CN109625025B - BTM equipment early warning system - Google Patents

BTM equipment early warning system Download PDF

Info

Publication number
CN109625025B
CN109625025B CN201811523602.XA CN201811523602A CN109625025B CN 109625025 B CN109625025 B CN 109625025B CN 201811523602 A CN201811523602 A CN 201811523602A CN 109625025 B CN109625025 B CN 109625025B
Authority
CN
China
Prior art keywords
btm
early warning
data
model
equipment
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
Application number
CN201811523602.XA
Other languages
Chinese (zh)
Other versions
CN109625025A (en
Inventor
田茂志
牟海涛
郭利俊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Jiaoda Signal Technology Co Ltd
Original Assignee
Beijing Jiaoda Signal Technology Co Ltd
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 Beijing Jiaoda Signal Technology Co Ltd filed Critical Beijing Jiaoda Signal Technology Co Ltd
Priority to CN201811523602.XA priority Critical patent/CN109625025B/en
Publication of CN109625025A publication Critical patent/CN109625025A/en
Application granted granted Critical
Publication of CN109625025B publication Critical patent/CN109625025B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L23/00—Control, warning, or like safety means along the route or between vehicles or vehicle trains
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B61—RAILWAYS
    • B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
    • B61L27/00—Central railway traffic control systems; Trackside control; Communication systems specially adapted therefor

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Train Traffic Observation, Control, And Security (AREA)

Abstract

The invention provides a BTM equipment early warning system.A data acquisition system acquires and receives data recorded by BTM equipment or other external detection equipment; the early warning processing system preprocesses data according to collected BTM equipment operation data, trains an early warning model by the preprocessed data, adjusts and optimizes the model, stores model parameters, predicts the state of the BTM equipment according to the latest operation data, and supplies generated early warning information to the terminal display system; the terminal display system can inquire and display the performance state of the BTM equipment, and when the BTM equipment is in a poor state, the early warning processing system automatically sends the early warning information of the BTM equipment to the terminal display system for warning prompt. The invention brings the following technical effects: BTM equipment maintainer accessible terminal knows the early warning information of BTM equipment, and the suggestion maintainer overhauls the maintenance to equipment, discovers potential risk.

