CN104260754B - Track transition forecasting system and method based on axle box vibration acceleration - Google Patents
Track transition forecasting system and method based on axle box vibration acceleration Download PDFInfo
- Publication number
- CN104260754B CN104260754B CN201410526294.1A CN201410526294A CN104260754B CN 104260754 B CN104260754 B CN 104260754B CN 201410526294 A CN201410526294 A CN 201410526294A CN 104260754 B CN104260754 B CN 104260754B
- Authority
- CN
- China
- Prior art keywords
- axle box
- vibration acceleration
- narx
- track transition
- data
- 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
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61K—AUXILIARY EQUIPMENT SPECIALLY ADAPTED FOR RAILWAYS, NOT OTHERWISE PROVIDED FOR
- B61K9/00—Railway vehicle profile gauges; Detecting or indicating overheating of components; Apparatus on locomotives or cars to indicate bad track sections; General design of track recording vehicles
- B61K9/08—Measuring installations for surveying permanent way
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a kind of track transition forecasting system and method based on axle box vibration acceleration.The system includes axle box vibration acceleration sensor, rotary pulsed tachogenerator, simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless network route, host computer.The axle box vibration acceleration sensor is arranged on train EEF bogie axle box, rotary pulsed tachogenerator is arranged in train axle end cap, the axle box vibration signal of the axle box vibration acceleration sensor collection and the GES of rotary pulsed tachogenerator collection first pass through simulation pretreatment circuit and carry out low-pass filtering treatment, again data signal is converted analog signals into through A/D change-over circuits, data signal is sent to host computer through wireless network route by embedded system and is processed, and host computer is predicted by vibration signal and obtains track transition result.The present invention has the advantages that low cost, engineering construction are good.
Description
Technical field
The present invention relates to the technical field of track irregularity prediction, particularly a kind of track based on axle box vibration acceleration
Longitudinal irregularity forecasting system and method.
Background technology
Track irregularity refers to the deviation of track geometry shape, size and locus with respect to its normal condition.Straight line rail
Road is uneven, not straight, to position of center line and orbit altitude, the deviation of width just size；Curve track is not smoother, deflection curve
Position of center line, deviates curvature, superelevation, the correct value of gauge, deviates along the track geometry deviations such as slope varying dimensions common name rail
Road irregularity.It is that rolling stock produces random vibration, track structure fatigue damage and rolling stock safety in operation to decline
Major influence factors.
The reason for producing track irregularity and influence factor are a lot.Track irregularity often originates from the defect of rail material
With the foozle or tolerance of the track component such as rail, and directly produced in line construction quality and work progress it is various just
Beginning irregularity.After track opens operation, under High-speed Train Loads, new track irregularity will further produce development, origin
To gradually increase deterioration in the various initial irregularity of manufacture and track construction process.During operation, the effect of rolling stock
It is track irregularity generation, development and the main cause for deteriorating.Additionally, the change of natural environment, orbital maintenance operation, track structure
The factor such as make has very important influence to the generation of track irregularity, development, deterioration.
Yang Wenzhong is studied axle box acceleration and track transition the (axle box acceleration of Yang Wen loyalty base Yu Xiaobos
With the research Tongji University Ph.D. Dissertation 2008 of track irregularity relation), the method passes through axle box acceleration double integrator meter
Calculation obtains track transition, but the method needs the accelerometer of multiple different bandwidths to realize that engineering construction is poor.
Curtain etc. proposes to hang down longitudinal irregularity in vehicle-rail system and vehicle using Hilbert-Huang transform method Lee again
It is analyzed that (Lee's curtain again, practices good, the Liu Xiao boats HHT frequency divisions in vehicle-rail system vertical vibration of pine to vibration acceleration relation
Application in analysis, vibration-testing with diagnosis 2013), using Empirical mode decomposition (empirical mode
Decomposition, abbreviation EMD) to survey longitudinal irregularity decomposed with Vertical Vibration of Vehicle acceleration signal, obtain
Both intrinsic mode functions；Then, by the time domain waveform and Hilbert energy spectrums of both comparative analysis intrinsic mode functions, say
Deterministic corresponding relation between bright longitudinal irregularity intrinsic mode functions and Vertical Vibration of Vehicle acceleration intrinsic mode functions, can
To recognize the bad section of track transition using Vertical Vibration of Vehicle acceleration, the method needs to be adopted according to track checking car
Collection vibration and track transition data, it is impossible to suitable for common vehicle in use.
