/CN104077552B/en
Publication of CN104077552A publication Critical patent/CN104077552A/en
Application granted granted Critical
Publication of CN104077552B publication Critical patent/CN104077552B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism

Abstract

O&M method and system are integrated the present invention relates to a kind of track traffic signal based on cloud computing.Using railroad, Railway Bureau, electricity business section, the level Four pattern at station, data center, data gather computer and terminal are disposed respectively in railroad, Railway Bureau, and data gather computer and terminal are only disposed at electricity business section, station;Real-time data analysis component is disposed simultaneously, and station, the real-time data analysis deployment of components of electricity business section are in data gather computer, and Railway Bureau, the real-time data analysis deployment of components of railroad are in data center.The present invention uses the deployment way of level, by the way that directly deployment real-time data analysis component realizes the real-time analysis to signal data on harvester, the storage demand to ever-increasing signal data is realized by being classified storage and index strategy, and realize quick positioning to failure by four layer index structures, the information such as time, type, position that can occur by failure quickly retrieves corresponding signal data when failure occurs.

Description

A kind of track traffic signal based on cloud computing integrates O&M method and system

Technical field

The present invention provides a kind of railway signal based on cloud computing and integrates O&M method and system, is related to railway signal number According to, railway communication data, railway knowledge data, system alarm data, machine learning, mobile unit, station equipment, center set The technical fields such as standby, trackside equipment, O&M problem encountered is integrated to solve railway signal.

Background technology

At present, track traffic (government railway, enterprise railway and urban track traffic) communication, the monitoring and maintenance in signal field Product mainly has three classes:There are CSM (centralized signal supervision system), each plant maintenance machine, communication network for trackside signal facility Guard system.In order to improve the modernization maintenance level of China railways signal system equipment, since the nineties, successively independent development The constantly centralized signal supervision CSM systems during upgrading such as TJWX-I types and TJWX-2000 types.Major part station is all used at present Centralized signal supervision system, realizes real-time monitoring to signaling at stations equipment state, and set with tracer signal by monitoring Standby main running status, grasps the current state of equipment for telecommunication and signaling branch and carries out crash analysis providing basic foundation, send out Important function is waved.Also, to Urban Rail Transit Signal equipment, Centralizing inspection CSM systems are also widely deployed in urban rail collection Middle station/rolling stock section etc., uses for urban rail O&M.There are DMS systems for vehicle-mounted signal device, mobile unit is dynamically supervised Survey.In addition, with the development of China Express Railway, the distinctive RBC systems of high ferro etc., being also faced with and including signal concentration prison The demand of examining system, be also faced with raising its monitoring capability, O&M ability, and equipment self-diagnosis ability demand.

In face of increasing communication data, signal data, and current network condition so that although the railway system produces Substantial amounts of Various types of data has been given birth to, but has not utilized these data well.Most data are stored in each at present Station, and station also simply data have been done with simple storage, when break down when check correlation data in order to Analyzing failure cause.The Various types of data that such case result in railway generation effectively can not be stored and utilized, and failure Analysis still need at present by artificial experience analyze judge, failure could be found when having there is accident in many cases, no The technical problem such as workload big, Fault monitoring and diagnosis inefficiency when having only resulted in Artificial Diagnosis railway signal system failure, Also increase the danger of driving.Therefore, railway signal integrated management O&M ability is improved, hidden danger is looked into, hidden danger is controlled, promotes failure to repair Develop to status maintenance, so that guarantee driving safety, raising transport power, are the active demands of field of track traffic.

The content of the invention

In order to solve signal data amount in the prior art is big, efficiency is low, risk is high, manual analysis failure labor intensity is big etc. Problem, the invention provides a kind of signal synthesis O&M method and system based on cloud computing.System includes data gather computer, end End and three kinds of hardware of data center.Software include data acquisition subsystem, data storage subsystem, data prediction subsystem, The real-time analyzing subsystem of data, off-line data excavate subsystem, displaying subsystem, data query system etc..

The technical solution adopted by the present invention is as follows:

A kind of track traffic signal based on cloud computing integrates operational system, and it includes:

Data gather computer, for acquisition trajectory data traffic signal;

Data center, connects the data gather computer, including Yunan County's holostrome, basic resource layer, virtualization layer, data storage Layer, computing engines layer, component layer and Web clusters;Wherein, Yunan County's holostrome is protected by data backup, data convert and access control The safety of barrier system;Basic resource layer is the hardware platform of data center;Virtualization layer is realized to hardware by virtualization software Virtualization, shield bottom hardware differences;Data storage layer includes distributed file system, columnar database, relation data Storehouse, is respectively used to unstructured, semi-structured and structuring the data of storage;Computing engines layer handles frame using cloud computing;Group Part layer is the component of specific processing miscellaneous service;Web clusters are the service issue layers based on load balancing;

Terminal, connects the data gather computer and the data center, for display data analysis result.

Further, using railroad, Railway Bureau, electricity business section, the level Four pattern at station, in railroad, Railway Bureau Deployment data center, data gather computer and terminal, data gather computer and terminal are only disposed at electricity business section, station respectively.AT STATION, Electricity business section, Railway Bureau, railroad dispose real-time data analysis component, wherein station, the real-time data analysis group of electricity business section Part is deployed in data gather computer, and Railway Bureau, the real-time data analysis deployment of components of railroad are in data center.

Further, computing engines layer includes off-line data processing framework MapReduce, real-time distributed processing framework Spark and data movement engine sqoop and underlying resource management system yarn, Yarn be used for being submitted according to user of the task come Distribution storage resource and computing resource are used to handle the task.

Further, component layer includes data management component, components of data analysis, fault diagnosis component etc..

