CN207552827U - A kind of rail system safe condition comprehensive monitoring and intelligent analysis system - Google Patents
A kind of rail system safe condition comprehensive monitoring and intelligent analysis system Download PDFInfo
- Publication number
- CN207552827U CN207552827U CN201721629476.7U CN201721629476U CN207552827U CN 207552827 U CN207552827 U CN 207552827U CN 201721629476 U CN201721629476 U CN 201721629476U CN 207552827 U CN207552827 U CN 207552827U
- Authority
- CN
- China
- Prior art keywords
- rail
- data
- monitoring
- stress
- temperature
- 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.)
- Active
Links
Landscapes
- Force Measurement Appropriate To Specific Purposes (AREA)
- Train Traffic Observation, Control, And Security (AREA)
- Machines For Laying And Maintaining Railways (AREA)
Abstract
The utility model discloses a kind of rail system safe condition comprehensive monitoring and intelligent analysis systems, three kinds of sensor monitoring technologies of the system globe area, based on low-frequency datas such as Fiber Bragg Grating technology monitoring of structures temperature, using high-frequency datas such as amendment ess-strain technical monitoring rail horizontal stroke, vertical stress, the sensitive structure of contact measurement is difficult to for point tongue etc., using video-aware technology observation structure large deformation and surface state, round-the-clock system monitoring of the railway rail system from appearance to inherence, from high frequency to low frequency is formed.Convergence analysis is carried out by the multi-source data to acquisition, effectively track condition can be assessed, diagnosed and predicted, and then realize the timely early warning of rail safety state.The utility model point layout is reasonable, and the automatic degree of monitoring process is high, accurate to line state evaluation, timely to abnormal conditions early warning, realizes the safely controllable of railway rail system military service, safety, even running for train provide reliable guarantee.
Description
Technical field
The utility model is related to railway engineering monitoring system, more particularly to a kind of rail system safe condition comprehensive monitoring
And intelligent analysis system.
Background technology
Backbone of the railway as integrated transportation system has important impetus to national economy and social development.With
The rapid development of China's economy, the rail transport using high-speed railway as representative develops rapidly, runs quality and construction scale is equal
Reach world lead level.But with the propulsion of high speed railway construction tide, the safe military service problem day of circuit and infrastructure
Benefit highlights.
The operating practice of high-speed railway shows under prolonged and repeated load action, using circuit, bridge tunnel as representative crucial base
Microcosmic hurt, the deterioration of the function of main structural components, infrastructure state and the performance development of Infrastructure are inevitable.Line
As the key structure contacted with wheel, itself stress deformation under train and temperature load is extremely complex on road, once hair
Situations such as raw rail expansion, broken rail, will result directly in train and the major accidents such as overturn, slides down.Roadbed, bridge tunnel structure are as line tower foundation
Important composition, be widely used in all kinds of track engineerings, stability directly affects the safety of line project, smooth-going.Its Road
In the case where construction quality is kept under strict control, temperature effect, train effect do not protrude for base, tunnel structure, line construction are influenced relatively small.
And bridge influences each other with track on stress deformation, forms a relationship complexity and the multilayered structure acted on by multi- scenarios method,
Bridge deformation can directly result in circuit stress destruction, deformation is seriously transfinited, and bring huge security risk.
At present, the domestic monitoring for rail system, mostly using displacement observation stake, collimator, chord_line method and thermometer into
Row observation and measurement, measuring accuracy is poor, and for test content based on Static State Index, detection cycle is also longer.In addition, although high ironworker
There is the detection method for ensureing rail smooth, confirming line security of complete set in business department, is such as manually patrolled in Window time
Confirmation train etc. is started, but can not grasp the safety clothes of track infrastructure in real time in road before periodically starting track checking car, daily operation
Labour state often can not find and renovate in time especially when circuit surface faces sudden disease.Meanwhile existing monitoring bases oneself upon section more
It grinds, focuses on regularity exploring, pay close attention to the crucial part of line project, such as point tongue, expansion and cotraction regulator, monitoring object is single more,
The considerations of system is lacked to non-fragment orbit, infrastructure.
Therefore, to ensure circuit operation security, it is necessary to from systematic entirety, unified monitoring content and method, structure
Integrated monitoring platform monitors the safe service state of track infrastructure in real time, and realizes the multi-source fusion of monitoring data, intelligence point
Analysis, assessment prediction, and then rational maintenance suggestion is proposed to the practical operation of rail system.The utility model proposes tracks
System safe condition comprehensive monitoring and intelligent analysis system, compensate for prior art deficiency well, and monitoring content fully considers
Basic military service behavior under line, used monitoring means can capture low frequency and high-frequency information simultaneously, and can effectively grasp structure
External deformation changes with internal stress, and various kinds of sensors easy construction, installation is firm, long-time stability are high.The utility model can
Realization alarms rapidly to demblee form disease, to rapid evolution and slowly development-oriented plant disease prevention early warning, has ensured circuit significantly
Operation security.
Utility model content
The utility model provides a kind of rail system safe condition comprehensive monitoring and intelligent analysis system, it is therefore intended that solves
Because prior art deficiency bring the problem of can not grasping the safe service state of rail system in real time, and excavate monitoring number in depth
On the basis of, to being likely to occur unsafe condition look-ahead, early warning.