Description

BTM equipment early warning system
Technical Field
The invention relates to the field of train control in the railway industry, in particular to a state prediction and evaluation scheme of BTM equipment.
Background
With the rapid development of high-speed railways in China, high-speed motor train units make great contribution to the development of economy in China. More and more high-speed motor train units are operated on line, and how to ensure the safety and reliability of the operation of the high-speed motor train units becomes a great challenge problem for relevant operation and maintenance departments of railways. The traditional maintenance mode of 'planned maintenance' has the defects of under maintenance and over maintenance, and is difficult to adapt to the requirement of the development of modern railways. State-based maintenance is a new equipment maintenance strategy following post-and periodic maintenance, and is a reverse development of future equipment maintenance.
The motor train unit is rapidly developed, the usage amount of the BTM equipment and the responder equipment is continuously increased, the situation of the train at a later point is caused on a line due to the faults of the BTM equipment or the responder equipment, and if the states of the BTM equipment and the responder equipment can be predicted in advance, the equipment with a poorer state is maintained in time, so that the maintenance efficiency can be improved, the maintenance cost is reduced, and the safety and reliability of the equipment are improved.
At present, the application of a BTM device and a transponder device early warning system is not available in China. The status of the BTM device and transponder device cannot be predicted.
Disclosure of Invention
In order to solve the above problems, the present invention provides an early warning scheme for a BTM device.
The invention provides a BTM (Business to management) equipment early warning system, which is mainly used for training an early warning model according to received operation data of BTM equipment and predicting the state of the BTM equipment according to the model, wherein the early warning system comprises a data acquisition system, an early warning processing system and a terminal display system;
the data acquisition system acquires and receives data recorded by BTM equipment or other external detection equipment for the early warning processing system to use;
the early warning processing system is used for preprocessing the data according to the BTM equipment operation data acquired by the data acquisition system, training an early warning model by using the preprocessed data, adjusting and optimizing the model and then storing model parameters, predicting the state of the BTM equipment according to the latest operation data, storing the generated early warning information of the BTM equipment into a database or a file, and supplying the early warning information to the terminal display system;
the operation data comprises data items such as good code rate, decoding times, decoding time, train transponder speed and the like;
the terminal display system can inquire and display the performance state of the BTM equipment, and when the BTM equipment is in a poor state, for example, the state predicted by the early warning model is lower than a set threshold value, the early warning processing system automatically sends the early warning information of the BTM equipment to the terminal display system for warning prompt.
Therefore, the invention brings the following technical effects: BTM equipment maintainer can know the early warning information of BTM equipment through the terminal, and the suggestion maintainer overhauls the BTM equipment and maintains, discovers potential risk. The maintenance efficiency can be improved, the maintenance cost is reduced, and the safety and reliability of the equipment are improved.
Drawings
FIG. 1 is a block diagram of the composition and processing flow of the early warning system of the present invention;
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention.
The transponder system device includes a BTM (Balise Transmission Module transponder information receiving unit) and a transponder device, and an early warning method for the transponder system device is described below by taking the BTM device as an example, and the method is also suitable for early warning of other transponder devices.
The BTM equipment early warning system is mainly used for training a BTM equipment early warning model according to received operation data of BTM equipment and then predicting the state of the BTM equipment according to the model, and comprises a data acquisition system, an early warning processing system and a terminal display system.
The composition and the processing flow of the early warning system are shown in fig. 1. The method comprises the following steps:
a data acquisition system:
the method mainly comprises the steps of collecting and receiving data (including data items such as good code rate, decoding times, decoding time, train transponder speed and the like) recorded by BTM equipment or other external detection equipment (such as a detector), and converting the data into a fixed format to be stored in a database or a file for an early warning processing system to use.
The early warning processing system:
according to BTM equipment operation data acquired by a data acquisition system, preprocessing the data (without data with noise or interference), removing data with larger noise and incomplete data, training an early warning model (such as a regression model, a probability statistical model and the like) by using the preprocessed data, optimizing the model according to the existing data, and finally storing model parameters. And then predicting the state of the BTM equipment according to the latest data (the predicted performance value is larger and better), and storing the early warning information of the BTM equipment into a database or a file for the terminal equipment to display.
Different early warning models (regression models, probability statistical models and the like) are used, the accuracy of prediction is influenced, and the specific type of early warning model to be used needs to be determined according to data and services, and meanwhile, the accuracy degree and the use condition of the final prediction of the model need to be considered.
The regression early warning model comprises the following steps: and analyzing the correlation between independent variables (data items) and dependent variables (results obtained according to the independent variables) in the data, establishing a regression equation between the variables on the basis, using the regression equation as an early warning model, and predicting the dependent variables according to the number change of the independent variables in the prediction process.
A probability statistics early warning model: embodied in a probability distribution, is typically described by a mathematical equation relating one or more random variables and other non-random variables.
A terminal display system:
inquiring and displaying the state of the BTM equipment, and when the state of the BTM equipment is poor (for example, the state (performance value score) predicted by the early warning model is lower than a set threshold), automatically sending early warning information of the BTM equipment to a terminal display system by the early warning processing system for warning prompt.
The early warning processing system comprises the following specific processing steps:
1) data pre-processing
And deleting the missing data, and using a part of data as a training data set and a part of data as a test data set.
2) Model training and tuning