A kind of track traffic signal based on cloud computing of use said system integrates O&M method, and its step includes:

1) signal synthesis operation platform firstly the need of solution be data acquisition the problem of, for current rail use iron Lu companies, Railway Bureau, electricity business section, the level Four management mode at station, data gather computer, which is also distributed about this four levels, to be used to gather Different types of signal data, and data are done with different processing;Data center, number are disposed respectively in railroad, Railway Bureau According to harvester and terminal, data gather computer and terminal are only disposed at electricity business section, station;AT STATION, electricity business section, Railway Bureau, railway Company disposes real-time data analysis component, and wherein station, the real-time data analysis deployment of components of electricity business section are in data gather computer Interior, Railway Bureau, the real-time data analysis deployment of components of railroad are in data center;

In view of the current network presence of railway, each railway internal signal system is entered using the cable network of 2M bandwidth Row data transfer, with the continuous improvement of signal data species and frequency that station is gathered, the single station letter per second collected Number is about that between 200-500kb, an electricity business section has about tens to hundreds of station;So result in the data at station Electricity business section can not possibly be completely transmitted to;The strategy that harvester is locally stored is used to Monitoring Data according to such case, only portion Conduct electricity business section and Railway Bureau, railroad in divided data;

2) data gather computer at station gathers the real time data at station, utilizes the real-time data analysis component in local disposition Signal data is pre-processed, feature extraction, feature selecting, then characteristic is divided in real time using analysis model Analysis, obtains the current running status of signal system, and (corresponding data can be included into the Partial Feature and analysis knot of extraction Really, may also comprise the initial data at each station) transmit to electricity business section;

3) data gather computer of electricity business section receives the data transmitted from each internal station, utilizes local disposition Real-time data analysis component the data at the different stations of reception are analyzed, obtain the analysis result of whole electricity business section, and Analysis result is transmitted to the data gather computer of Railway Bureau;

4) data gather computer of Railway Bureau receives the data from each electricity business section, and by data conversion storage to data center, Then data are handled and analyzed using the real-time data analysis component of data center deployment, and by analysis result and part Feature is transferred to the data gather computer of railroad;

5) data gather computer of railroad receives the data that Railway Bureau sends over, and by data conversion storage into data The heart, is analyzed and processed using the real-time data analysis component of data center, obtains the global analysis result of railroad;

6) using the terminal for being deployed in railroad, Railway Bureau, electricity business section and station, the exhibition of data results is carried out Show.

Further, in addition to step 7), the step to railroad, Railway Bureau, electricity business section and station storage data Index is set up, when breaking down according to the index quick obtaining relevant time period and the various types of signal data of position, in order to It is rapidly performed by accident analysis, positioning and solves.

Further, the data gather computer can be multiplexed existing machine, it is only necessary to install corresponding data storage group Part, data prediction component, real-time data analysis component and data transfer components.Railroad, the data gather computer of Railway Bureau Only need to dispose data transfer components, it is not necessary to complete data storage, the function of processing.The data storage of data gather computer is used Storage strategy is classified, the characteristic of extraction directly can be transferred to upper strata in the case that railway network bandwidth is enough, this Sample harvester locally avoids the need for stored signal data.

Further, the data gather computer is stored in local system for the local signal data collected, then The storage organizations of data is managed on upper strata in the form of metadata.The main innovation of the present invention is hierarchical data point Cloth storage mode, is not simple distributed storage, but a kind of new layer that the network based on railway is limited and designed Secondary data storage architecture, i.e.,:Initial data is respectively stored in collected level, and upper strata is in addition to the data for storing this layer, also Need to store lower floor metadata (metadata generally refers to index data here, that is, lower floor's True Data storage letter Breath, such as some data storages where), for the particular location of quick location data, upper strata by metadata come Manage the data of lower floor.When data are calculated, nor simply transmitting lower floor's analysis result upwards, divide except needing to transmit Analyse outside result, in addition it is also necessary to the foundation that hop initial data is analyzed as upper layer data.So as to form a kind of number of hierarchical According to storage, processing framework.Such as electricity business section includes two class data, and a class is the data that electric section of being engaged in is gathered in itself, is in area under control in addition Each station metadata, including the information such as Station XXX, device numbering, time interval, storage location, so that ensure can be fast The location data of speed.Upwards can be by that analogy.

Further, step 2) in the data gather computer at station signal data, machine of standing are obtained by the signal station machine at station Whether need to be transmitted by existing rule judgment first, so as to be compressed signal data processing.As voltage (also may be used To be other signal values) value in the case where standard value fluctuation is no more than 20% (can also be other threshold values), it is believed that number According to not changing, then it need not transmit.Data gather computer is set up socket by the signal station machine with station and is connected, station Signal station machine compression is handled after data transfer to data gather computer, data gather computer is entered using existing rule to data Row reduction.

Further, step 2)~5) described in real-time data analysis component be mainly and carried out according to obtained characteristic Failure predication, is classified using disaggregated model to feature, judges whether to produce failure.Specifically, real-time data analysis is root The disaggregated model that analysis mining obtains being directed to all kinds of failures is carried out according to the historical signal data marked.Real-time data analysis group The data analysis algorithm that part is used includes SVMs, Bayes classifier, rough set, decision tree, neutral net etc., utilizes Algorithm carries out unified processing to the signal data that collects in data center, and excavating the model of Fault Identification, there is provided to reality When analysis establishment use.

Further, step 6) in:The terminal at described station is mainly used in showing each alanysis knot inside the station And various failure analysis results really, including real-time running status,;The terminal of electricity business section is mainly used in showing in electricity business section Overall running status, when the station in area under control is broken down, the terminal is also required to show corresponding fault message;Iron The terminal of road bureau, the running status for showing Railway Bureau's aspect, various fault alarm informations etc.;The terminal of railroad, is used In the running status for showing whole railroad, quickly displaying is also required to when station is broken down and is tracked.