In order to achieve the above objectives, the utility model is first according to a kind of high-speed railway of patent and urban track traffic track knot
Seamless turnout sets on the high-speed railway overhead station of structure Experimental mimic system (ZL200910242417.8), in length and breadth vertical coupled
Meter method (ZL200910236546.6) and a kind of seamless turnout structural system on bridge and its method of dynamic analysis
(ZL200910236922.1), monitoring content and point layout position are determined.
On the basis of above-mentioned determining monitoring position, the utility model adopts the following technical solution：
A kind of rail system safe condition comprehensive monitoring and intelligent analysis system, the system work step include：
S1, the low-frequency datas such as rail system stress, temperature and thin tail sheep are acquired using fiber-optic grating sensor；
S2, rail system big displacement data are acquired using video surveillance；
The high-frequency datas such as S3, the vertical stress that patch acquisition rail is spent using stress and lateral stress carry out data；
S4, step S1 to the S3 orbital datas collected are analyzed and processed, using BP neural network model and more
First linear regression under conventional sense track structure stress, displacement predicts；
S5, code requirement and statistical result given threshold are combined to the possible destruction occurred to a certain degree with cluster analysis
Upper carry out early warning.
Preferably, the step S1 includes the stress acquisition of rail system using fiber-optic grating sensor：
Optical fiber optical grating stress sensor is pasted onto to the surface of the rail of rail system, track plates, base board, acquires steel
The stress data of rail, track plates and base board.
Preferably, the step S1 includes the temperature acquisition of rail system using fiber-optic grating sensor：
Track plates are punched, temperature sensor is arranged in hole and do encapsulation process, measure track plate temperature
Position is included in plate, edges of boards and plate angle；
Preferably, it according to local track plates reinforcing bar layout drawing, is punched using at the injected hole in track plate, places temperature
Spend sensor, measurement plate medium temperature gradient；
Temperature sensor is placed using punching between two sleeper of track plates to measure edges of boards temperature；
Plate angle temperature is measured using placing temperature sensor away from punching at edges of boards 150mm and 70mm.
Preferably, the step S1 includes the thin tail sheep data acquisition of rail system using fiber-optic grating sensor：
It is punched in the track plates appropriate location of rail foot, the fixing end of displacement sensor is mounted on track plates, then
Fiber grating displacement sensor mobile terminal is mounted on rail foot using fixture block is installed, adjusts rope capacity and fixation,
It is made to meet sensor displacement range.
Preferably, the step S2 includes：
Hollow camera mounting rod is fixed on guardrail using fixture block, camera is fixed on camera mounting rod
Transmission line is connect with upper level transmission device from the hollow space of camera mounting rod along going out by top.
Preferably, the step S3 includes：
The high-frequency datas such as vertical stress and the lateral stress of patch acquisition rail are spent using stress, and paste compensation on the steel plate
Piece realizes temperature self-compensation.
Preferably, the step S4 to the orbital data collected carry out analyzing and processing prediction include：Using one-dimensional fast
Fast Fourier interpolation method is increased 3 times of sampling to temperature forecast data and establishes BP neural network model, monitored with continuous 48 temperature
Data are input neuron, and repetition training and prediction are carried out to indexs such as rail stress, tongue displacement, rail temperature.Establish polynary line
Property regression model, using the rail temperature time series of continuous 12 hours as independent variable, by the Fitting Calculation to rail stress, tongue position
The key indexes such as shifting, rail temperature are predicted.Wherein BP neural network predicts conventional sense lower rail temperature, stress, displacement,
It is gentle that the conventional sense refers to temperature Change, the normal state of amplitude；Multivariate regression models, under extreme condition rail temperature,
Situations such as stress, displacement predicted, the extreme condition includes environment suddenly cold and hot and continuous high temperature or low temperature.
Preferably, the step S5 includes comparing rail system current state information and preset alarm threshold value
It is right, alarm is sent out if transfiniting；If the difference of current measurement value and the arithmetic mean of instantaneous value of this data history data is more than 3 times
Historical data root-mean-square-deviation, then judge the measured value for bad value, be not involved in the processing and analysis of data；If a period of time
It is interior continuous or repeatedly bad value data occur, then judge that monitoring device occurs extremely.
Wherein alarm threshold value setting is divided into following three ranks：
Using in ballastless track of high-speed railway line rule to the limit value of tongue displacement and stock rail displacement as level-one
Alarm threshold value, data over run carry out level-one early warning；
The arithmetic mean of instantaneous value of remaining each monitoring item historical data in addition to the currently monitored value is added, subtracts three times root mean square work
For secondary alarm threshold value, data over run carries out secondary alarm prompting；
The data of the collected different attribute of sample data each time point are formed into the space vector with multiple parameters,
Cluster analysis is carried out in the case where cluster amount is identical with vectorial latitude, sets three-level alarm threshold value, data over run carries out three-level report
Alert prompting.