Training the early warning model by using the training data set, testing the model by using the testing data set, adjusting the training data set and the testing data set according to the predicted error, training again and checking the error until the predicted error is not reduced or the reduced degree is very small, and stopping the training of the model. And save the parameters of the model.
3) Data early warning
By inputting the latest data into the early warning model, the future equipment state value pred (y) (namely performance value score) is predicted, and when the predicted BTM equipment state (score) is lower than a set threshold value, the early warning information is saved.
And (4) acquiring a threshold value when the state of the BTM equipment is poor by combining the use condition of the field BTM equipment, and giving an alarm prompt when the state of the BTM equipment is predicted to be lower than the threshold value.
The early warning model may be regarded as an equation, where y is a × x, where x is an input parameter vector (e.g., a row vector), a is a parameter vector after model training, a vector set (e.g., a column vector) corresponds to the number of columns of the vector x, and y is a result calculated by the model. The final predicted value of the model can be expressed by the following formula:
Figure BDA0001903618440000041
wherein xiCorresponding to the input column vector described above, several values in the vector are represented by i, and n identifies the number of attributes; beta is aiTraining each attribute item in the corresponding parameter vector a through a model, and then obtaining a parameter when the variance is minimum; beta is a0Is a threshold value that is initially set (initial threshold value setting method see later).
The BTM equipment receives and processes information in the transponder, the relation between the data accuracy of the collected BTM equipment and the speed of the train passing through the transponder is large, when the speed of the train is low, the value of the collected data item (such as decoding times, generally, the number of times of the train stopping on the transponder is 100 times) is large, when the speed of the train is high, the value of the collected data item is small, if all data are used for prediction, the prediction accuracy is low, therefore, when the model is trained, screening is carried out according to the speed, a low speed value is initially determined, when the model is trained and tested, only data higher than the speed value is used for processing, and then the speed value is gradually increased for carrying out the training and the testing of the model again. And acquiring the model with the highest accuracy (the predicted value is closest to the estimated value) as the final early warning model.
The speed of the train passing through the transponder is an important data item for training the BTM early warning model, and meanwhile, in order to enable the model to have higher accuracy, the speed is used as a screening condition to preprocess data in the model training process, and the training data set and the test data set use the data with the speed higher than the specified speed to train and test the model. And adjusting the speed value for multiple times of iterative training, and selecting the model with the minimum variance as the final early warning model.
Due to different line conditions of train running, the same data items of BTM equipment data have larger difference, and in order to improve the accuracy, an independent early warning model is trained for each BTM equipment. Meanwhile, a set of universal early warning model can be provided by integrating a plurality of BTM devices so as to be used by other external devices.
The threshold value determining method comprises the following steps:
and (3) evaluating a uniform value as an initial threshold value according to the opinion of an expert at first, and updating the threshold value according to the state of the BTM equipment in the operation process of the early warning system, for example, after the BTM equipment is early warned, the actual BTM equipment is overhauled, and the threshold value is comprehensively evaluated (according to a predicted value and an expert evaluation value) according to the overhaul maintenance condition and is updated into the early warning system.
Or according to the transponder evaluation system applied by the applicant, combining a large amount of data to obtain performance values when a plurality of devices are in fault, and taking a larger performance value as a threshold value; that is, when the device can be used in large quantities, the device can be determined according to actual conditions, for example, the performance value of the device during the failure is determined, and then the threshold is properly increased.
Example one
1. Data acquisition system
BTM data acquisition mainly comprises the following two modes, namely, index data of the BTM are copied to a system through storage equipment such as a U disk; secondly, the BTM recorder or BTM wirelessly transmits the real-time operation data of the BTM to a system server through a vehicle-to-ground data transmission network, for example, a DSC network with traffic control in a subway, that is, an existing server network that can be authorized to use. The mainstream mode is the first mode at present, but with the development of the network, the second mode has a larger development space; regardless of the manner in which the data is collected, the data is eventually collected to a server for processing.
2. Early warning processing system
After index data of a BTM data acquisition system are received, performance of equipment is scored according to the data of the equipment, multiple people can score, and finally a final performance score is determined according to an average value, wherein a specific calculation method comprises the following steps:
the following is the index data (average calculated from the raw data) as the model training and test data set, where the decoding time refers to the time from the first time the electromagnetic signal is received to the first time the code is correctly decoded, the distance between the transponder and the train is basically one meter, the train is 200 + 300 km/h, and the shorter the time, the better.
Figure BDA0001903618440000051
Figure BDA0001903618440000061
Figure BDA0001903618440000071
1) Training the data of the training data set by using a regression model, and reducing prediction errors by using the data of the test data set to obtain a model formula as follows:
pred(y)≈1.80×x1-0.19×x2+0.19×x3+15.25
wherein x1Representing the decoding order value, x2Decoding time, x3Indicating a good code rate.
2) Results predicted by trained models
The following table is the results predicted from the trained model
Figure BDA0001903618440000072
If the threshold value is 70 points, the BTM number (such as ID, with uniqueness) and the score value with the prediction score lower than 70 points are pushed to a terminal display system for early warning.
3. Terminal display system
And the terminal display system displays according to the received early warning result, and supports display terminals such as a PC (personal computer), a mobile phone, a tablet personal computer and the like.
The scheme of the invention brings remarkable technical advantages: BTM equipment maintainer can know the early warning information of BTM equipment through the terminal, and the suggestion maintainer overhauls the BTM equipment and maintains, discovers potential risk. The maintenance efficiency can be improved, the maintenance cost is reduced, and the safety and reliability of the equipment are improved.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