Further, step 7) described in process need first each station according to the data of storage set up index, so The index file at station is transferred to electricity business section afterwards, electricity business section is built to the data locally gathered and the station index file received Vertical secondary index, then transmits index to the data center of Railway Bureau, the Various types of data of the data center of Railway Bureau to reception Three level list is set up, is then transmitted index to the data center of railroad, the data center of railroad is to each of reception Class data set up level Four index, form a railroad, Railway Bureau, electricity business section, the level Four at station index.Can during specific implementation To set up index according to modes such as Railway Bureau, electricity business section, station, equipment.And then by means of the index file of foundation, can basis The time of failure, positional information are directly obtained specific signal data.

Compared with prior art, it is an advantage of the invention that:

The present invention has unified the integrated management of railway signal data, including the collections of data, data storage, data analysis with And result displaying.The life cycle of whole signal data is managed collectively, an organic whole is formed, improves signal data Utilization rate.

The present invention will be dispersed in each electricity business section, the various signal datas at station and unified by data collecting system Collection and processing, so as to change the mutually isolated present situation of various signal datas.Signal data will be carried out for future further Excavate and provide good acquisition platform.

The present invention has adapted to railroad, Railway Bureau, electricity business section and the station of railway by the deployment way of level Level Four management mode, and the deployment way of this level also applies to the current network presence of railway, optimizes data The mode of concentration, greatly reduces the data volume for being transferred to electricity business section and railroad.

By the present invention in that using Model Identification failure, substantial amounts of human cost is saved, it is no longer necessary to which artificial goes observation Monitoring information and then progress Fault Identification and analysis;The accuracy rate of track traffic Monitoring Data Fault Identification can be improved, is shortened Fault correction time, greatly improves the troubleshooting efficiency of track traffic, improves O&M ability.

The present invention on harvester by directly disposing real-time data analysis component, it is possible to achieve to the real-time of signal data Analysis, improves the real-time of data analysis, analysis result directly can be locally shown.

The present invention realizes the storage demand to ever-increasing signal data by being classified the strategy for storing and indexing, and And quick positioning to failure is realized by four layer index structures, time that related personnel can occur by failure, type, position The information such as put and quickly retrieve corresponding signal data when failure occurs.So technical staff and administrative staff can be more Quick directly positioning failure, and propose solution.

Brief description of the drawings

Fig. 1 is railway signal system institutional framework schematic diagram.

Fig. 2 is the integrated stand composition of the synthesis operational system of the present invention.

Fig. 3 is the integral deployment figure of data gather computer, data center and the terminal of the present invention.

Fig. 4 is data center architecture figure of the present invention.

Effect schematic diagram of the synthesis operational system of Fig. 5 present invention in Digital Railway " integration " platform.

Fig. 6 is the flow chart of a 25Hz phase-sensitive track circuits differentiation indoor and outdoor failure in embodiment.

Embodiment

Below by specific embodiments and the drawings, the present invention is described in detail.

The synthesis O&M scheme of the railway signal of the present invention is used to solve signal data decentralized processing, work in the prior art The technical problems such as amount is big, inefficiency, risk height, fault inquiry difficulty.The user object of the present invention is mainly government railway With domestic large enterprise's railway, such as urban track traffic.It will be seen from figure 1 that being divided into railway inside railway signal system Parent company, Railway Bureau, electricity business section and station level Four.Introduced with reference to Fig. 1 iron signal system inner stages department of state and Their corresponding functions.

1) railway parent company

Railway parent company telecommunication and signaling branch is responsible for the railway signals equipment of national each road bureau, is the final remittance of all monitoring datas Collect point, equipment coverage rate is most wide, it is most comprehensive to be related to device category, the data volume maximum for analysis.Main task includes: (1) equipment angle:The working condition of the whole signalling arrangements of each road bureau is grasped at any time;By statistical report form function, various letters are understood The information such as number fault rate, failure mode, the fault impact of equipment in actual applications;Grasp the maintenance feelings of each road bureau's signalling arrangement Condition;By statistical report form function, the combination property to different manufacturers, variety classes, the signalling arrangement of different model is compared, Data, which are provided, for equipment purchase, operation, maintenance etc. supports and quantify reference;Analyzed by Data Integration, find signalling arrangement Weak link, specific aim suggestion is provided for the introducing of the upgrading and new system, new equipment of equipment;(2) O&M angle:Grasp The O&M situation of each Railway Bureau, arranges Test Gauge, rectification emphasis, prevents trouble before it happens targeted specifically.

2) Railway Bureau

The signalling arrangement species of each road bureau is comparatively relatively more unified, and main task includes:(1) equipment angle:Check The warning message of each signalling arrangement;By statistical report form function, understand various signalling arrangements fault rate in actual applications therefore Hinder the information such as species, fault impact;The disposition of failure is understood in time;Monitor the working condition of institute's supervisor's signalling arrangement;Hair During raw failure, warning message, state monitoring information of Comprehensive Correlation mobile unit and ground installation etc., judgement cause the root of failure This reason;By functions such as statistical report form and data analyses, different vendor's equipment is compared, and phase is provided to equipment manufacturer Answer demand and opinion;(2) O&M angle:The O&M situation of each electricity business section, arranges Test Gauge, whole targeted specifically belonging to grasping Change emphasis, prevent trouble before it happens.

3) electricity business section

All signalling arrangements in its pipe are carried out with overall monitoring, guidance is operated to each workshop of its subordinate.Carry out The specific maintenance work of equipment, is concerned about aggregate analysis processing and the comparative result of Various types of data again.

4) station/work area

Station is included between common workshop, vehicle-mounted workshop, fortune shunt, and some stations also include high speed workshop etc..It is respectively used to The different monitoring device of management, including signalling arrangement and Block signaling equipment, mobile unit, CTC, RBC, TDCS etc. are set in station It is standby.The reason for finding failure in time by comprehensive monitoring system and recognize failure.Failure is recognized by various statistical report forms The information such as place, equipment, the monitoring of emphasis is carried out for the higher position of rate of breakdown, the equipment easily broken down is entered Row statistical analysis, is that equipment replacement and buying provide foundation.