According to above-mentioned space vector, our setup parameters are vectorial Its time series is [Mo1,Mo2,Mo3..., Mok..., Mon]T(k=1,2,
3 ...),
T in formulaTemperatureTo measure temperature, TRail temperature 1To measure left side rail temperature, TRail temperature 2To measure right side rail temperature, SBeam-endsTo measure beam-ends
Locate rail stress, SRailway frogTo measure rail stress at railway frog, SHeart railTo measure rail stress at heart rail, SHold-down supportTo measure fixed branch
Rail stress at seat, SThe point of switchTo measure rail stress at the point of switch, DThe point of switchTo measure displacement at the point of switch
In view of left and right stock rail rail temperature is not exactly the same, T' is enabledRail temperature 2=TRail temperature 2-TRail temperature 1,
T'Rail temperature 2Left and right rail temperature difference value, TRail temperature 1For left rail rail temperature, TRail temperature 2For right rail rail temperature；
In view of the rail stress of each point and rail temperature are closely related, for the factor is avoided to repeat to consider impact analysis as a result, enabling
S'i=εi=Si-ai-biTRail temperature 1,
In formula, SiFor the rail stress of different location, εiIt is the factor in regression model in addition to rail temperature to the shadow of stress
It rings；ai,biRespectively stress is to regression constant and regression coefficient in rail temperature linear regression correlation analysis, S'iFor rail stress
Gap between predicted value and measured value.
In view of the steel rail displacement of each point also has certain correlation with rail temperature, D' is enabledi=εi=Di-ai-biTRail temperature,
In formula, DiFor the steel rail displacement of different location, εiIt is the factor in regression model in addition to rail temperature to the shadow of displacement
It rings, ai,biRespectively displacement is to slope regression constant and regression coefficient in rail temperature Linear correlative analysis, D'iIt is pre- for steel rail displacement
Gap between measured value and measured value.
The above-mentioned factor in addition to rail temperature refer to sensor it is unstable due to itself during operation caused by random error,
And sensor inaccurate this kind of systematic error of existing precision etc. in itself.
With reference to three rank early warning, can the more rigorous safety and steady operation for train provide safeguard.
The utility model provides a kind of rail system safe condition comprehensive monitoring and intelligent analysis system, the system include：
Data acquisition unit carries out data for the temperature of acquisition trajectory system, stress, displacement etc.；
Database module, for storing the orbital data of data acquisition unit acquisition；
Data processing and inversion module carries out processing analysis for transferring orbital data from database module, obtains rail
Road current state information；
Predicting unit, for predicting the development tendency of following week age track force and deformation state；
Alarm unit, for the alarm threshold value of track current state information and systemic presupposition to be compared, if analysis knot
Structure then sends out alarm beyond the threshold value of setting；
Power supply unit, for providing electric power support for the unit module in system in addition to fiber-optic grating sensor.
Preferably, the data acquisition unit includes：
Fiber-optic grating sensor carries out rail system the acquisition of the low-frequency datas such as stress, thin tail sheep and temperature；
Video sensor carries out point tongue etc. using video monitoring the video data acquiring of telescopic displacement；
Stress flower patch, carries out rail system the acquisition of the high-frequency datas such as vertical stress and lateral stress.
Preferably, the predicting unit includes：
BP neural network prediction module utilizes the key indexes numbers such as BP neural network model prediction track structure stress deformation
According to variation.
Multiple linear regression prediction module is referred to using keys such as multiple linear regression model predicted orbit structure stress deformations
Mark the variation of data.
Preferably, the alarm unit includes：
Threshold value comparing module for data results and preset threshold value to be compared, and is tied according to comparison
Fruit sends out early warning instruction；
Alert data library module, for storing comparing result.
Preferably, which further includes：
Filter module, the video data for being obtained to track big displacement data collecting module collected are filtered except dry place
Reason；
Compensating module compensates for the error caused by track external environment, so that high-frequency type orbital data is adopted
Collection module accurately acquires track stress data.
Picture recognition module for carrying out image identification to tongue picture, obtains the telescopic displacement of tongue..
The beneficial effects of the utility model are as follows：
The utility model is monitored the locations of structures difference of railway rail system, is divided according to the characteristics of each monitoring position
Fiber Bragg Grating technology, video-aware technology and the comprehensively monitoring mode for correcting stress-strain technology are not employed, are formd to rail
Road structure is from external-to-internal, and from part to entirety, Hybrid Decision-making monitoring system from low to high is realized to track
The long-term safety of system monitors in real time.Automatic collection is carried out by the data to railway rail system and processing is analyzed, Neng Goushi
When monitor rail system safety military service behavior, while prediction and mathematical model of decision by establishing rail system state, according to
The stress and deformation measurement data of track and its component carry out prediction and warning to the destruction that may occur, and are the safety and steady of train
Operation provides safeguard.
Description of the drawings
Specific embodiment of the present utility model is described in further detail below in conjunction with the accompanying drawings；
Fig. 1 shows the schematic diagram of rail system safe condition comprehensive monitoring and intelligent analysis method；
Fig. 2 shows the schematic diagrames of rail stress sensor installation procedure in the present embodiment；
Fig. 3 shows the schematic diagram that steel rail displacement sensor is installed in the present embodiment；
Fig. 4 shows to arrange the schematic diagram of temperature sensor punch position in the present embodiment；
Fig. 5 shows the schematic diagram that temperature sensor gradient is arranged in the present embodiment mesoporous；
Fig. 6 shows the present embodiment middle orbit plate, screed, base board temperature gradient point layout；
Fig. 7 shows camera and the schematic diagram of monitoring object relative position in the present embodiment；
Fig. 8 shows the schematic diagram of vertical force test road and bridge connection mode in the present embodiment；
Fig. 9 shows the schematic diagram of cross force test road and bridge connection mode in the present embodiment；
Figure 10 shows a kind of rail system safe condition comprehensive monitoring and intelligent analysis system schematic diagram；
Figure 11 shows the schematic diagram of the acquisition transmission mode of monitoring data in the present embodiment.