  1. The BTM equipment early warning system is mainly used for training an early warning model according to received operation data of BTM equipment and predicting the state of the BTM equipment according to the model, and comprises a data acquisition system, an early warning processing system and a terminal display system;
    the data acquisition system acquires and receives data recorded by the BTM equipment or other external detection equipment;
    the data acquisition system collects data into a server for processing, and the server is used by the early warning processing system;
    the early warning processing system preprocesses data according to BTM equipment operation data acquired by the data acquisition system, wherein the operation data comprises data items of good code rate, decoding times, decoding time and train responder passing speed;
    training an early warning model by using the preprocessed data, screening according to the speed when the model is trained, storing model parameters after the model is adjusted and optimized, predicting the state of BTM equipment according to the latest running data, storing the generated early warning information of the BTM equipment into a database or a file, and supplying the early warning information to the terminal display system;
    the terminal display system can inquire and display the performance state of the BTM equipment, when the BTM equipment is in a poor state, the state predicted by the early warning model is lower than a set threshold value, and the early warning processing system automatically sends the early warning information of the BTM equipment to the terminal display system for warning prompt.
  2. 2. The BTM device alert system of claim 1, wherein the preprocessing refers to removing noisy and incomplete data, with a portion of the data being a training data set and a portion of the data being a testing data set.
  3. 3. The BTM equipment early warning system of claim 2, wherein the training and tuning of the early warning model is to train the early warning model with a training data set, test the model with a testing data set, adjust the training data set and the testing data set according to a predicted error, train again and check the error until the predicted error is not reduced, and stop the training of the model.
  4. 4. The BTM equipment early warning system of claim 3, wherein a lower speed value is initially determined during model training, only data higher than the speed value is used for processing during model training and testing, then the speed value is gradually increased to perform model training and testing again, and a model with the highest accuracy is obtained as a final early warning model.
  5. 5. The BTM device early warning system of claim 1, wherein the set threshold is initially evaluated as an initial threshold according to expert opinions, the threshold is updated according to the state of the BTM device during operation of the early warning system, after the BTM device is early warned, actual BTM devices are overhauled, and the threshold is comprehensively evaluated and updated into the early warning system according to the conditions of overhaul and maintenance.
  6. 6. The BTM device alert system of claim 1, wherein the threshold is set by combining a large amount of data to obtain performance values at the time of a plurality of device failures, and taking a performance value with a larger value as a threshold, or determining a performance value at the time of a device failure and then appropriately increasing the threshold.
  7. 7. The BTM device alert system of claim 1, wherein to improve accuracy, an independent alert model is trained for each BTM device, and a common alert model can be provided for other external devices by integrating multiple BTM devices.
  8. 8. The BTM device pre-warning system of claim 1, wherein the data collection system collects BTM data in two ways, one is to copy BTM index data to the server processing system through the mobile storage device; secondly, the BTM recorder or the BTM sends the real-time operation data of the BTM to the early warning processing system in a wireless mode through a vehicle-ground data transmission network; the vehicle-ground data transmission network is an existing server network that can be authorized for use.
  9. 9. The BTM device pre-warning system of claim 1, wherein the terminal display system supports PCs, cell phones, tablets and other display terminal devices.
  10. 10. The BTM device alert system according to claim 1, wherein the transponder system device comprises a BTM device and a transponder device, and the prediction method of the BTM device alert system is also applicable to the state prediction of the transponder device.
CN201811523602.XA 2018-12-13 2018-12-13 BTM equipment early warning system Active CN109625025B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811523602.XA CN109625025B (en) 2018-12-13 2018-12-13 BTM equipment early warning system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811523602.XA CN109625025B (en) 2018-12-13 2018-12-13 BTM equipment early warning system