The data storage and processing mode of the stagewise of the present invention are highly suitable for the classification feature of state's iron and at different levels Data analysis requirements.The present invention is i.e. using railroad, Railway Bureau, electricity business section, the level Four pattern at station, in railroad, iron Road bureau disposes data center, data gather computer and terminal respectively, and data gather computer and terminal are only disposed at electricity business section, station. Station, electricity business section, Railway Bureau, railroad's deployment real-time data analysis component, wherein station, the real-time data analysis of electricity business section Deployment of components is in data gather computer, and Railway Bureau, the real-time data analysis deployment of components of railroad are in data center.Railway Office, railway parent company also need to dispose off-line data digging system in data center, for analysis mining real-time data analysis Used model.The solution of the present invention mainly includes the content of following six parts:

(1) data acquisition:

Need the various types of signal data acquisition for producing railway, including communication data, mobile unit data, station are set Standby data, central apparatus data, trackside equipment data etc..

Railway signal includes polytype, such as communication data, vehicle-mounted data, trackside data, station data, the position of generation Putting also includes station, electricity business section, administrative center of Railway Bureau, railroad center etc., therefore data acquisition subsystem is also required to portion Administration AT STATION, electricity business section, Railway Bureau and signal center of railroad.Data acquisition subsystem needs the rule according to signal data Model, receives and parses through signal data.The signal data collected is transferred to data prediction subsystem by signal acquisition subsystem, Signal data is analyzed and processed.

(2) data prediction:

Data prediction subsystem receives the signal data collected from data acquisition subsystem, according to different events Barrier, finds the signal data related to the failure, then data is carried out with the pretreatment such as duplicate removal, denoising first, it is ensured that data Validity.Vector space model is converted data to according to signal data, and as needed, such as believed using feature selecting algorithm Breath gain algorithm etc. is extracted to feature, duplicate removal and selection, is found to failure modes useful feature, is ultimately generated suitable for mould Type training and the characteristic vector analyzed in real time.

(3) disaggregated model is excavated:

Manually marked first for the data collected, all kinds of fault datas marked pass through these data It is trained, corresponding disaggregated model and parameter can be produced, the signal data for next step is analyzed.

The subsystem is directed to by the analyses of the Historical Monitoring data to manually marking, first to number that these have been marked According to data prediction is carried out, the characteristic vector that band marks classification is produced, appropriate feature and initial parameter is then selected to these Data are trained, so as to obtain the model of failure modes.

Selection meets the model of rule, and model is trained, and finds and finds corresponding classification mould for specific failure Type and parameter, make it have best classifying quality to such failure.

(4) signal data is analyzed in real time:

Including steps such as data filtering, duplicate removal, feature extraction, feature selecting, normalization, then using fault model to this A little data carry out failure predication and classification.The subsystem is received by the live signal number after data prediction subsystem processes According to the disaggregated model obtained using disaggregated model excavation subsystem carries out analysis calculating to these data, obtains current system Running status.Do not have normally to be shown when failure, need to remind in the way of alerting when breaking down Related personnel.

(5) analysis result is shown:

The subsystem is various display terminals, is deployed in the signal center of station, electricity business section, Railway Bureau and railroad, There is provided to site technology for running status, timely reminding alarm information, inquiry signal data for showing administrative area in real time etc. Personnel and administrative staff use.It is easy to related personnel can be with the current running status of more intuitive understanding system.For difference The exhibition method and scope of the result of the signal data analysis of level (station, electricity business section, Railway Bureau, railroad) are not yet Together, station is the real-time analysis situation inside the real-time exhibition station;Electricity business section shows the part point for administering all stations Analyse result and all fault messages;Railroad station section show in company it is all electricity business sections, the running status at station and Analysis result.

For above step (2)~(4), AT STATION with electricity business section two-stage because of no deployment data center, therefore just It is (i.e. the real-time data analysis component) completed in harvester;And then completed in Railway Bureau and railroad in data center. When it is implemented, Railway Bureau and railroad can also not need special data gather computer, directly data acquisition components portion Administration realizes the function of data gather computer on central server.

(6) to railroad, Railway Bureau, electricity business section and station storage data set up index, break down when according to The index quick obtaining relevant time period and the various types of signal data of position, in order to be rapidly performed by accident analysis, positioning and Solve.

A) structure of hierarchical index

Railway signal system departments at different levels are only inquired about and analyzed to the signal data in administrative area, therefore demand data With obvious locality.The characteristics of according to the current network presence of the railway system and demand data, entered by the way of classification Row signal data is stored, for carry out data query rapidly and efficiently, using the data directory strategy of stagewise.Namely in car Stand, electricity business section, Railway Bureau, railroad are respectively adopted different data directory structures to ensure the speed of data query.Here The concept of data block (Block) is introduced, is the memory space with 64M fixed sizes, by the data center services of railroad Device carries out unified numbering to the signal data storage device of the whole network.

B) station data directory structure, as shown in table 1.

The station data directory structure of table 1.

workid deviceid blockid offset length

workid:The numbering in workshop, internal AT STATION is unique;

deviceid:The numbering of monitoring device in workshop;

blockid:Data block is numbered, and this is globally unique numbering, and the particular location of data storage can be by this Individual id maps directly to corresponding physical storage locations;

offset:Original position of the signal data in the data block;

length:The size of signal data, is also the memory length that data take;

The index data at station is generated in data storage, and is stored in local machine, and station terminal carries out data and looked into During inquiry, index file is read into local machine first, is then gone to read real data according to data block, offset.

C) electricity business segment data index structure

Electricity business section stores two kinds of data, is the actual signal data and index data collected respectively.Index data It is also classified into two classes:The index of station index data;The index of local data.Local data index is similar with station data, simply Workid becomes to represent the depotid of electricity business section, as shown in table 2.