Specific embodiment
In order to illustrate more clearly of the utility model, the utility model is done into one with reference to preferred embodiments and drawings
The explanation of step.Similar component is indicated with identical reference numeral in attached drawing.It will be appreciated by those skilled in the art that below
Specifically described content is illustrative and be not restrictive, and should not limit the scope of protection of the utility model with this.
Specifically, it is described with reference to the drawings.Attached drawing 1-11 shows the embodiment and effect of the utility model each section.
As shown in Figure 1, the utility model is on the basis of monitoring content and point layout position is determined, by railroad track system
Safe military service status information of uniting is divided into three kinds of data types, means of different is taken to be monitored respectively.The first be track stress,
The low frequencies orbital data such as micro-displacement and temperature；Second is that tongue telescopic displacement, track switch and bridge integrality etc. can rely on
The orbital data of visual recognition；The third be rail hang down, the high frequencies orbital data such as lateral stress.Base is acquired in real time in multi-source data
On plinth, established rail system State Forecasting Model and intelligent analysis system are utilized, monitoring data are handled and are analyzed,
Comprehensive assessment is carried out to circuit service state, realizes the prediction and early warning of automation.
Embodiment 1
This gives a detailed process using the utility model monitoring rail system low-frequency data.To frequency
Relatively low orbital data, is acquired using Fiber Bragg Grating technology.Fiber grating is made of the light sensitivity using optical fiber.Work as light
When temperature, stress, strain or the other physical quantitys of fine grating local environment change, the period of grating or fiber core refractive index will
It changes, so as to which the wavelength for making reflected light changes, by measuring the variation of reflected light wavelength before and after physical quantity variation, just
The situation of change of measured physical quantity can be obtained.The characteristics of using fiber grating, the utility model provide and high-speed iron rail
Optical fiber optical grating stress, displacement and the temperature sensor installation method that road structure matches.
As shown in Fig. 2, for rail stress sensor installation procedure figure.Strain gauge rail to be measured is chosen first
Position and surface of polishing in flange of rail installation folder block, strain gauge are pasted onto at steel rail grinding, welding stress sensor light
Fibre, mount stress sensor protective cover on the web of the rail, while water-proofing treatment is done to protective cover grating connector.By by strain gauge
The surface of rail, track plates, base board is pasted onto, grating deformation can be caused down in load actions such as train, temperature, so as to reflect
The stress variation of rail, track plates and base board.
As shown in figure 3, the displacement sensor installation diagram for rail, it is preferable that used displacement sensor measurement range is
Sensor activity end by the way that the fixing end of displacement sensor is mounted on track plate surface, is mounted on rail foot by 50mm, and
Two parts are connected with the traction steel wire through insulation processing.When relative displacement occurs for the two, grating deformation is driven through traction steel wire,
Thus the relative displacement of rail and track plates is monitored.Maximum distance for 50mm, (i.e. move by sensor maximum between terminal A and terminal B
Journey), the initial distance of terminal A to terminal B is set as 25mm, ensures that sensor terminal A can be to far from end along the axis of terminal B
Point B directions movement 25mm (at this time up to maximum traverse, sensor probe is fully extended), and terminal A can be with along the axis of terminal B
To close to terminal B direction movement 25mm (AB is closely connected at this time, and sensor probe is fully retracted), the range of ± 25mm is realized with this.
CD is traction steel wire, and C, E spacing are the spacing between instrument and track plates sleeper.
This installation method is equally applicable to the relative displacement between base board and bridge, bridge structure after appropriate adjustment
Monitoring.Since steel rail displacement sensor is mounted on rail foot, the high-speed cruising of train will not be had an impact, therefore can guarantee
The safe operation of circuit.And traditional displacement sensor installation site is on the outside of rail, although easy for installation, to traffic safety
Property produces a very large impact, and is not easy to practical driving.
Rail system temperature sensor installation method is, by the way that fiber-optical grating temperature sensor and monitoring object is closely connected tight
Gu using institute's geodesic structure under temperature change expand with heat and contract with cold drive grating deform, so as to monitor atmospheric temperature, rail temperature,
The data such as track plates temperature gradient and bridge temperature.It is as follows for the specific monitoring method of different objects：
(1) by temperature sensor exposure in air, temperature can be measured；
(2) temperature sensor is pasted onto rail waist, rail temperature can be measured；
(3) by the way that temperature sensor is embedded in track plates, track plate temperature can be measured.
In the present embodiment, in order to measure the temperature gradient of track plates, while analyze in the plate of track plates, edges of boards and plate angle
Temperature difference, when track plates punching is layouted, selection is in plate, edges of boards and plate angle position.
As shown in figure 4, since track plates reinforcing bar is intensive, when punching, will avoid reinforcing bar, with reference to local track slab steel during punching
Muscle layout drawing.Preferably, to carrying out temperature survey among track plates it when, selects to punch at injected hole；To the plate of track plates
Side carries out temperature survey, selects to punch between two sleepers；When carrying out temperature survey to the plate angle of track plates, select away from edges of boards
It is punched at 150mm and 70mm, close proximity to the position of plate angle.
It is punched on track plates first before measure track plate temperature gradient, then intercepted length is identical with hole depth determines
The sensor for being fixed with positioning iron wire in the corresponding position installation fixed temperature sensor of iron wire, is put into track plates by position iron wire
The cement mortar with track plates phase same material is poured into after in hole, finally does water-proofing treatment on surface.