The content of the invention
It is an object of the invention to provide the good track based on axle box vibration acceleration of a kind of low cost, engineering construction
Longitudinal irregularity forecasting system and method are uneven to rail height by gathering the vibration acceleration signal on vehicle in use axle box
It is suitable to carry out real time on-line monitoring.
Realizing the technical solution of the object of the invention is：
A kind of track transition Forecasting Methodology based on axle box vibration acceleration, comprises the steps of：
Step 1, sets on train EEF bogie axle box and rotation is set in axle box vibration acceleration sensor, train axle end cap
Turn pulse tachogenerator, and simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless is set on operation train
Network route, host computer；
Step 2, it is pre- that the axle box vibration acceleration sensor, the output end of rotary pulsed tachogenerator access simulation
Process circuit, the output end of simulation pretreatment circuit accesses embedded system by A/D change-over circuits；
Step 3, vehicle gathers axle box vibration acceleration during operation, by axle box vibration acceleration sensor to be believed
Number, rotary pulsed tachogenerator collection GES, the signal for being gathered is filtered through simulation pretreatment circuit, then passes through
A/D change-over circuits convert analog signals into data signal input embedded system, and embedded system passes through the data of collection
Wireless network route is sent to host computer；
Step 4, host computer is using existing axle box vibration acceleration signal as input, track transition spectrum as defeated
Go out, trained using exogenous nonlinear Recurrent neural network NARX, obtain network-related parameters, the network-related parameters include
Link weight coefficients, each Node B threshold between node；
Step 5, host computer passes through NARX neutral net real-time estimates using the axle box vibration acceleration signal for collecting
Current orbit longitudinal irregularity, it is specific as follows：
(1) according to existing axle box vibration acceleration signal and the track transition data of historical accumulation, to the axle box
Vibration acceleration signal and track transition data are normalized respectively, and normalization formula is：
In formula,It is the data after normalization, xiFor in vibration acceleration signal or track transition data
I-th data, xminIt is minimum value, x in vibration acceleration signal or track transition datamaxIt is vibration acceleration signal
Or maximum in track transition data；
(2) determine NARX neural network structures, including input number of nodes, output node number, hidden layers numbers, set NARX nets
The input and output layer neuron number of network, input layer is 1, and output layer neuron is 1, selects hidden node and output layer section
The activation primitive of point, activation primitive includes threshold function table, piecewise linear function and nonlinear function；
(3) hidden node number is determined, using experience traversal, i.e., by choosing different Hidden nodes come training net
Network, hidden node number when selection performance is optimal；
(4) determine time delay exponent number, take input time delay exponent number and be consistent with output time delay exponent number, use
Experience traversal constructs the NARX neutral nets that one group of different time postpones exponent number, selection test root-mean-square error minimum time
Postpone exponent number；
(5) training algorithm for selecting to be adapted to the NARX neutral nets is Regularization algorithms；
(6) it is high as the input data of NARX neutral nets, track by the use of the axial vibration acceleration information of historical accumulation
Low irregularity data are trained as the output data of NARX neutral nets to NARX, obtain NARX neural metwork trainings and refer to
Mark, the NARX neural metwork trainings index includes root-mean-square error and network output and the coefficient correlation of reality output, compares
The coefficient correlation between NARX neutral nets output valve and real system output valve obtained by training evaluates network performance, just
Root error is smaller, and coefficient correlation shows that network performance is more superior closer to 1；
(7) the real-time axle box vibration acceleration signal that will be gathered is input into as NARX neutral nets, and predicted orbit height is not
Smooth-going, according to the Rail inspection allowable deviation that railway interests's part irregularity amplitude transfinites in point system to longitudinal irregularity
To judge track transition state, train operation distance and track transition data are calculated with reference to rate signal, pushed away
Calculate track transition position.
Compared with prior art, its remarkable advantage is the present invention：(1) track irregularity is detected on vehicle in use, it is to avoid
Traditional detection method needs the operation expense that special track checking car brings；(2) engineering construction is good, and the axle box vibration adds
Velocity sensor and the convenient installation of rotary pulsed tachogenerator, reliability are high.
Brief description of the drawings
Fig. 1 is the structure chart of track transition forecasting system of the present invention based on axle box vibration acceleration.