The local data index structure of the electricity business section of table 2.

depotid deviceid blockid offset length

Table 3 is the data structure of the index of station index data:

The index structure of the station index data of the electricity business section of table 3.

Stationid workid blockid offset length

Stationed:The numbering at station, is unique in the range of system-wide;

The index adds a Station XXX relative to the index at station, and for recognizing station, an index record is just Represent the particular location of the index file storage in station work area, and index file length etc..

D) Railway Bureau's data directory structure

Railway Bureau is similar with electricity business section, and the data of storage are divided into two classes:Data and index that Railway Bureau collects in itself Data.Wherein index data is divided into the index of electricity business segment index in local data index and area under control.

Local data index data structure is as shown in table 4:

The local data index data structure of the Railway Bureau of table 4.

officeid deviceid blockid offset length

Officeid:It is the numbering of Railway Bureau, the unique encodings for recognizing the Railway Bureau.

The index data structure of electricity business segment index is as shown in table 5:

The index data structure of the electricity business segment index of the Railway Bureau of table 5.

deoptid Stationid blockid offset length

The index adds an electricity business segment number relative to the index of electricity business section, for recognizing electricity business section, an index Record just represents the particular location of the index file storage of an electricity business section, and index file length etc..

E) railroad's data directory structure

Railroad is similar with Railway Bureau, and the data of storage are divided into two classes:Data that railroad collects in itself and Index data.Wherein index data is divided into local data index and the index of each Railway Bureau's index data.

Local data index data structure is as shown in table 6:

The local data index data structure of the railroad of table 6.

companyid deviceid blockid offset length

The index data structure of Railway Bureau's index is as shown in table 7:

The index data structure of Railway Bureau's index of the railroad of table 7.

officeid deoptid blockid offset length

The index adds Railway Bureau's numbering relative to the index of Railway Bureau, for recognizing Railway Bureau, an index Record just represents the particular location of the index file storage of Railway Bureau, and index file length etc..

Pass through above-mentioned data storages at different levels and index structure, it is ensured that all departments at different levels can be according to the index text of itself Part quickly inquires all signal datas in area under control, has taken into account the actual feelings of railway signal system current network, memory space The access speed of signal data is ensure that in the case of condition.

The present invention realizes the centralized management to railway signal data using comprehensive operation platform, realizes all kinds of to railway The unified of signal data is gathered and centrally stored.Data prediction subsystem is unified to all kinds of Monitoring Datas collected Pretreatment, extracts useful feature to be supplied to data mining and analysis in real time to use.Disaggregated model excavates subsystem and passed through Data to mark are analyzed and processed, and find out suitable disaggregated model and parameter for specific failure problems, and will excavate As a result real-time data analysis subsystem is transferred to use.Data analytics subsystem is using disaggregated model to believing in real time after pretreatment Number characteristic is classified, and obtains failure analysis result, and the result is transferred into result displaying subsystem, as a result shows son System shows corresponding analysis result according to the location of itself, including station is internal, electric intersegmental part and the whole railway of being engaged in The running situation of company.The idiographic flow of this method is illustrated below by diagram and example:

Fig. 2 is the integrated stand composition of system, the system include data collecting system (i.e. data gather computer), data center and Terminal, data center include structural data storage, semi-structured data storage, unstructured data storage, at off-line data The modules such as reason, the processing of real-time big data, statistical analysis, data mining, fault pre-alarming, query engine, push engine.Terminal can be with It is various types of receiving terminals, including PC computers, notebook, Pad, smart mobile phone etc..

Fig. 3 is the deployment schematic diagram of system, and point four levels are disposed, respectively railroad, Railway Bureau, electricity business section And station.

A) railroad

The level needs deployment data center, data gather computer and terminal.The specific framework of data center refers to Fig. 3, because Need to manage and handle all electricity business sections, the various types of signal data at station in whole railroad for this data center, because This needs larger storage, computing capability.Data gather computer is mainly used in receiving the data that each Railway Bureau sends over, the machine Deployment data acquisition subsystem, data prediction subsystem are needed on device, after the responsible signal data pretreatment that will be collected It is transferred to data center;Terminal is connected to show the running status of whole company, including live signal with data center Displaying, signal data inquiry and fault alarm.

B) Railway Bureau

Railway Bureau needs one small-sized data center of deployment, data gather computer and terminal.Data center of Railway Bureau uses In the signal data that electric section of being engaged in is uploaded in receiving and processing administration, it is necessary to dispose including distributed file system, parallel processing frame The components such as structure, real-time data analysis;Data gather computer is mainly used in gathering the various types of signal data that Railway Bureau's aspect is produced, and right Data are transferred to data center after being pre-processed.Terminal be used for show, inquire about and early warning Railway Bureau in various types of signal data.

C) electricity business section

Electricity business section and station need not dispose data center, it is only necessary to data gather computer and terminal.These harvesters except Outside function with signal acquisition, the also ability with data prediction, real-time data analysis, data storage and data transfer. After data gather computer gathered data, data are pre-processed and analyzed in real time, analysis result is carried out by local terminal Displaying, and partial analysis result and signal data are transferred in data center of Railway Bureau.The data collecting system of electricity business section connects Receive each station is transmitted in area under control signal data and analysis result.Terminal be used for show and inquire about various signal datas and Analysis result.

D) station

Station only needs to deployment data gather computer and terminal as electricity business section, and its function is also similar, simply the number at station Only need to gather the signal data at this station according to harvester, and these data are subjected to pretreatment and analysis, and according to need in real time Seek the harvester that partial data and analysis result are transferred to electricity business section.Terminal is mainly used in various signal datas in displaying station And the real-time analysis situation of data.