As shown in figure 5, arrangement temperature sensor.Preferably, punching depth is 30cm, which can pass through track plates, sand
Pulp layer and base board, in order to each layer of progress temperature survey, temperature sensor is sent into hole, and root using iron wire is positioned
According to every layer of thickness, setting at least every layer of sensor measures every layer of temperature.Fig. 6 illustrates finally formed rail
Guidance tape, screed, base board temperature gradient measuring point horizontal layout situation.
Embodiment 2
This gives a detailed process using the utility model monitoring high-speed railway track switch tongue stroke.For
Point tongue etc. can not install the track sensitive structure of fiber-optic grating sensor, and the utility model connects using video-aware technology is non-
It is tactile, multiple dimensioned and the characteristics of contain much information, tripod head type camera is installed near the sensitive part of point tongue tip, is coordinated in point
Scale is pasted at rail center, realizes non-contact real-time monitoring.
Video identification is mainly at the analysis of acquisition and transmission, the video detection of centre and rear end including head end video information
Manage three links.The utility model is provided steady and audible using the point tongue displacement data acquisition module of video acquisition video camera
Vision signal；Denoising is filtered to video data by filter module again；Finally by data processing module, video is drawn
Abnormal conditions in face do target and track label.By analyzing video image, exclude inhuman in monitoring scene
Disturbing factor, activity condition of the accurate judgement target in video image.
As shown in fig. 7, the utility model utilizes video-aware technical monitoring tongue stroke in embodiment, while pass through cloud
Platform rotates, and grasps track switch, the isostructural integrality of bridge in real time.Video-aware idiographic flow is：
(1) by pasting scale on the stock rail web of the rail by tongue, point is shot using tripod head type camera fixed angle
Rail obtains the steady and audible vision signal with tongue and scale；
(2) denoising is filtered to video data by filter module；
(3) by data processing and inversion module, video pictures is identified, detect, are analyzed, obtain the point of switch pair
The scale label of position is answered, so as to be accurately judged to the dilatation of tongue.
In the present embodiment, due to consider high-speed railway safety requirements it is very high, do not allow camera be mounted on bridge top retaining wall with
It is interior, while there is no the positions with proper height of installation camera again for the actual conditions at scene.Therefore, pacified using camera
It fills bar camera is mounted on Bridge guardrail.Position above and below Bridge guardrail respectively sets a folder that can carry out displacement fine tuning
Block, for fixing and adjusting camera mounting rod.Mounting rod is hollow steel tube, in addition to support, fixed cradle head camera, inside
Space for threading, protect the complete of transmission cable.
Embodiment 3
This gives a detailed process using the utility model monitoring rail system high-frequency data.This practicality
It is novel based on correct stress-strain technology self compensation, self-correction, it is steady in a long-term the advantages of, with reference to the environmental condition at scene, pass through
Mount stress flower patch monitoring rail lateral stress, vertical stress.
As shown in figure 8, vertical force test strain rosette is pasted onto near rail natural axis, it is in longitudinally ± 45 DEG C with rail.It hangs down
Full-bridge is used to stress bridge.In the bridge by vertical stress, A, C correspond to control source, and it is defeated that B, D correspond to signal
Go out.In embodiment, stress flower patch pastes (60kg/m rail, neutral axis is apart from flange of rail 8.123cm) in pairs centered on neutral axis,
Strain rosette center spacing is 22cm.
As shown in figure 9, lateral stress test strain rosette is pasted onto flange of rail upper surface, it is in longitudinally ± 45 DEG C with rail.By
Into the bridge of lateral stress, A, C are control source, and B, D are exported for signal.Strain rosette to be at flange of rail edge 2.5cm
Center is pasted (60kg/m rail) in pairs, and strain rosette center spacing is 22cm.
The utility model realizes temperature self-compensation by pasting compensating plate on the steel plate of automatic telescopic.It is self-complementary through excess temperature
After repaying processing, vertical stress test bridge surveys strain only comprising vertical strain, and cross force test bridge surveys strain and there was only horizontal stroke
To strain.The calibration of wheel track vertical stress carries out quasi-static calibration using special equipment in example；Wheel-rail lateral force utilizes very heavy
Top and boosting frame field calibration.
Embodiment 4
As shown in Figure 10, the example shows a kind of rail system safe condition comprehensive monitoring and intelligent analysis systems, should
System composition includes：
Data acquisition unit, for data such as the temperature of acquisition trajectory system, stress, displacements,
The data acquisition unit includes sensor light fiber grating sensor, for acquisition trajectory system at low frequency
Stress, displacement and temperature data；Video sensor, for monitoring the displacement data of point tongue in rail system；Stress flower patch,
For the vertical stress in high frequency of rail in acquisition trajectory system and lateral stress data；
Database Unit, for storing the rail system data of data acquisition unit acquisition；
Data processing and inversion unit carries out processing analysis for transferring orbital data from Database Unit, obtains rail
Road system current state information；
The current state information obtained from data processing and inversion unit is carried out prediction processing, for pre- by predicting unit
Survey the development tendency of the track force and deformation state in one week.BP neural network model and multiple linear regression can be passed through
Two kinds of model realizes the prediction.
Skilled person will appreciate that predicting unit can be realized by software form, it can also be by firmware, such as burn
FPGA or the microprocessor of logical program are had to realize.