Publications (2)

Publication Number Publication Date
CN109625025A CN109625025A (en) 2019-04-16
CN109625025B true CN109625025B (en) 2021-02-09

Family

ID=66073430

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811523602.XA Active CN109625025B (en) 2018-12-13 2018-12-13 BTM equipment early warning system

Country Status (1)

Country Link
CN (1) CN109625025B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112541597A (en) * 2019-09-04 2021-03-23 上海杰之能软件科技有限公司 Multi-equipment maintenance method and device, storage medium and terminal
CN110689317A (en) * 2019-09-20 2020-01-14 浙江盾安轨道交通设备有限公司 Intelligent maintenance method and device, computer equipment and storage medium

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101607958B1 (en) * 2014-09-18 2016-04-01 현대로템 주식회사 relay system between cab of ERTMS applied railway vehicle and the signal processing method
CN104219012B (en) * 2014-10-09 2016-08-17 北京交大思诺科技股份有限公司 The EMC test system of transponder transport module and method of testing
CN107547150B (en) * 2016-06-29 2023-05-12 河南蓝信科技有限责任公司 BTM on-line detection method and device thereof
CN205754357U (en) * 2016-06-30 2016-11-30 北京交大思诺科技股份有限公司 Portable transponder response time tester
CN106406295B (en) * 2016-12-02 2019-02-26 南京康尼机电股份有限公司 Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state
CN107862763B (en) * 2017-11-06 2020-12-29 中国人民解放军国防科技大学 Train safety early warning evaluation model training method, module and monitoring evaluation system

Also Published As

Publication number Publication date
CN109625025A (en) 2019-04-16

Similar Documents

Publication Publication Date Title
US10109122B2 (en) System for maintenance recommendation based on maintenance effectiveness estimation
CN109189019B (en) Standardized monitoring system for locomotive crew member value taking
CN110171757A (en) Elevator maintenance system based on the support of Internet of Things big data
WO2011118290A1 (en) System for supporting determination of abnormality of moving object
CN109747685B (en) Fault early warning platform of responder system
CN114267178B (en) Intelligent operation maintenance method and device for station
CN109625025B (en) BTM equipment early warning system
CN112449696A (en) Time series data diagnosis device, additional learning method, and program
CN112232405B (en) Fault prediction, monitoring and diagnosis method of gearbox and corresponding device
KR102078654B1 (en) System and method for predicting error of electric rail car
CN111806516A (en) Health management device and method for intelligent train monitoring and operation and maintenance
CN112862233A (en) Fault relevance analysis system and method based on Internet of vehicles data
CN101149780A (en) System and method for automatic measurement for radio frequency identification service system
CN110809280B (en) Detection and early warning method and device for railway wireless network quality
CN112381958A (en) ETC portal system state online monitoring method and system
CN116279700A (en) Method for realizing fault early warning and management of electronic related equipment of railway signal system
KR20050060251A (en) Prevention system and the method for maintenance of railway car using rams analysis
CN108960220B (en) Signal system communication data analysis method for reliability based on state machine model
CN112040433B (en) Data processing method and device
CN116546534A (en) Remote ESIM card data air operation and maintenance method
CN116452154A (en) Project management system suitable for communication operators
CN114167837B (en) Intelligent fault diagnosis method and system for railway signal system
US20010047416A1 (en) Network administration system and method of re-arranging network resources
CN114861830A (en) Automatic fault diagnosis method for vehicle-mounted equipment of subway signal system
CN115171242A (en) Remote fault detection system for industrial vehicle

Legal Events

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