Fig. 4 is the Organization Chart of data center, mainly includes Yunan County's holostrome, basic resource layer, virtualization layer, storage system Layer, computing engines layer, component layer and web cluster.Yunan County's holostrome is ensured by data backup, data convert and access control The safety of system.The basic resource layer building hardware platform of data center, the part can be multiplexed existing server apparatus. Virtualization layer realizes virtualization to hardware by virtualization software, shields the hardware differences of bottom, forms a High Availabitity, can Flexible, expansible cluster.Data storage layer is made up of distributed file system, columnar database, relational database, respectively For storing unstructured, semi-structured and structuring data.Computing engines layer includes off-line data processing framework MapReduce, real-time distributed processing framework spark and data movement engine sqoop and underlying resource management system yarn. The task that Yarn is used to be submitted according to user is used to handle the task to distribute storage resource and computing resource.Component layer is specific Handle the component of miscellaneous service, including data management component, components of data analysis, fault diagnosis component etc..Various data acquisitions, The result of analysis is required for releasing by B/S or C/S framework, and Web clusters are the service hairs based on load balancing Layer of cloth.

The synthesis operational system based on cloud computing of the present invention, can be integrated with other processing platforms.Such as Fig. 5 institutes A kind of Digital Railway unified platform is shown as, is built by the way of " general purpose module " and " business plug-in unit ".Wherein " data center With big data management & data mining analysis platform " it can be realized based in the way of general purpose module one present invention, it is that data are adopted The core of collection, storage and data processing, and with equipment complex monitor supervision platform, shipping platform, operation platform, scheduling collaborative platform etc. Integrate, be collectively forming Digital Railway unified platform.

The workflow of platform is described in detail below by a specific example:

Fig. 6 is the flow chart that a 25Hz phase-sensitive track circuits distinguish indoor and outdoor failure.Track circuit failure reason point Analysis is a classification problem, is especially suitable for carrying out analysis mining using the method for data mining.Relative signal acquisition is come From in station equipment.It is the model training stage first, existing data is manually marked, it is faulty to mark out those Data.Material is thus formed a training set, these data sets are trained using data mining subsystem, Bayes is selected Grader obtains corresponding various model parameters as training pattern.Then these disaggregated models are disposed into data AT STATION In harvester, data gather computer receives the Monitoring Data from each monitoring device first, and then these data are located in advance Reason, obtains the characteristic vector for analysis.Real-time data analysis component is divided real time data using obtained disaggregated model Analysis is calculated, and obtains system whether faulty at present.Then the result of accident analysis is shown by display systems, and by phase The analysis result and feature of pass are transferred in electricity business section, Railway Bureau and railway center, for the analysis in higher level.

After data prediction and feature selecting, the feature extraction result of completion is:

Junction box is by terminal voltage Throw cable terminal outside voltage away Sending end voltage Data acquisition time is stabbed 25.00 25.00 25.00 521365 24.00 24.00 25.00 521365 27.00 27.00 27.00 521365 0.00 0.00 0.00 521365

For the purpose of simplifying the description, the normal voltage value of three test points is disposed as 25v in upper table.The type of failure is divided into three Class:

(1) fault-free;

(2) failure is indoors;

(3) failure is in outdoor;

(4) indoor short circuit;

(5) indoor open circuit;

Above-mentioned data are subjected to vectorization, calculated in order to improve Bayes classifier:

Instance data position:

0 1:25.0 2:25.0 3:25.0

0 1:25.0 2:25.0 3:25.0

0 1:25.0 2:25.0 3:25.0

4 1:30.0 2:25.0 3:25.0

4 1:30.0 2:35.0 3:20.0

1 1:0.0 2:0.0 3:0.0

2 1:0.0 2:25.0 3:25.0

3 1:0.0 2:50.0 3:25.0

3 1:15.0 2:50.0 3:25.0

1 1:0.0 2:0.0 3:0.0

1 1:0.0 2:0.0 3:0.0

The type of one column of figure representing fault of foremost:

● 0 represents no failure

● 1 represents failure indoors

● 2 represent failure in outdoor

● 3 represent indoor short circuit

● 4 represent indoor open circuit

The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, the ordinary skill of this area Personnel can modify or equivalent substitution to technical scheme, without departing from the spirit and scope of the present invention, this The protection domain of invention should be to be defined described in claim.

Claims (7)