Alarm unit, for the alarm threshold value of track current state information and systemic presupposition to be compared, if analysis knot
Structure exceeds the threshold value of setting, then sends out alarm,
The alarm unit includes threshold value comparing module, for data results and preset threshold value to be carried out pair
Than, and early warning instruction is sent out according to comparing result；Alert data library module, for storing comparing result；
Power supply unit, for providing electric power support for the unit module in system in addition to fiber-optic grating sensor.
Preferably, the system also includes filter module, for what is obtained to track big displacement data collecting module collected
Video data is filtered except dry processing；Compensating module compensates for the error caused by track external environment, so that
High-frequency type orbital data acquisition module accurately acquires track stress data；
Picture recognition module for carrying out image identification to tongue picture, obtains the telescopic displacement of tongue.
This example uses the time series of temperature to become certainly during being predicted using existing monitoring data
Amount, without the temperature record using single point.This is because the uneven and track liter that the complexity of track structure, system are heated
The state changes such as the hysteresis quality of temperature, stress, the displacement of rail system are not completely the same with temperature, in the shape that certain moment shows
State is not only related to current load, but related with the accumulation of the load of the past period.Before the projection, profit is needed
Weather forecast data are increased with fast Fourier interpolation method and are sampled.Two kinds of specific predictions are as follows：
(1) BP neural network is predicted：Using the temperature record of continuous 12 hours as independent variable, it is contemplated that prediction mean square error is most
It is small, the parameters such as transmission function, training function, Inport And Outport Node quantity, frequency of training in prediction model are integrated
Analysis and optimization establish more feasible neural network prediction scheme, and to many indexs such as stress, displacement, temperature than choosing
Carry out look-ahead.
(2) multiple linear regression is predicted：Assuming that in monitoring item, sample
Y in formulatiIt is then the project to be predicted, in tiThe predicted value at moment can be rail temperature, plate temperature, rail stress etc.
Deng, the monitoring temperature that T refers to,The t referred toiThe monitoring temperature at moment,It is tiThe temperature monitoring number of 1 hour before moment
According to similarlyIt is tiThe temperature monitoring data of n hours, β before momenti-n,βi-n+1,...,βi-2,βi-1,βiWhen respectively each
Multiple linear regression coefficient of the monitoring temperature in entire multiple predictors is carved, c is coefficient to be estimated, and ε is error amount.
Multiple linear regression model is established using MATLAB, the data of acquisition are primary for acquisition in every 15 minutes, with continuous 48
The temperature forecast data of a hour are fitted calculating for independent variable, so as to fulfill the prediction to multinomial monitoring index data.
The alarm strategy taken in this example is：Pass through the detection data arrived to system acquisition and preset warning level
Value is compared, and warning message is generated if transfiniting, and is automatically stored in alarm database, by sound or it is watchful in a manner of carry out
Prompting.Simultaneously according to the correlation between the spatial and temporal distributions and monitoring parameter of field monitoring data, it is pre- to establish rail system state
It surveys and mathematical model of decision, a degree of prediction of development progress to subsequent rail state carries out the destruction that may occur
A degree of early warning.
In data processing, since data volume is huge, it is most likely that exceptional value occur.In order to avoid exceptional value causes
The alarm of mistake or the prediction result for influencing track condition, the abnormity removing method that the utility model is taken are：By more
Year observation data obtain the arithmetic mean of instantaneous value and root-mean-square-deviation of every monitoring index, and drawing is used to reach principle combination quantile case
Type drawing method judges exceptional value jointly, when two kinds of method of discrimination results are consistent, then using data measured as abnormality value removing, if
It is continuous whithin a period of time or wrong data repeatedly occur, then judge that field monitoring equipment is likely to occur failure.Two kinds differentiate original
Then specific method is as follows：
(1) draw to reach principle, defining exceptional value is | x- μ | the data of 3 σ of ＞, wherein x are the currently monitored value, and μ is monitoring number
According to average value, σ is poor for data standard；
(2) box figure differentiates, defines exceptional value Yi and is defined as：Yi<Q1- α × IQR or Yi>Q1+ α × IQR, wherein Q1For under
Quartile, IQR are interquartile range, Q3For upper quartile, α is adjustability coefficients, is set as 1.5.
As shown in figure 11, the data transmission flow that monitoring and warning system is established in this example is：Field monitoring data
Acquisition is transmitted the monitoring data at scene by being laid with special optical cable between monitoring field and its neighbouring data acquisition center
Into the acquisition server of data acquisition center.By wireless network, monitoring data are transferred in the processing server of rear end.
To sum up embodiment is monitored railroad track using the utility model, and the program can monitor iron in real time for a long time
Rail temperature, flexible additional force, vertical force, cross force and displacement in the rail system of road, point tongue displacement, the temperature of track plates
Spend gradient and stress, base board-bridge relative displacement, the temperature of bridge, displacement.On this basis, the utility model is using now
Correlation between the spatial and temporal distributions and monitoring parameter of field monitoring data, is analyzed by the processing of data, to circuit service state
Comprehensive assessment is carried out, realizes the early warning and alert of automation.The utility model does not destroy track structure, and the monitoring on track is set
It is standby passive, track circuit will not be had an impact, while itself strong antijamming capability, no drift, can ensure the precision of test
And accuracy.The characteristics of the utility model can adapt to railways train operation speed height, and density is big, and Window time is short can adapt to existing
Field adverse circumstances, round-the-clock monitoring rail system, and realize data storage and secure communication automatically.The utility model has sensitivity
It is reasonable that point is laid, and captures timely, the advantages that the influence degree real time reaction of structure, meets the needs of safety of railway operation,
Solves gapless track safety military service state controllability technical barrier, safety, even running for train provide reliable guarantee.