Fig. 2 is the scheme of installation of sensor in present system.
Fig. 3 is the flow chart of track transition Forecasting Methodology of the present invention based on axle box vibration acceleration.
Fig. 4 is the NARX neutral nets output valve and real system output valve obtained by being trained in embodiment.
Fig. 5 is the NARX neutral nets output valve and real system output valve obtained by being tested in embodiment.
Specific embodiment
Below in conjunction with the accompanying drawings and specific embodiment is described in further detail to the present invention.
With reference to Fig. 1, track transition forecasting system of the present invention based on axle box vibration acceleration, including axle box vibration
Acceleration transducer, rotary pulsed tachogenerator, simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless network
Network route, host computer；The axle box vibration acceleration sensor is arranged on train EEF bogie axle box, the rotary pulsed sensing that tests the speed
Device is arranged in train axle end cap, the axle box vibration acceleration sensor, rotary pulsed tachogenerator output end it is equal
Simulation pretreatment circuit is accessed, the output end of simulation pretreatment circuit accesses embedded system by A/D change-over circuits；
What the axle box vibration signal and rotary pulsed tachogenerator of the axle box vibration acceleration sensor collection were gathered
GES first passes through simulation pretreatment circuit and carries out low-pass filtering treatment, then is converted analog signals into through A/D change-over circuits
Data signal, data signal is sent to host computer through wireless network route by embedded system and is processed, and host computer is by shaking
Dynamic signal estimation obtains track transition result, and result is shown, is stored.
With reference to Fig. 1~2, track transition Forecasting Methodology of the present invention based on axle box vibration acceleration, comprising following
Step：
Step 1, sets on train EEF bogie axle box and rotation is set in axle box vibration acceleration sensor, train axle end cap
Turn pulse tachogenerator, and simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless is set on operation train
Network route, host computer；
Step 2, it is pre- that the axle box vibration acceleration sensor, the output end of rotary pulsed tachogenerator access simulation
Process circuit, the output end of simulation pretreatment circuit accesses embedded system by A/D change-over circuits；
Step 3, vehicle gathers axle box vibration acceleration during operation, by axle box vibration acceleration sensor to be believed
Number, rotary pulsed tachogenerator collection GES, the signal for being gathered is filtered through simulation pretreatment circuit, then passes through
A/D change-over circuits convert analog signals into data signal input embedded system, and embedded system passes through the data of collection
Wireless network route is sent to host computer；
Step 4, host computer is using existing axle box vibration acceleration signal as input, track transition spectrum as defeated
Go out, using exogenous nonlinear Recurrent neural network NARX (Nonlinear Auto-Regressive with eXogenous
Input Neural Networks) training, network-related parameters are obtained, the network-related parameters include the connection between node
Weight coefficient, each Node B threshold；
Step 5, host computer passes through NARX neutral net real-time estimates using the axle box vibration acceleration signal for collecting
Current orbit longitudinal irregularity.
With reference to Fig. 3, host computer is pre- in real time by NARX neutral nets using the axle box vibration acceleration signal for collecting
Current orbit longitudinal irregularity is surveyed, it is specific as follows：
(1) according to existing axle box vibration acceleration signal and the track transition data of historical accumulation, to the axle box
Vibration acceleration signal and track transition data are normalized respectively, and normalization formula is：
In formula,It is the data after normalization, xiFor in vibration acceleration signal or track transition data
I-th data, xminIt is minimum value, x in vibration acceleration signal or track transition datamaxFor vibration acceleration is believed
Number or track transition data in maximum；
(2) determine NARX neural network structures, including input number of nodes, output node number, hidden layers numbers, set NARX nets
The input and output layer neuron number of network, input layer is 1, and output layer neuron is 1, selects hidden node and output layer section
The activation primitive of point, activation primitive includes threshold function table, piecewise linear function and nonlinear function；
(3) hidden node number is determined, using experience traversal, i.e., by choosing different Hidden nodes come training net
Network, hidden node number when selection performance is optimal；
(4) determine time delay exponent number, take input time delay exponent number and be consistent with output time delay exponent number, use
Experience traversal constructs the NARX neutral nets that one group of different time postpones exponent number, selection test root-mean-square error minimum time