1. a kind of track traffic signal based on cloud computing integrates operational system, it is characterised in that including:
Data gather computer, for acquisition trajectory data traffic signal;
Data center, connects the data gather computer, including Yunan County's holostrome, basic resource layer, virtualization layer, data storage layer, Computing engines layer, component layer and Web clusters;Wherein, Yunan County's holostrome is ensured by data backup, data convert and access control The safety of system;Basic resource layer is the hardware platform of data center;Virtualization layer is realized to hardware by virtualization software Virtualization, shields the hardware differences of bottom;Data storage layer includes distributed file system, columnar database, relational database, It is respectively used to unstructured, semi-structured and structuring the data of storage;Computing engines layer handles frame using cloud computing;Component layer It is the component of specific processing miscellaneous service;Web clusters are the service issue layers based on load balancing;
Terminal, connects the data gather computer and the data center, for display data analysis result;
Said system is using railroad, Railway Bureau, electricity business section, the level Four pattern at station, in railroad, difference portion of Railway Bureau Data center, data gather computer and terminal are affixed one's name to, data gather computer and terminal are only disposed at electricity business section, station;AT STATION, electricity business Section, Railway Bureau, railroad dispose real-time data analysis component, wherein station, the real-time data analysis assembly portion of electricity business section It is deployed in data gather computer, Railway Bureau, the real-time data analysis deployment of components of railroad are in data center;
The data gather computer is stored in local system for the local signal data collected, then on upper strata with metadata Form manage the storage organization of data, upper strata also stores the metadata of lower floor, in data in addition to the data for storing this layer During calculating, except to upper layer transport, lower floor's analysis result is gone back into the foundation that hop initial data is analyzed as upper layer data, So as to implementation level type data distribution formula storage mode;
Data to railroad, Railway Bureau, electricity business section and station storage set up index, fast according to the index when breaking down Speed obtains the various types of signal data of relevant time period and position, in order to be rapidly performed by accident analysis, positioning and solve;It is described Setting up the method for index is:Set up and indexed according to the data of storage at each station first, then pass the index file at station Defeated to arrive electricity business section, electricity business section sets up secondary index to the data locally gathered and the station index file received, then by rope Draw and transmit to the data center of Railway Bureau, the data center of Railway Bureau sets up three level list to the Various types of data of reception, then will Index is transmitted to the data center of railroad, and the data center of railroad sets up level Four index to the Various types of data of reception, Form a railroad, Railway Bureau, electricity business section, the level Four at station index;By the strategy realization pair for being classified storage and index The storage demand of ever-increasing signal data, and realize the quick positioning to failure by four layer index structures.
2. the system as claimed in claim 1, it is characterised in that:The computing engines layer includes off-line data processing framework MapReduce, real-time distributed processing framework spark and data movement engine sqoop and underlying resource management system yarn, The task that Yarn is used to be submitted according to user is used to handle the task to distribute storage resource and computing resource.
3. the system as claimed in claim 1, it is characterised in that:The component layer includes data management component, data analysis group Part, fault diagnosis component.
4. a kind of track traffic signal based on cloud computing of system described in use claim 1 integrates O&M method, its step Including:
1) using railroad, Railway Bureau, electricity business section, the level Four pattern at station, data are disposed respectively in railroad, Railway Bureau Center, data gather computer and terminal, data gather computer and terminal are only disposed at electricity business section, station;AT STATION, electricity business section, railway Office, railroad dispose real-time data analysis component, and wherein station, the real-time data analysis deployment of components of electricity business section are in data In harvester, Railway Bureau, the real-time data analysis deployment of components of railroad are in data center;
2) data gather computer at station gathers the real time data at station, using the real-time data analysis component in local disposition to letter Number is pre-processed, feature extraction and feature selecting, and then characteristic is analyzed in real time using analysis model, obtained To the current running status of signal system, and the section that corresponding data transfer to electricity can be engaged in;
3) data gather computer of electricity business section receives the data transmitted from each internal station, utilizes the reality of local disposition When components of data analysis the data at the different stations of reception are analyzed, obtain the analysis result of whole electricity business section, and will point Analysis result is transmitted to the data gather computer of Railway Bureau;
4) data gather computer of Railway Bureau receives the data from each electricity business section, and by data conversion storage to data center, then Data are handled and analyzed using the real-time data analysis component of data center deployment, and by analysis result and Partial Feature It is transferred to the data gather computer of railroad;
5) data gather computer of railroad receives the data that Railway Bureau sends over, and by data conversion storage to data center, it is sharp Analyzed and processed with the real-time data analysis component of data center, obtain the global analysis result of railroad;
6) using the terminal for being deployed in railroad, Railway Bureau, electricity business section and station, the displaying of data results is carried out;
7) data to railroad, Railway Bureau, electricity business section and station storage set up index, when breaking down according to the index Quick obtaining relevant time period and the various types of signal data of position, in order to be rapidly performed by accident analysis, positioning and solve;Institute Stating the method for foundation index is:Set up and indexed according to the data of storage at each station first, then by the index file at station Electricity business section is transferred to, electricity business section sets up secondary index to the data locally gathered and the station index file received, then will Index is transmitted to the data center of Railway Bureau, and the data center of Railway Bureau sets up three level list to the Various types of data of reception, then Index is transmitted to the data center of railroad, the data center of railroad sets up level Four rope to the Various types of data of reception Draw, form a railroad, Railway Bureau, electricity business section, the level Four at station index.
5. method as claimed in claim 4, it is characterised in that in the level Four index:
The numbering of monitoring device, data block numbering, signal data exist in the numbering of station data directory structure including workshop, workshop The original position of data block, the size of signal data;The index data at station is generated in data storage, and is stored in local In device, when station terminal carries out data query, index file is read into local machine first, then according to data block, skew Amount goes to read real data;
Electricity business segment data index structure includes the index of local data index and station index data;
Railway Bureau's data directory structure includes the index that local data indexes segment index data of being engaged in electricity in area under control;
Railroad's data directory structure includes local data index and the index of each Railway Bureau's index data.
6. method as claimed in claim 4, it is characterised in that:The data gather computer is for the local signal data that collects It is stored in local system, upper strata manages the data of lower floor by metadata.
7. method as claimed in claim 4, it is characterised in that:Step 2) in station the data gather computer signal that passes through station Machine of standing obtains signal data, and whether the signal station machine at station needs to be transmitted by existing rule judgment, so as to signal Data are compressed processing;The data gather computer at station and the signal station machine at station set up socket connections, the signal station at station Machine compression is handled after data transfer to data gather computer, data gather computer is gone back using the existing rule to data It is former.
CN201410321145.1A 2014-07-07 2014-07-07 A kind of track traffic signal based on cloud computing integrates O&M method and system CN104077552B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410321145.1A CN104077552B (en) 2014-07-07 2014-07-07 A kind of track traffic signal based on cloud computing integrates O&M method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410321145.1A CN104077552B (en) 2014-07-07 2014-07-07 A kind of track traffic signal based on cloud computing integrates O&M method and system
PCT/CN2015/075007 WO2016004775A1 (en) 2014-07-07 2015-03-25 Method and system for integrated operation and maintenance of rail transportation signals based on cloud computing

Publications (2)

Publication Number Publication Date
CN104077552A CN104077552A (en) 2014-10-01
CN104077552B true CN104077552B (en) 2017-08-25

Family

ID=51598802

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410321145.1A CN104077552B (en) 2014-07-07 2014-07-07 A kind of track traffic signal based on cloud computing integrates O&M method and system

Country Status (2)

Country Link
CN (1) CN104077552B (en)
WO (1) WO2016004775A1 (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104077552B (en) * 2014-07-07 2017-08-25 北京泰乐德信息技术有限公司 A kind of track traffic signal based on cloud computing integrates O&M method and system
US10282676B2 (en) * 2014-10-06 2019-05-07 Fisher-Rosemount Systems, Inc. Automatic signal processing-based learning in a process plant
CN104504956A (en) * 2014-12-22 2015-04-08 中国神华能源股份有限公司 Railway electrical training system based on virtual reality
CN104908783B (en) * 2015-05-27 2017-01-18 中国铁路总公司 System architecture of comprehensive monitoring and maintaining system for railway electricity
CN104890702A (en) * 2015-05-27 2015-09-09 中国铁路总公司 Vehicle-ground railway signal comprehensive analyzing and monitoring system
CN105574593B (en) * 2015-12-18 2020-05-05 中南大学 Track state static detection and control system and method based on cloud computing and big data
CN105676842B (en) * 2016-03-14 2019-06-18 中国铁路总公司 A kind of high-speed rail train control on board equipment method for diagnosing faults
CN106250429A (en) * 2016-07-26 2016-12-21 浪潮软件股份有限公司 A kind of data pick-up method based on sqoop
CN106341467B (en) * 2016-08-30 2019-11-29 国网江苏省电力公司电力科学研究院 Power information based on big data parallel computation acquires equipment state analysis method
CN108075917A (en) * 2016-11-18 2018-05-25 中兴通讯股份有限公司 The management method and system of a kind of data center
CN107316158A (en) * 2017-07-05 2017-11-03 云能服(北京)科技有限公司 A kind of multidimensional work data treating method and apparatus
CN110032090A (en) * 2019-04-15 2019-07-19 深圳众维轨道交通科技发展有限公司 A method of realizing that tramcar is remotely monitored and authorized based on BIM+AI
CN110182244A (en) * 2019-04-23 2019-08-30 深圳众维轨道交通科技发展有限公司 A kind of tramcar cloud platform based on cloud computing and AI intelligence

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101917237A (en) * 2010-07-27 2010-12-15 北京全路通信信号研究设计院 Railway signal monitoring method and system
WO2012119879A1 (en) * 2011-03-07 2012-09-13 Siemens Aktiengesellschaft Railway control system
CN103338261A (en) * 2013-07-04 2013-10-02 北京泰乐德信息技术有限公司 Storage and processing method and system of rail transit monitoring data
CN103391185A (en) * 2013-08-12 2013-11-13 北京泰乐德信息技术有限公司 Cloud security storage and processing method and system for rail transit monitoring data

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103264717B (en) * 2013-05-21 2015-12-02 北京泰乐德信息技术有限公司 A kind of track traffic synthetic monitoring and scheduling is worked in coordination with and O&M information system
CN104077552B (en) * 2014-07-07 2017-08-25 北京泰乐德信息技术有限公司 A kind of track traffic signal based on cloud computing integrates O&M method and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101917237A (en) * 2010-07-27 2010-12-15 北京全路通信信号研究设计院 Railway signal monitoring method and system
WO2012119879A1 (en) * 2011-03-07 2012-09-13 Siemens Aktiengesellschaft Railway control system
CN103338261A (en) * 2013-07-04 2013-10-02 北京泰乐德信息技术有限公司 Storage and processing method and system of rail transit monitoring data
CN103391185A (en) * 2013-08-12 2013-11-13 北京泰乐德信息技术有限公司 Cloud security storage and processing method and system for rail transit monitoring data

Also Published As

Publication number Publication date
WO2016004775A1 (en) 2016-01-14
CN104077552A (en) 2014-10-01

Similar Documents

Publication Publication Date Title
CN103812052B (en) A kind of for without the centralized monitoring system of man-machine polling transmission line and monitoring method
CN103941745B (en) For without the mobile substation of man-machine polling transmission line and method of work
CN103116825B (en) Intelligent city management
CN105183609B (en) A kind of real-time monitoring system for being applied to software system and method
CN103888287B (en) Information systemintegration O&M monitor service early warning platform
JP6100160B2 (en) Intelligent network
CN101789630B (en) Monitoring system uniform platform of on-line monitoring and auxiliary equipment of transformer substation
Martínez et al. Human-caused wildfire risk rating for prevention planning in Spain
CN102193555B (en) Panoramic-state monitoring system for centralized control centers
CN102761174B (en) Fault intelligent diagnosis and analysis system and method based on dispatching integrated data platform
CN104503399B (en) The regimental Wind turbines status monitoring of one kind collection and fault diagnosis platform
CN102149103B (en) Network optimizing system and method
CN103593361B (en) Movement space-time trajectory analysis method in sense network environment
CN105302098B (en) A kind of railcar interoperability maintenance support platform and its building method based on IETM
CN105049814B (en) A kind of method and system of monitoring analysis personnel motion trail
CN105407159B (en) Logistics transportation position service system
CN103338261B (en) The storage of a kind of track traffic Monitoring Data and processing method and system
CN102929903B (en) Rapid video retrieval method based on layered structuralized description of video information
CN102882969B (en) A kind of safety production cloud service platform of industrial and mining enterprises
CN104944240A (en) Elevator equipment state monitoring system based on large data technology
CN105553957A (en) Network safety situation awareness early-warning method and system based big data
CN106165345A (en) Mark is for dissolving the failture evacuation option of network failure
CN104616205B (en) A kind of operation states of electric power system monitoring method based on distributed information log analysis
Deng et al. A research on subway physical vulnerability based on network theory and FMECA
CN102407867B (en) Urban rail transit signal maintenance supporting system

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
GR01 Patent grant