In conclusion railroad track is monitored by the utility model, and by Fiber Bragg Grating technology, video-aware skill
Art and the fusion for correcting stress-strain technology, form to track structure from appearance to inherence, are seen from thin to macroscopic view, from low frequency
Hybrid Decision-making to high frequency monitors system, realizes the long-term real-time monitoring to rail system.By being carried out to monitoring data
Automatic collection and processing are analyzed, being capable of real-time testing rail system safety military service behavior.Simultaneously by establishing rail system state
Prediction and mathematical model of decision carry out prediction and warning, so as to which the safety for train is put down to a certain extent to the destruction that may occur
Steady operation provides safeguard.The utility model is suitable for high-speed railway and city rail traffic route, has very high application value
With business promotion prospect.
Obviously, above-described embodiment of the utility model is only intended to clearly illustrate the utility model example, and
It is not the restriction to the embodiment of the utility model, for those of ordinary skill in the art, in above description
On the basis of can also make other variations or changes in different ways, all embodiments can not be exhaustive here,
It is every belong to obvious changes or variations that the utility model extends out still in the scope of protection of the utility model it
Row.
Claims (4)
1. a kind of rail system safe condition comprehensive monitoring and intelligent analysis system, which is characterized in that the system comprises：
Data acquisition unit, for carrying out data acquisition to the temperature of rail system, stress, displacement；
Database Unit, for storing the orbital data of data acquisition unit acquisition；
Data processing and inversion unit carries out processing analysis for transferring surveyed rail system data from database module, obtains
Obtain rail system current state information；
Predicting unit, according to rail system current state information, the variation tendency of predicted orbit force and deformation state；
Alarm unit, for the alarm threshold value of track current state information and systemic presupposition to be compared, if analysis result surpasses
Go out the threshold value of setting, then send out alarm；
Power supply unit, for providing electric power support for the unit module in system in addition to fiber-optic grating sensor.
2. system according to claim 1, which is characterized in that
The data acquisition unit includes：Fiber-optic grating sensor, for stress, displacement and the temperature under acquisition trajectory system low frequency
Degree；Video sensor, for the displacement data of acquisition trajectory system；Stress flower patch, for the rail under acquisition trajectory system high-frequency
Road lateral stress and vertical stress.
3. system according to claim 1, which is characterized in that
The alarm unit includes：Threshold value comparing module, for data results and preset threshold value to be compared,
And early warning instruction is sent out according to comparing result；Alert data library module, for storing comparing result.
4. system according to claim 1, which is characterized in that the system also includes：
Filter module, the video data for being obtained to track big displacement data collecting module collected are filtered except dry processing；
Compensating module compensates for the error caused by track external environment；
Picture recognition module for carrying out image identification to tongue picture, obtains the telescopic displacement of tongue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201721629476.7U CN207552827U (en) | 2017-11-29 | 2017-11-29 | A kind of rail system safe condition comprehensive monitoring and intelligent analysis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201721629476.7U CN207552827U (en) | 2017-11-29 | 2017-11-29 | A kind of rail system safe condition comprehensive monitoring and intelligent analysis system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN207552827U true CN207552827U (en) | 2018-06-29 |
Family
ID=62678055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201721629476.7U Active CN207552827U (en) | 2017-11-29 | 2017-11-29 | A kind of rail system safe condition comprehensive monitoring and intelligent analysis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN207552827U (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108880668A (en) * | 2018-07-30 | 2018-11-23 | 北京交通大学 | A kind of railway optical cable real-time monitoring system for state and method |
| CN109302811A (en) * | 2018-09-28 | 2019-02-01 | 天津斌德鑫力电子科技有限公司 | A kind of suspension railway vehicle static debugging system |
| CN110059403A (en) * | 2019-04-16 | 2019-07-26 | 上海宝钢工业技术服务有限公司 | The fatigue detecting analysis and assessment system of track girder group |
| CN110733534A (en) * | 2019-10-24 | 2020-01-31 | 西南交通大学 | rail crawling observation method and system |
| CN111474386A (en) * | 2020-04-20 | 2020-07-31 | 北京交通大学 | Vehicle shaking disease data deviation correction method and system |
| CN112330073A (en) * | 2020-11-30 | 2021-02-05 | 天津大学 | Method for predicting key effect quantity of monitoring data of long-distance water transfer project |
| CN113255151A (en) * | 2021-06-08 | 2021-08-13 | 长安大学 | Tunnel construction deformation prediction method and system based on composite neural network |
| CN113657041A (en) * | 2021-08-25 | 2021-11-16 | 哈尔滨工业大学 | Intelligent sensing and forecasting system for physical and mechanical states of roadbed in alpine region |
| US20220170976A1 (en) * | 2020-12-02 | 2022-06-02 | Siemens Aktiengesellschaft | Assistance apparatus for localizing errors in a monitored technical system |