Postpone exponent number；
(5) training algorithm of the suitable NARX neutral nets is selected, NARX neural network BP training algorithms include：Real-time recurrence
Learning algorithm (RTRL), the BP algorithm (BPTT) with the time, dynamic BP algorithm (DBP), hierarchical optimization algorithm (Layer-By-
Layer optimizing), Bayesian regularization (BR) algorithm etc., the NARX neutral nets are unable to on-line operation due to BPTT,
DBP computation complexities are higher, and RTRL is less efficient, and BR algorithms can reduce effective network parameter and be missed with making up larger network
Difference, therefore selection BR algorithms；
(6) by the use of historical accumulation axial vibration acceleration information as NARX neutral nets input, rail height not
Smooth-going data are trained as the output of NARX neutral nets to NARX, obtain NARX neural metwork training indexs, described
NARX neural metwork trainings index includes the coefficient correlation of root-mean-square error and network output and reality output, compares training gained
Coefficient correlation between the NARX neutral nets output valve and real system output valve that arrive, evaluates network performance, root-mean-square error
Smaller, coefficient correlation shows that network performance is more superior closer to 1；
(7) the real-time axle box vibration acceleration signal that will be gathered is input into as NARX neutral nets, and predicted orbit height is not
Smooth-going, according to the Rail inspection allowable deviation that railway interests's part irregularity amplitude transfinites in point system to longitudinal irregularity
To judge track transition state, train operation distance and track transition data are calculated with reference to rate signal, pushed away
Calculate track transition position.
Embodiment 1
With reference to Fig. 1, track transition forecasting system of the present invention based on axle box vibration acceleration, including axle box vibration
Acceleration transducer, rotary pulsed tachogenerator, simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless network
Network route, host computer；The axle box vibration acceleration sensor is arranged on train EEF bogie axle box, the rotary pulsed sensing that tests the speed
Device is arranged in train axle end cap, the axle box vibration acceleration sensor, rotary pulsed tachogenerator output end it is equal
Simulation pretreatment circuit is accessed, the output end of simulation pretreatment circuit accesses embedded system by A/D change-over circuits；The axle
The axle box vibration signal of case vibration acceleration sensor collection and the GES of rotary pulsed tachogenerator collection are first passed through
Simulation pretreatment circuit carries out low-pass filtering treatment, then converts analog signals into data signal, numeral letter through A/D change-over circuits
Number sent to host computer through wireless network route by embedded system and to be processed, host computer obtains rail by vibration signal prediction
Road longitudinal irregularity result.
The DH112 type piezoelectric acceleration transducers that the axle box vibration acceleration sensor is produced using the test of east China,
Range is 1000m/s2, frequency response range is 0.5~1KHz, and sensitivity is 0~5mV/ms-2。
It is the embedded system device of core that the embedded system uses arm processor, and the equipment includes wireless network
Module, the model AT91SAM9263 of the arm processor, dominant frequency 200MHz possess peripheral hardware resource in abundant piece, can run
Embedded Linux system；The embedded system is integrated with MAC circuit, using the Ethernet interface of PHY chip DM9161, leads to
Crossing setting IP address and MAC carries out Ethernet connection, is configured with the USB-WIFI modules based on RT3070 chips, the USB-WIFI
Module is connected by USB interface with mainboard, transmission rate 150Mbps, by the drive for adding the USB-WIFI modules in the motherboard
The dynamic wireless connection for being capable of achieving simulation platform and wireless routing.
The wireless network is route using the AR151W-P/AR151W-P- type wireless routers of Huawei Company, and IMX is
100Mbps, memory size is 512M, with serial auxiliary/console port.
The ITX3010 core main frames that the host computer is produced using Sheng Bo scientific ＆ technical corporation, the main frame uses Intel's atom
Processor D525, support Surface Mount internal memory, the extension of DIMM bars, internal memory up to 2GB/4GB, support 2 SATA, support 18 VGA with it is only
It is vertical double aobvious, Linux, VxWorks, Windows can be run, there is provided PC/104 and PC/104+ bus extensions.
With reference to Fig. 3~5, the algorithm flow using axle box vibration acceleration signal predicted orbit longitudinal irregularity is：
(1) according to existing axial vibration acceleration information and track transition data, data are normalized with place
Reason, in the range of data normalization to [0 1]：
In formula,It is the data after normalization, xiFor in vibration acceleration signal or track transition data
I-th data, xminIt is minimum value, x in vibration acceleration signal or track transition datamaxIt is vibration acceleration signal
Or maximum in track transition data.
(2) NARX neural network structures are determined, the input layer for setting NARX networks is axle box vibration acceleration, defeated
Ingress number is 1, and output layer neuron is track transition, and output node number is 1, and selection hidden node activation primitive is
Sigmoid functions, output node layer activation primitive is linear function.
(3) hidden node number is determined, empirically, it is 17 to select suitable hidden node number.
(4) determine time delay exponent number, take input time delay exponent number and be consistent with output time delay exponent number, foundation
Experience or test determine that time delay exponent number is 45.
(5) training algorithm for selecting the NARX neutral nets is Bayesian Regularization algorithms.
(6) Fig. 4, Fig. 5 give the NARX neutral nets output valve and real system output valve obtained by training and test
Between comparing, training coefficient R be 0.8030, test coefficient R be 0.7730.It can be seen that, the output of NARX neutral nets
The good relationship exported with real system,
(7) the real-time axle box vibration acceleration signal that will be gathered is input into as NARX neutral nets, and predicted orbit height is not
Smooth-going, according to threshold determination longitudinal irregularity, track transition position can be calculated with reference to rate signal.
In sum, the present invention detects track irregularity on vehicle in use, it is to avoid traditional detection method needs special
The operation expense that track checking car brings；Engineering construction is good, the axle box vibration acceleration sensor and rotary pulsed tests the speed
Sensor is convenient to be installed, and reliability is high.
Claims (1)
1. a kind of track transition Forecasting Methodology based on axle box vibration acceleration, it is characterised in that comprise the steps of：
Step 1, sets in axle box vibration acceleration sensor, train axle end cap on train EEF bogie axle box and sets rotation arteries and veins
Tachogenerator is rushed, and simulation pretreatment circuit, A/D change-over circuits, embedded system, wireless network are set on operation train
Route, host computer；
Step 2, the axle box vibration acceleration sensor, the output end of rotary pulsed tachogenerator access simulation pretreatment
Circuit, the output end of simulation pretreatment circuit accesses embedded system by A/D change-over circuits；
Step 3, vehicle gathers axle box vibration acceleration signal, rotation during operation by axle box vibration acceleration sensor
Turn pulse tachogenerator collection GES, the signal for being gathered is filtered through simulation pretreatment circuit, then turns by A/D
Change circuit and convert analog signals into data signal input embedded system, the data that embedded system will be gathered pass through wireless network
Network route is sent to host computer；
Step 4, host computer is composed existing axle box vibration acceleration signal as input, track transition as output,
Trained using exogenous nonlinear Recurrent neural network NARX, obtain network-related parameters, the network-related parameters include section
Link weight coefficients, each Node B threshold between point；
Step 5, host computer is current by NARX neutral net real-time estimates using the axle box vibration acceleration signal for collecting
Track transition, it is specific as follows：
(1) according to existing axle box vibration acceleration signal and the track transition data of historical accumulation, the axle box is vibrated
Acceleration signal and track transition data are normalized respectively, and normalization formula is：
In formula,It is the data after normalization, xiIt is i-th in vibration acceleration signal or track transition data
Individual data, xminIt is minimum value, x in vibration acceleration signal or track transition datamaxFor vibration acceleration signal or
Maximum in track transition data；
(2) NARX neural network structures, including input number of nodes, output node number, hidden layers numbers are determined, setting NARX networks
Input and output layer neuron number, input layer is 1, and output layer neuron is 1, selects hidden node with output node layer
Activation primitive, activation primitive includes threshold function table, piecewise linear function and nonlinear function；
(3) hidden node number is determined, using experience traversal, i.e., by choosing different Hidden nodes come training network, choosing
Take performance it is optimal when hidden node number；
(4) determine time delay exponent number, take input time delay exponent number and be consistent with output time delay exponent number, using experience
Traversal constructs the NARX neutral nets that one group of different time postpones exponent number, the minimum time delay of selection test root-mean-square error
Exponent number；
(5) training algorithm for selecting to be adapted to the NARX neutral nets is Regularization algorithms；
(6) by the use of historical accumulation axial vibration acceleration information as NARX neutral nets input data, rail height not
Smooth-going data are trained as the output data of NARX neutral nets to NARX, obtain NARX neural metwork training indexs, institute
Stating NARX neural metwork trainings index includes root-mean-square error and network output and the coefficient correlation of reality output, compares training institute
Coefficient correlation between the NARX neutral nets output valve and real system output valve that obtain evaluates network performance, root-mean-square error
Smaller, coefficient correlation shows that network performance is more superior closer to 1；
(7) the real-time axle box vibration acceleration signal that will be gathered is input into as NARX neutral nets, predicted orbit longitudinal irregularity,
Being transfinited according to railway interests's part irregularity amplitude the Rail inspection allowable deviation of longitudinal irregularity is sentenced in point system
Disconnected track transition state, train operation distance and track transition data are calculated with reference to rate signal, are extrapolated
Track transition position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410526294.1A CN104260754B (en) | 2014-10-08 | 2014-10-08 | Track transition forecasting system and method based on axle box vibration acceleration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410526294.1A CN104260754B (en) | 2014-10-08 | 2014-10-08 | Track transition forecasting system and method based on axle box vibration acceleration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104260754A CN104260754A (en) | 2015-01-07 |
| CN104260754B true CN104260754B (en) | 2017-06-27 |
Family
ID=52152390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410526294.1A Active CN104260754B (en) | 2014-10-08 | 2014-10-08 | Track transition forecasting system and method based on axle box vibration acceleration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104260754B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105372080B (en) * | 2015-12-22 | 2018-04-24 | 成都市新筑路桥机械股份有限公司 | A kind of tramcar and its embedded tracks Coupled Dynamics test device and method |
| CN106840717B (en) * | 2017-01-15 | 2024-03-29 | 华东交通大学 | Train wheel vibration testing method based on axle box acceleration electromagnetic interference resistance |
| CN107264569A (en) * | 2017-06-15 | 2017-10-20 | 山东交通职业学院 | A kind of track quality state evaluation method and system |
| CN109034192B (en) * | 2018-06-20 | 2022-03-08 | 西南交通大学 | Track-vehicle body vibration state prediction method based on deep learning |
| CN112014593B (en) * | 2019-05-28 | 2022-05-17 | 浙江德盛铁路器材股份有限公司 | Device and method for monitoring and evaluating quality condition of railway track basic equipment |
| CN110789566B (en) * | 2019-11-11 | 2021-01-12 | 成都西交智众科技有限公司 | Track defect monitoring method and monitoring equipment based on axle box acceleration signal |
| CN111979859B (en) * | 2020-08-19 | 2022-03-11 | 中国铁道科学研究院集团有限公司 | Track irregularity detection system and method |
| CN113358053B (en) * | 2021-04-27 | 2023-02-24 | 中车青岛四方机车车辆股份有限公司 | Track irregularity detection and evaluation system and method, electronic equipment and track vehicle |
| CN114993461B (en) * | 2022-08-08 | 2022-11-04 | 成都久和建设设备有限责任公司 | System and method for detecting vibration of motor of tower crane mechanism |
| CN116307302B (en) * | 2023-05-23 | 2023-07-25 | 西南交通大学 | Inversion method, system and storage medium for track irregularity dynamic and static detection data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102548828A (en) * | 2009-08-13 | 2012-07-04 | 代尔夫特工业大学 | Method and instrumentation for detection of rail defects, in particular rail top defects |
| CN102798458A (en) * | 2012-08-01 | 2012-11-28 | 北京铁科工程检测中心 | Low-frequency wireless accelerometer |
| CN102890143A (en) * | 2012-10-19 | 2013-01-23 | 西南交通大学 | Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information |
| CN103332207A (en) * | 2013-06-11 | 2013-10-02 | 大连理工大学 | Method and device for on-line acquisition of steel rail temperature, strain and acceleration data |
| CN104228872A (en) * | 2014-09-28 | 2014-12-24 | 江苏润仪仪表有限公司 | Portable track irregularity online monitoring device and method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2720172B2 (en) * | 1988-08-12 | 1998-02-25 | 帝都高速度交通営団 | Parallel recording method of train sway data and track data |
| JPH08184426A (en) * | 1994-12-28 | 1996-07-16 | Hitachi Ltd | Misalignment detector for rail based on oscillatory acceleration of axle box |
| JP3411861B2 (en) * | 1999-08-30 | 2003-06-03 | 財団法人鉄道総合技術研究所 | Inertia Masaya method |
| AUPR221900A0 (en) * | 2000-12-20 | 2001-01-25 | Central Queensland University | Vehicle dynamics prediction system and method |
| KR101250228B1 (en) * | 2010-06-09 | 2013-04-03 | 한국철도기술연구원 | Rail Irregularity Measurement System Using Accelerometers and Method Thereof |
-
2014
- 2014-10-08 CN CN201410526294.1A patent/CN104260754B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102548828A (en) * | 2009-08-13 | 2012-07-04 | 代尔夫特工业大学 | Method and instrumentation for detection of rail defects, in particular rail top defects |
| CN102798458A (en) * | 2012-08-01 | 2012-11-28 | 北京铁科工程检测中心 | Low-frequency wireless accelerometer |
| CN102890143A (en) * | 2012-10-19 | 2013-01-23 | 西南交通大学 | Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information |
| CN103332207A (en) * | 2013-06-11 | 2013-10-02 | 大连理工大学 | Method and device for on-line acquisition of steel rail temperature, strain and acceleration data |
| CN104228872A (en) * | 2014-09-28 | 2014-12-24 | 江苏润仪仪表有限公司 | Portable track irregularity online monitoring device and method |
Non-Patent Citations (1)
| Title |
|---|
| 基于轴箱谱的轨道短波不平顺识别方法研究;张伟;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20080915;第26-32页，图4-4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104260754A (en) | 2015-01-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104260754B (en) | Track transition forecasting system and method based on axle box vibration acceleration | |
| Shao et al. | Modified stacked autoencoder using adaptive Morlet wavelet for intelligent fault diagnosis of rotating machinery | |
| CN107941537B (en) | A kind of mechanical equipment health state evaluation method | |
| CN103901882B (en) | A kind of system and method for train dynamics system on-line monitoring fault diagnosis | |
| CN108845028B (en) | Method and device for dynamically detecting high-speed railway rail corrugation | |
| CN103884506B (en) | Production equipment on-line monitoring and fault diagnosis system and method based on wireless network | |
| CN102636577B (en) | Method for extracting rim ultrasonic flaw detection signal on basis of wavelet transformation | |
| CN108489719B (en) | A kind of combined failure of rotating machinery diagnostic method based on the unusual spectral factorization of G-P | |
| CN109858104A (en) | A kind of rolling bearing health evaluating and method for diagnosing faults and monitoring system | |
| CN105424366A (en) | Bearing fault diagnosis method based on EEMD adaptive denoising | |
| CN104794312A (en) | Method and device for evaluating track regularity | |
| CN103530511A (en) | Flutter boundary prediction method in wind tunnel flutter test under turbulence excitation condition | |
| CN102682180A (en) | Evaluation method for performance degradation of rotary mechanical equipment | |
| CN103163215A (en) | Pulse eddy current detection method and device of fatigue crack of vibrating screen for large mine | |
| CN106656669A (en) | Equipment parameter abnormity detection system and method based on self-adaptive setting of threshold | |
| CN111695201A (en) | Data-based monitoring method for running state of magnetic-levitation train | |
| CN113074807A (en) | Real-time monitoring system for vibration and deformation of cultivation fence facility structure | |
| CN114330096B (en) | Rail transit motor bearing fault degree classification model construction method based on deep learning | |
| Dai et al. | Acceleration-guided acoustic signal denoising framework based on learnable wavelet transform applied to slab track condition monitoring | |
| CN109919511A (en) | The linear appraisal procedure of existing railway and system | |
| CN114595721A (en) | Vehicle part strain load prediction method based on road condition recognition | |
| Du et al. | Study on optical fiber gas-holdup meter signal denoising using improved threshold wavelet transform | |
| CN107849904A (en) | Movement overseas compensation winch real-time performance monitoring and prediction | |
| CN116499742A (en) | Rotary machine fault monitoring method based on ensemble learning | |
| CN115062867A (en) | Bearing fault prediction method based on Teager energy spectrum and improved CNN model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210113 Address after: 210046 room 296, No.1 Tianzuo Road, Sheshan Xingcheng, Xigang street, Qixia District, Nanjing City, Jiangsu Province Patentee after: Nanjing hangxuan Rail Transit Technology Co.,Ltd. Address before: 210094 No. 200, Xiaolingwei, Jiangsu, Nanjing Patentee before: NANJING University OF SCIENCE AND TECHNOLOGY |