| CN115535026A (en) * | 2022-12-05 | 2022-12-30 | 北京智弘通达科技有限公司 | Railway track flatness detection method and system |
| CN116385979A (en) * | 2023-06-07 | 2023-07-04 | 河北铁达科技有限公司 | Beam end steel rail expansion regulator monitoring method, terminal, monitoring system and medium |
| CN117232451A (en) * | 2023-08-30 | 2023-12-15 | 中铁第一勘察设计院集团有限公司 | Track 3D geometric form high-frequency monitoring system and method |
| CN117991278A (en) * | 2024-04-03 | 2024-05-07 | 任丘市东之风专用量仪有限公司 | Digital signal acquisition intelligent system and device of measuring instrument based on big data |
-
2017
- 2017-11-29 CN CN201721629476.7U patent/CN207552827U/en active Active
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108880668A (en) * | 2018-07-30 | 2018-11-23 | 北京交通大学 | A kind of railway optical cable real-time monitoring system for state and method |
| CN109302811A (en) * | 2018-09-28 | 2019-02-01 | 天津斌德鑫力电子科技有限公司 | A kind of suspension railway vehicle static debugging system |
| CN109302811B (en) * | 2018-09-28 | 2023-08-22 | 天津斌德鑫力电子科技有限公司 | Suspension railway vehicle static debugging system |
| CN110059403A (en) * | 2019-04-16 | 2019-07-26 | 上海宝钢工业技术服务有限公司 | The fatigue detecting analysis and assessment system of track girder group |
| CN110733534A (en) * | 2019-10-24 | 2020-01-31 | 西南交通大学 | rail crawling observation method and system |
| CN111474386A (en) * | 2020-04-20 | 2020-07-31 | 北京交通大学 | Vehicle shaking disease data deviation correction method and system |
| CN112330073A (en) * | 2020-11-30 | 2021-02-05 | 天津大学 | Method for predicting key effect quantity of monitoring data of long-distance water transfer project |
| US20220170976A1 (en) * | 2020-12-02 | 2022-06-02 | Siemens Aktiengesellschaft | Assistance apparatus for localizing errors in a monitored technical system |
| CN113255151A (en) * | 2021-06-08 | 2021-08-13 | 长安大学 | Tunnel construction deformation prediction method and system based on composite neural network |
| CN113255151B (en) * | 2021-06-08 | 2024-03-12 | 长安大学 | Tunnel construction deformation prediction method and system based on composite neural network |
| CN113657041A (en) * | 2021-08-25 | 2021-11-16 | 哈尔滨工业大学 | Intelligent sensing and forecasting system for physical and mechanical states of roadbed in alpine region |
| CN115535026A (en) * | 2022-12-05 | 2022-12-30 | 北京智弘通达科技有限公司 | Railway track flatness detection method and system |
| CN116385979A (en) * | 2023-06-07 | 2023-07-04 | 河北铁达科技有限公司 | Beam end steel rail expansion regulator monitoring method, terminal, monitoring system and medium |
| CN116385979B (en) * | 2023-06-07 | 2023-09-05 | 河北铁达科技有限公司 | Beam end steel rail expansion regulator monitoring method, terminal, monitoring system and medium |
| CN117232451A (en) * | 2023-08-30 | 2023-12-15 | 中铁第一勘察设计院集团有限公司 | Track 3D geometric form high-frequency monitoring system and method |
| CN117991278A (en) * | 2024-04-03 | 2024-05-07 | 任丘市东之风专用量仪有限公司 | Digital signal acquisition intelligent system and device of measuring instrument based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN207552827U (en) | A kind of rail system safe condition comprehensive monitoring and intelligent analysis system | |
| CN107858883A (en) | A kind of rail system safe condition comprehensive monitoring and intelligent analysis method | |
| CN104401360A (en) | Multi-means fusion-based safety real-time monitoring method and multi-means fusion-based safety real-time monitoring system for railway track system | |
| CN206563898U (en) | It is a kind of automatic data collection and the bridge monitoring system of integrated management to be carried out | |
| CN102343922B (en) | On-line monitoring system for vibration characteristics of rapid railway turnout based on wireless sensor network | |
| RU2444449C1 (en) | Method and system of diagnostics and remote monitoring of railway overhead contact system | |
| CN109278053B (en) | Railway tunnel operation state detection and monitoring method | |
| CN113159475B (en) | Infrastructure full life cycle monitoring platform and method | |
| CN111060270B (en) | Movable rapid monitoring and intelligent evaluation method for urban viaduct | |
| CN103031787A (en) | Method and device for fully automatically and accurately monitoring changes of geometric parameters of track | |
| US20130066600A1 (en) | Method and apparatus for real-time line rating of a transmission line | |
| CN108280969A (en) | A kind of high slope surface deformation monitoring and warning system and its method for early warning | |
| CN104655101A (en) | High-precision lead type whole-section slope stability monitoring pre-alarming system and monitoring method thereof | |
| CN110821560A (en) | Tunnel inspection system | |
| CN104015756B (en) | Rail temperature based on wireless senser and pressure detection method | |
| CN115601008A (en) | Engineering settlement deformation monitoring system and method based on digital twinning | |
| CN111829585A (en) | Highway subgrade disease monitoring system and method based on OFDR | |
| CN210707430U (en) | Switch application on-line state monitoring system | |
| Moschas et al. | High accuracy measurement of deflections of an electricity transmission line tower | |
| CN104121945A (en) | Distributed sag online monitoring system and method for optical fiber composite overhead ground wire | |
| CN208172969U (en) | A kind of high slope surface deformation monitoring and warning system | |
| JP6396100B2 (en) | Early use resumption support system at the time of earthquake | |
| CN215491553U (en) | Steel rail displacement detection system | |
| CN115682948A (en) | Urban rail monitoring method and system based on distributed optical fiber | |
| CN109490892A (en) | A kind of method for real-time monitoring for compensation device of casting anchor, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |