CN110111446A - A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition - Google Patents
A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition Download PDFInfo
- Publication number
- CN110111446A CN110111446A CN201910070738.8A CN201910070738A CN110111446A CN 110111446 A CN110111446 A CN 110111446A CN 201910070738 A CN201910070738 A CN 201910070738A CN 110111446 A CN110111446 A CN 110111446A
- Authority
- CN
- China
- Prior art keywords
- data
- failure
- speed rail
- real
- vehicle
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/008—Registering or indicating the working of vehicles communicating information to a remotely located station
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0808—Diagnosing performance data
Abstract
The present invention proposes a kind of adaptive acquisition system of high-speed rail train data and method based on operating condition, comprising: ground system, vehicle-mounted TCMS system and radio transmitting device;When there is no when failure, vehicle-mounted TCMS system is saved in the historical data base of ground system normal operation by sending data in the normal operation data buffer area that sample rate is T1 in radio transmitting device terrestrial system for high-speed rail train;When high-speed rail train breaks down, vehicle-mounted TCMS system is stored data into ground system Mishap Database by sending data in the fault data buffer area that sample rate is T2 in radio transmitting device terrestrial system;The present invention monitors for high-speed rail train fault and diagnosis provides more abundant historical failure case data, more reasonably uses onboard system hard disk resources, provides the more effective adaptive acquisition method of high-speed rail train data.
Description
Technical field
The invention belongs to the acquisition of high-speed rail train data, analysis and fault diagnosis technology fields, and in particular to one kind is based on fortune
The adaptive acquisition system of high-speed rail train data and method of row operating condition.
Background technique
The operation average speed about 300km/h of China's high-speed rail train will cause catastrophic effect once breaking down.Institute
It is particularly significant with the data of high-speed rail train at runtime.Wherein, it may determine that the health status of bearing by monitoring bearing temperature,
It can be seen that the importance of bearing temperature data.And historical failure case number of cases abundant is not only needed in the fault diagnosis of high-speed rail train
According to, and microcosmic waveform is needed failure is described in detail, so needing the data of quick sampling rate.
Existing high-speed rail train data acquisition method is to transmit data to ground by wireless-transmission network by onboard system
Plane system.Equipment information interaction uses MVB (Multifunction Vehicle Bus) to vehicle-mounted TCMS system inside the vehicle
Multifunctional vehicle bus acquires sensing data, and data are transmitted high-speed rail train radio transmitting device.Data pass through
The public network platforms such as GPRS, 3G/4G, WLAN local area network and railway are unified MQ (Message Queue) transmission platform and are real-time transmitted to
Ground system, when encountering tunnel vehicle-mounted TCMS system by the data buffer storage of acquisition in systems.But due to onboard system
The limitation of network speed in limited storage space and network transmission process, the data sampling rate stored at present are slower.And in height
The data using fast sample rate are needed in the monitoring of iron train fault and diagnosis, so current high-speed rail train data acquisition method is not
It is enough to provide historical failure case data abundant and fault waveform microcosmic enough.
Summary of the invention
In view of the above shortcomings of the prior art, it is adaptive to provide a kind of high-speed rail train data based on operating condition by the present invention
Acquisition system and method provide more abundant historical failure case data for the monitoring of high-speed rail train fault and diagnosis, more rationally
Use onboard system hard disk resources, provide the more effective adaptive acquisition method of high-speed rail train data.
The adaptive acquisition method thought of high-speed rail train data, which is carried out, using high-speed rail train bearing temperature data illustrates that thought is such as
Under: firstly, determining storage rule using dynamic internal model pivot analysis algorithm.Secondly, using multi-direction in data storage procedure
Reconstructing method finds failure variable, adds faulty tag while storing data for failure variable.
In summary thought, the present invention propose a kind of adaptive acquisition system of high-speed rail train data based on operating condition and
Method, specially a kind of adaptive acquisition system of high-speed rail train data based on operating condition, comprising: ground system, vehicle-mounted
TCMS system and radio transmitting device;
Ground system includes: the historical data base of Mishap Database, normal operation;
Vehicle-mounted TCMS system includes: normal operation data buffer area, the sample rate T2 that malfunction monitoring area, sample rate are T1
Fault data buffer area;
The historical data base operated normally in ground system is connected with vehicle-mounted TCMS system malfunction monitoring area, vehicle-mounted TCMS
The fault data buffer area that the normal operation data buffer area and sample rate that sample rate is T1 in system are T2 is passed with wireless respectively
Defeated device is connected, and radio transmitting device is connected with the historical data base of Mishap Database in ground system and normal operation respectively
It connects;
The historical data operated normally in the historical data base of normal operation is sent vehicle-mounted TCMS system by ground system
Malfunction monitoring area in, malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure
Signature, learns whether high-speed rail train breaks down from failure signature;When high-speed rail train is there is no when failure, vehicle
TCMS system is carried by sending number in the normal operation data buffer area that sample rate is T1 in radio transmitting device terrestrial system
According to, and be saved in the historical data base of ground system normal operation;When high-speed rail train breaks down, vehicle-mounted TCMS system is logical
It crosses in radio transmitting device terrestrial system and sends data in the fault data buffer area that sample rate is T2, and store data into
In ground system Mishap Database, until failure starts the data whole end of transmission terminated to failure, restore vehicle-mounted TCMS system
System is saved by sending data in the normal operation data buffer area that sample rate is T1 in radio transmitting device terrestrial system
In the historical data base operated normally to ground system;
Mishap Database saves the fault data of high-speed rail train;
The historical data base of normal operation saves the historical data that high-speed rail train operates normally;
Malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure variable mark
Label;
Sample rate is the normal operation data buffer area of T1, saves the data that high-speed rail operates normally by sample rate of T1;
Sample rate is the fault data buffer area of T2, using T2 as the data of sample rate preservation high-speed rail failure operation, and T2 >
T1;
Fault data in vehicle-mounted TCMS system is transmitted to Mishap Database in ground system by radio transmitting device, will just
Regular data is transmitted to the historical data base operated normally in ground system;Unify MQ including GPRS, 3G/4G public network platform, railway
Transmission platform, if the WLAN local network transport platform for utilizing station if still faulty into station;
A method of the adaptive acquisition system of high-speed rail train data based on operating condition, using one kind based on operation work
The adaptive acquisition system of the high-speed rail train data of condition is realized, is specifically comprised the following steps:
Step 1: vehicle-mounted TCMS system receives the bearing temperature history that the high-speed rail train that ground system transmission comes operates normally
Data, as training set;In malfunction monitoring area, training set is established high-speed rail train by dynamic internal model pivot analysis algorithm and is normally transported
Bearing temperature data fault monitoring model when row;
Wherein, xiBearing temperature sample for historical data in moment i, tiFor xiThe dynamic latent variable extracted in moment i
By ti=xiP is calculated, and P is dynamic load matrix,For dynamic latent variable tiEstimated value, can be by the dynamic at the preceding m moment of i
Latent variable is linearly expressed, and β is linear coefficient, viIt is the prediction error of i moment dynamic latent variable, βjFor the vector that the jth of β arranges, j
=1,2 ..., m, eiFor xiIt is extracted dynamic latent variable tiStatic residual error portion later, PrFor eiBy static pivot analysis
Algorithm models obtained load matrix, tr,iFor eiDynamic latent variable, er,iFor eiAfter the modeling of static pivot analysis algorithm
Static residual error portion;
Step 2: the bearing temperature data acquired in real time are acquired as test set by malfunction monitoring model monitoring in real time
Bearing temperature data detect whether to break down;When high-speed rail train does not have failure, storage rule is vehicle-mounted TCMS system
Storage temperature sensor saves the bearing temperature data that high-speed rail operates normally by sample rate of T1, after high-speed rail train breaks down
And continue in period that failure terminates, storage rule is that vehicle-mounted TCMS system storage temperature sensor is protected by sample rate of T2
Deposit the bearing temperature data of high-speed rail failure operation, and T2 > T1;
Step 2.1: using the monitoring of the real-time bearing temperature data dynamic part acquired to temperature sensor, at the k moment
High-speed rail train bearing temperature sample xkExpression formula are as follows:
Wherein, tkFor xkThe dynamic latent variable extracted in moment k is by tk=xkP is calculated, and P is dynamic load matrix,For
Dynamic latent variable tiEstimated value, ekFor xkIt is extracted dynamic latent variable tkStatic residual error portion later, β are linear coefficient,
βjFor the vector that the jth of β arranges, j=1,2 ..., m, PrFor ekObtained load matrix is modeled by static pivot analysis algorithm,
tr,kFor ekDynamic latent variable, er,kFor ekStatic residual error portion after the modeling of static pivot analysis algorithm;
Can find out from above formula to the monitoring of real-time data collection be bytr,k、er,kIt determines, but becauseIt is
It is dynamic and and it is unstable, will cause higher rate of false alarm if be monitored to it, for vkFor, it between each variable still
There is correlation, so using v in the dynamic part of real-time data collectionkTo replace pairMonitoring, to vkIt carries out static main
The modeling monitoring of meta analysis algorithm, uses overall target in monitoring processFault detection is carried out to real-time data collection, it is comprehensive
IndexExpression formula are as follows:
Wherein, QvAnd Tv 2Respectively vkT2Statistic and Q statistical magnitude,WithRespectively vkT2Statistic and Q statistics
The control of amount limits, ΦvFor symmetrical positively definite matrix, I is unit matrix, ΛvError is predicted for the dynamic latent variable of real-time data collection
The pivot covariance matrix of v, PvPredict that error component static state pivot analysis algorithm models the resulting moment of load for dynamic latent variable
Battle array;Overall targetControl limitExpression formula are as follows:
Wherein,gvForWithBetween coefficient of relationship, SvFor the covariance of v
Matrix;Be freedom degree be hv, confidence level be 0.95 chi square distribution critical value;WhenWhen, then it is assumed that in dynamic part
Failure has occurred;
When vehicle-mounted TCMS system detects that failure occurs since real-time data collection, vehicle-mounted TCMS system storage temperature
Sensor saves the bearing temperature data of high-speed rail failure operation using T2 as sample rate, until failure terminates.Because failure still may
Occur in the static part of real-time data collection, so needing step 2.2.
Step 2.2: using the monitoring of the real-time bearing temperature data inactivity part acquired to temperature sensor, to real-time
After acquiring data progress dynamic latent variable extraction, ekFor static residual error portion, to ekStatic pivot analysis algorithm modeling is carried out, is built
Formwork erection type are as follows:
ek=Prtr,k+er,k
Overall target is used in static residual error portionThe static part of real-time data collection is monitored, overall targetExpression formula are as follows:
Wherein, QrAnd Tr 2Respectively ekT2Statistic and Q statistical magnitude, ΦrFor symmetrical positively definite matrix,WithRespectively
ekT2The control of statistic and Q statistical magnitude limits, and I is unit matrix, ΛrFor the master of the static residual error portion e of real-time data collection
First covariance matrix;Overall targetControl limit be written as:
Wherein,SrFor the covariance matrix of e;WhenWhen, then it is assumed that
Failure has occurred in static part.It is identical when breaking down with the dynamic part in step 2.1, when vehicle-mounted TCMS system is adopted from real time
Detect that static part failure starts in collection data, vehicle-mounted TCMS system storage temperature sensor is saved by sample rate of T2
The bearing temperature data of high-speed rail failure operation, until failure terminates.
Step 3: in malfunction monitoring area, determining failure signature using multi-direction reconstructing method;After the failure occurred, sharp
Failure variable is determined with multi-direction reconstructing method, and is added while vehicle-mounted TCMS system saves in data to data buffer area
Failure signature.
Step 4: vehicle-mounted TCMS system sends data to ground system in real time;Data are stored data into vehicle-mounted TCMS system
While in buffer area, the data of storage are sent to high-speed rail train radio transmitting device, pass through GPRS, 3G/4G public network platform
Unify MQ (Message Queue) transmission platform with railway and be sent to ground system in real time, passes through when high-speed rail train enters the station
WLAN local network transport data are to ground system.
Determine that the multi-direction reconstructing method thought of failure signature is as follows in the step 3:
As the real-time data collection sample x for detecting the failure generation momentfAfterwards, the direction weight along the i-th column is carried out to it
Structure, the sample after reconstruct are as follows:
zi=xf-ξifi
Wherein ξiIt is the unit column vector that the i-th row element is 1, fiFor the failure size on fault direction i.Sample after reconstruct
This ziOverall targetValue is needed in control limit hereinafter, overall targetFor formula:
Wherein, Φ is symmetrical positively definite matrix, is written as:
Wherein, I is unit matrix, and Λ is the pivot covariance matrix of training set, and P is the load matrix of training set, δ2And χ2
Respectively ziT2The control of statistic and Q statistical magnitude limits;
By overall target to failure size fiDerivation can find out the comprehensive index value of faulty componentIt can be written as public affairs
Formula 11:
Wherein,Indicate indexAlong the contribution after the reconstruct of variable i direction.
Because failure variable may be the ξ in multiple bearing temperature variable institutes above formulaiIt can be indicated by matrix Ξ, by above formula
It rewrites are as follows:
Wherein, ()+For a mole Peng Ruosi generalized inverse, the overall target after reconstruct is indicated are as follows:Work as time
After selecting failure variable to determine, it is equipped with l Candidate Fault variable, its contribution degree Cont is asked respectively to l Candidate Fault variablejIt writes
Are as follows:
The step 3 the following steps are included:
Step 3.1: determining real-time data collection in the failure variable of dynamic part;For the dynamic state part of real-time data collection
Point, vkThe bearing temperature data sample x acquired in real time for the k momentkDynamic latent variable prediction error, because of vkIn pivot
In subspace, so needing fault moment f by vfIt is projected back in in original variable space, vfFor f moment sample xfDynamic it is latent
The prediction error of variable;The dynamic part of real-time data collection breaks down the original variable space x of moment fv,fExpression formula such as
Under:
xv,f=(w (pTw)-1)vf
Calculate the contribution degree Cont of the dynamic part Candidate Fault variable of real-time data collectionjIt is written as:
Dynamic part failure variable number can be determined by the size of contribution degree and determines failure signature, deposited
Failure signature is added in the data of storage during storage real-time data collection.Due to the static state in real-time data collection
Part can also break down, so needing the failure variable of the static part of the determining real-time data collection of step 3.2.
Step 3.2: determining real-time data collection in the failure variable of static part;It is latent that real-time data collection is extracted dynamic
Static residual error portion after variable is ef, calculate the contribution degree Cont of the static part Candidate Fault variable of real-time data collectioni
Expression formula is as follows:
Static part failure variable number can be determined by contribution degree size and determines failure signature, stored
Failure signature is added in the data of storage during real-time data collection.
When high-speed rail train does not break down, the real-time data collection of sample rate T1 is transmitted to the ground using public network platform,
Since network speed limits, transmission data might have lag, but when vehicle-mounted TCMS system detects that failure occurs, stop passing
The data of defeated sample rate T1 start to transmit the temperature sensor of the sample rate T2 of vehicle-mounted TCMS system storage collected number in real time
According to, until failure starts the data end of transmission terminated to failure, restore transmission sample rate it is slower be buffered in vehicle-mounted TCMS system
Data in system.When passing through a station, data of the quick transmission buffer of WLAN in vehicle-mounted TCMS system are used.
Advantageous effects:
In view of the above shortcomings of the prior art, it is adaptive to provide a kind of high-speed rail train data based on operating condition by the present invention
Acquisition method provides more abundant historical failure case data for the monitoring of high-speed rail train fault and diagnosis, more reasonably uses
Onboard system hard disk resources provide the more effective adaptive acquisition method of high-speed rail train data.
Existing collecting method is the acquisition that data are carried out by certain low speed sample frequency, because of existing number
Even if fault case data can not be collected according to the too low onboard system discovery failure of acquisition method sample frequency.So cannot
Historical failure case data abundant is provided, and because low speed sample frequency system can not provide effective microcosmic fault wave
Shape.Further more, the memory that existing collecting method occupies onboard system is high.Pass through the high-speed rail based on operating condition of this patent
The adaptive acquisition method of train data can reduce the use memory of onboard system, store more train operating datas, provide
Fault waveform and historical failure case data abundant microcosmic enough.
Detailed description of the invention
Fig. 1 is the structure chart of the data collection strategy of the embodiment of the present invention;
Fig. 2 is case sample axle box provided in an embodiment of the present invention and motor stator temperature;
Fig. 3 is the data segment index that case sample train monitoring needs provided in an embodiment of the present invention save high-frequency acquisition
Figure;
Fig. 4 is the failure that case sample provided in an embodiment of the present invention is determined based on the multidirectional reconstruct contribution plot of monitoring index
Signature;
Fig. 5 is the failure variable that the multidirectional reconstruct contribution plot of case sample monitoring index provided in an embodiment of the present invention determines
Label;
Fig. 6 is a kind of method of adaptive acquisition system of high-speed rail train data based on operating condition of the embodiment of the present invention
Flow chart.
Specific embodiment
Invention is described further with specific implementation example with reference to the accompanying drawing;
A kind of adaptive acquisition system of high-speed rail train data based on operating condition, as shown in Figure 1, comprising: ground system,
Vehicle-mounted TCMS system and radio transmitting device;
Ground system includes: the historical data base of Mishap Database, normal operation;
Vehicle-mounted TCMS system includes: normal operation data buffer area, the sample rate T2 that malfunction monitoring area, sample rate are T1
Fault data buffer area;
The historical data base operated normally in ground system is connected with vehicle-mounted TCMS system malfunction monitoring area, vehicle-mounted TCMS
The fault data buffer area that the normal operation data buffer area and sample rate that sample rate is T1 in system are T2 is passed with wireless respectively
Defeated device is connected, and radio transmitting device is connected with the historical data base of Mishap Database in ground system and normal operation respectively
It connects;
The historical data operated normally in the historical data base of normal operation is sent vehicle-mounted TCMS system by ground system
Malfunction monitoring area in, malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure
Signature, when there is no when failure, vehicle-mounted TCMS system in radio transmitting device terrestrial system by sending out for high-speed rail train
Data in the normal operation data buffer area that sample rate is T1 are sent, and are saved in the historical data base of ground system normal operation
In;When high-speed rail train breaks down, vehicle-mounted TCMS system is by sending sample rate in radio transmitting device terrestrial system
Data in the fault data buffer area of T2, and store data into ground system Mishap Database, until failure starts to event
Hinder the data whole end of transmission terminated, restores vehicle-mounted TCMS system by sending sampling in radio transmitting device terrestrial system
Rate is data in the normal operation data buffer area of T1, and is saved in the historical data base of ground system normal operation;
Mishap Database saves the fault data of high-speed rail train;
The historical data base of normal operation saves the historical data that high-speed rail train operates normally;
Malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure variable mark
Label;
Sample rate is the normal operation data buffer area of T1, saves the data that high-speed rail operates normally by sample rate of T1;
Sample rate is the fault data buffer area of T2, using T2 as the data of sample rate preservation high-speed rail failure operation, and T2 >
T1;
Fault data in vehicle-mounted TCMS system is transmitted to Mishap Database in ground system by radio transmitting device, will just
Regular data is transmitted to the historical data base operated normally in ground system;Unify MQ including GPRS, 3G/4G public network platform, railway
Transmission platform, if the WLAN local network transport platform for utilizing station if still faulty into station.
The embodiment of the present invention is the high-speed rail train bearing temperature data adaptive acquisition based on operating condition.The present embodiment
Middle high-speed rail train shares 36 temperature sensors and respectively corresponds 36 critical positions on high-speed rail train, and 36 temperature sensors are used for
Monitor the working condition of train bearing, it would be desirable to bearing temperature be carried out to this 36 temperature sensor variables and adaptively acquired.
A method of the adaptive acquisition system of high-speed rail train data based on operating condition, using one kind based on operation work
The adaptive acquisition system of the high-speed rail train data of condition is realized, as shown in fig. 6, specifically comprising the following steps:
Step 1: vehicle-mounted TCMS system receives the bearing temperature history that the high-speed rail train that ground system transmission comes operates normally
Data, as training set;In malfunction monitoring area, training set is established high-speed rail train by dynamic internal model pivot analysis algorithm and is normally transported
Bearing temperature data fault monitoring model when row;
Wherein, xiBearing temperature sample for historical data in moment i, tiFor xiThe dynamic latent variable extracted in moment i
By ti=xiP is calculated, and P is dynamic load matrix,For dynamic latent variable tiEstimated value, can be dived by the dynamic at the preceding m moment of i
Linear variable displacement expression, β is linear coefficient, viIt is the prediction error of i moment dynamic latent variable, βjFor the vector that the jth of β arranges, j=
1,2 ..., m, eiFor xiIt is extracted dynamic latent variable tiStatic residual error portion later, PrFor eiIt is calculated by static pivot analysis
Method models obtained load matrix, tr,iFor eiDynamic latent variable, er,iFor eiIt is quiet after the modeling of static pivot analysis algorithm
State residual error portion;
Step 2: the bearing temperature data acquired in real time are acquired as test set by malfunction monitoring model monitoring in real time
Bearing temperature data detect whether to break down;When high-speed rail train does not have failure, storage rule is vehicle-mounted TCMS system
Storage temperature sensor saves the bearing temperature data that high-speed rail operates normally by sample rate of T1, after high-speed rail train breaks down
And continue in period that failure terminates, storage rule is that vehicle-mounted TCMS system storage temperature sensor is protected by sample rate of T2
Deposit the bearing temperature data of high-speed rail failure operation, and T2 > T1;
Step 2.1: using the monitoring of the real-time bearing temperature data dynamic part acquired to temperature sensor, at the k moment
High-speed rail train bearing temperature sample xkExpression formula are as follows:
Wherein, tkFor xkThe dynamic latent variable extracted in moment k is by tk=xkP is calculated, and P is dynamic load matrix,For
Dynamic latent variable tiEstimated value, ekFor xkIt is extracted dynamic latent variable tkStatic residual error portion later, β are linear coefficient,
βjFor the vector that the jth of β arranges, j=1,2 ..., m, PrFor ekObtained load matrix is modeled by static pivot analysis algorithm,
tr,kFor ekDynamic latent variable, er,kFor ekStatic residual error portion after the modeling of static pivot analysis algorithm;
Can find out from above formula to the monitoring of real-time data collection be bytr,k、er,kIt determines, but becauseIt is
State and and it is unstable, will cause higher rate of false alarm if be monitored to it, for vkFor, it still has between each variable
Correlation, so using v in the dynamic part of real-time data collectionkTo replace pairMonitoring, to vkCarry out static pivot point
Algorithm modeling monitoring is analysed, uses overall target in monitoring processFault detection is carried out to real-time data collection, according to as follows
Formula can calculate the comprehensive index value of k moment dynamic part
Wherein, QvAnd Tv 2Respectively vkT2Statistic and Q statistical magnitude,WithRespectively vkT2Statistic and Q statistics
The control of amount limits, ΦvFor symmetrical positively definite matrix, I is unit matrix, ΛvError is predicted for the dynamic latent variable of real-time data collection
The pivot covariance matrix of v, PvPredict that error component static state pivot analysis algorithm models the resulting moment of load for dynamic latent variable
Battle array;
Comprehensive index valueControl limit can be calculated by following formula:
Wherein,gvForWithBetween coefficient of relationship, SvFor the covariance of v
Matrix;Be freedom degree be hv, confidence level be 0.95 chi square distribution critical value;WhenWhen, then it is assumed that in dynamic part
Failure has occurred.
When failure starts, vehicle-mounted TCMS system starts to store the fast bearing temperature data to break down of acquisition rate until event
Barrier terminates.
Step 2.2: using the monitoring of the real-time bearing temperature data inactivity part acquired to temperature sensor, to real-time
After acquiring data progress dynamic latent variable extraction, ekFor static residual error portion, to ekStatic pivot analysis algorithm modeling is carried out, is built
Formwork erection type are as follows:
ek=Prtr,k+er,k
Overall target is used in static residual error portionThe static part of real-time data collection is monitored, according to as follows
Formula can calculate the comprehensive index value of k moment static part
Wherein, QrAnd Tr 2Respectively ekT2Statistic and Q statistical magnitude, ΦrFor symmetrical positively definite matrix,WithRespectively
ekT2The control of statistic and Q statistical magnitude limits, and I is unit matrix, ΛrFor the master of the static residual error portion e of real-time data collection
First covariance matrix;Comprehensive index valueControl limit can be calculated by following formula:
Wherein,SrFor the covariance matrix of e;WhenWhen, then it is assumed that
Failure has occurred in static part.
When failure starts, vehicle-mounted TCMS system starts to store the fast bearing temperature data to break down of acquisition rate until event
Barrier terminates.
Step 3: in malfunction monitoring area, determining failure signature using multi-direction reconstructing method;After the failure occurred, sharp
Failure variable is determined with multi-direction reconstructing method, and is added while vehicle-mounted TCMS system saves in data to data buffer area
Failure signature.
Step 3.1: determining real-time data collection in the failure variable of dynamic part;The dynamic state part of real-time data collection is distributed
The original variable space x of raw fault moment fv,fExpression formula it is as follows:
xv,f=(w (pTw)-1)vf
Calculate the contribution degree Cont of the dynamic part Candidate Fault variable of real-time data collectionj, expression formula is as follows:
Dynamic part failure variable number can be determined by the size of contribution degree and determines failure signature, deposited
Failure signature is added in the data of storage during storage real-time data collection.
Step 3.2 determines real-time data collection in the failure variable of static part;It is latent that real-time data collection is extracted dynamic
Static residual error portion after variable is ef, calculate the contribution degree Cont of the static part Candidate Fault variable of real-time data collectioni
Expression formula is as follows:
Static part failure variable number can be determined by contribution degree size and determines failure signature, stored
Failure signature is added in the data of storage during real-time data collection.
In order to embody contribution degree size, the dynamic part and static part contribution plot of real-time data collection, tribute are drawn respectively
Offering figure abscissa is sample number, and ordinate corresponds to 36 bearing vibrations of train, and contribution degree is indicated by the light levels degree of color, face
The brighter expression of color, contribution degree are bigger.Failure variable contribution degree can the obvious bright contribution degree in its dependent variable.It is possible thereby to determine event
Hinder signature.
Step 4: vehicle-mounted TCMS system sends data to ground system in real time;Data are stored data into vehicle-mounted TCMS system
While in buffer area, the data of storage are sent to high-speed rail train radio transmitting device, pass through GPRS, 3G/4G public network platform
Unify MQ (Message Queue) transmission platform with railway and be sent to ground system in real time, passes through when high-speed rail train enters the station
WLAN local network transport data are to ground system.
When high-speed rail train does not break down, the slower real-time acquisition number of sample rate is transmitted to the ground using public network platform
According to.Since network speed limits, transmission data might have lag, but when vehicle-mounted TCMS system detects that failure occurs, stop
The only slower data of transmission sample rate, the faster temperature sensor of sample rate for starting to transmit vehicle-mounted TCMS system storage are adopted in real time
The data collected start the data end of transmission terminated to failure until failure, restore transmission sample rate it is slower be buffered in it is vehicle-mounted
Data in TCMS system.When passing through a station, data of the quick transmission buffer of WLAN in vehicle-mounted TCMS system are used.
In the present embodiment, sample number 12000, temperature sensor sampling period are 1s, share 36 variables in sample
It as shown in Fig. 2, failure takes place in variable 17 since at 622 samples, and is more than alarm limit at 630-634 sample, in sample
Restore at sheet 724 normal.And starts sustained fault bearing temperature at sample 952 persistently and transfinite and is extensive at sample 11955
It is multiple normal.In high-speed rail train normally travel, data storage rule is that vehicle-mounted TCMS system storage temperature sensor sample rate is slower
Bearing temperature data, after the failure occurred and continue to that storage rule is that vehicle-mounted TCMS system is protected in period that failure terminates
Deposit the faster bearing temperature data of temperature sensor sampling rate.
Two monitoring indexes of train travelling process bearing stateWithAs shown in Figure 2, it can be seen that in sample 622
Place, the failure of variable 17 occur successfully to be detected in dynamic and static two parts by two indexs.The failure has continueed to sample
At 724, it can be seen that Fig. 2 dashed rectangle period breaks down in dynamic and static part, it can be seen that failure in Fig. 3
QuiltWithIndex successfully detects, and passes through the vehicle-mounted faster bearing temperature data of TCMS system storage temperature sensor sample rate.
It is arrived at sample 951 at sample 725, as shown in Figure 3 two monitoring indexesWithIt does not transfinite, i.e. solid line boxes in Fig. 2
Period, vehicle-mounted TCMS system restores the slower bearing temperature data of storage temperature sensor sample rate at this time.At sample 952
Start, variable 17 starts persistently to break down, and has mark in Fig. 2, and ultra-high temperature alarming line this failure continues to sample 11955
Place, this failure occurs in static part as shown in Figure 3, byMonitoring Indexes success.Data collector starts to save temperature at this time
Spending the bearing temperature data of sensor high-frequency sampling, to restore storage temperature sensor at the sample 11955 after failure vanishes low
The bearing temperature data of frequency sampling.
Fig. 3 is respectively monitoring indexWithMultidirectional reconstruct contribution plot, can mainly be occurred with the failure of variable 17 from Fig. 3
In static part, it can be seen that the contribution of the front-end and back-end variable 17 of sample is significantly greater than its dependent variable in Fig. 4, so dynamic
The failure signature of polymorphic segment is determined as variable 17, it is obvious that variable 17 in the failure of static part in Fig. 5
Contribution degree is significantly greater than its dependent variable so being confirmed as failure signature.
As can be seen from the above embodiments, the adaptive acquisition method success of high-speed rail train data based on operating condition is in height
Storage temperature sensor acquires fast bearing temperature data in the period that iron train breaks down.And it can successfully avoid passing
The problems such as high-frequency data does not save when system rule-based approach, and can complete to add event while saving fault data
The hindering label of the task, historical failure case data abundant can be provided for researcher and can be mentioned in the failure period of right time
For fault waveform microcosmic enough.
Claims (4)
1. a kind of adaptive acquisition system of high-speed rail train data based on operating condition characterized by comprising ground system,
Vehicle-mounted TCMS system and radio transmitting device;
Ground system includes: the historical data base of Mishap Database, normal operation;
Vehicle-mounted TCMS system include: malfunction monitoring area, sample rate be T1 normal operation data buffer area, sample rate be T2 event
Hinder data buffer area;
The historical data base operated normally in ground system is connected with vehicle-mounted TCMS system malfunction monitoring area, vehicle-mounted TCMS system
The fault data buffer area that the normal operation data buffer area and sample rate that middle sample rate is T1 are T2 is filled with wireless transmission respectively
It sets and is connected, radio transmitting device is connected with the historical data base of Mishap Database in ground system and normal operation respectively;
Ground system sends the historical data operated normally in the historical data base of normal operation to the event of vehicle-mounted TCMS system
Hinder in monitoring section, malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure variable
Label, learns whether high-speed rail train breaks down from failure signature;It is vehicle-mounted when high-speed rail train is there is no when failure
TCMS system is data in the normal operation data buffer area of T1 by transmission sample rate in radio transmitting device terrestrial system,
And it is saved in the historical data base of ground system normal operation;When high-speed rail train breaks down, vehicle-mounted TCMS system passes through
Data in the fault data buffer area that sample rate is T2 are sent in radio transmitting device terrestrial system, and store data into ground
In plane system Mishap Database, until failure starts the data whole end of transmission terminated to failure, restore vehicle-mounted TCMS system
By sending data in the normal operation data buffer area that sample rate is T1 in radio transmitting device terrestrial system, and it is saved in
In the historical data base that ground system operates normally;
Mishap Database saves the fault data of high-speed rail train;
The historical data base of normal operation saves the historical data that high-speed rail train operates normally;
Malfunction monitoring area establishes malfunction monitoring model with historical data using normal operation, and determines failure signature;
Sample rate is the normal operation data buffer area of T1, saves the data that high-speed rail operates normally by sample rate of T1;
Sample rate is the fault data buffer area of T2, and the data of high-speed rail failure operation, and T2 > T1 are saved using T2 as sample rate;
Fault data in vehicle-mounted TCMS system is transmitted to Mishap Database in ground system, by normal number by radio transmitting device
According to being transmitted to the historical data base operated normally in ground system.
2. a kind of adaptive acquisition system of high-speed rail train data based on operating condition, feature exist according to claim 1
In the radio transmitting device, including GPRS, 3G/4G public network platform, railway unify MQ transmission platform, if still having into station
Failure then utilizes the WLAN local network transport platform at station.
3. a kind of method that the high-speed rail train data based on operating condition adaptively acquires, a kind of base according to claim 1
It is realized in the adaptive acquisition system of the high-speed rail train data of operating condition, which is characterized in that specifically comprise the following steps:
Step 1: vehicle-mounted TCMS system receives the bearing temperature historical data that the high-speed rail train that ground system transmission comes operates normally,
As training set;In malfunction monitoring area, when training set establishes high-speed rail train normal operation by dynamic internal model pivot analysis algorithm
Bearing temperature data fault monitoring model;
Wherein, xiBearing temperature sample for historical data in moment i, tiFor xiThe dynamic latent variable extracted in moment i is by ti
=xiP is calculated, and P is dynamic load matrix,For dynamic latent variable tiEstimated value, can be by the dynamic creep at the preceding m moment of i
The linear expression of amount, β is linear coefficient, viIt is the prediction error of i moment dynamic latent variable, βjFor β jth arrange vector, j=1,
2 ..., m, eiFor xiIt is extracted dynamic latent variable tiStatic residual error portion later, PrFor eiBy static pivot analysis algorithm
Model obtained load matrix, tr,iFor eiDynamic latent variable, er,iFor eiStatic state after the modeling of static pivot analysis algorithm
Residual error portion;
Step 2: the bearing temperature data acquired in real time are as test set, the bearing acquired in real time by malfunction monitoring model monitoring
Temperature data detects whether to break down;When high-speed rail train does not have failure, storage rule is the preservation of vehicle-mounted TCMS system
Temperature sensor saves the bearing temperature data that high-speed rail operates normally by sample rate of T1, after high-speed rail train breaks down and holds
Continue in the period that failure terminates, storage rule is that vehicle-mounted TCMS system storage temperature sensor saves height by sample rate of T2
The bearing temperature data of iron failure operation, and T2 > T1;
Step 2.1: using the monitoring of the real-time bearing temperature data dynamic part acquired to temperature sensor, in k moment high-speed rail
Train bearing temperature sample xkExpression formula are as follows:
Wherein, tkFor xkThe dynamic latent variable extracted in moment k is by tk=xkP is calculated, and P is dynamic load matrix,Dynamically to dive
Variable tiEstimated value, ekFor xkIt is extracted dynamic latent variable tkStatic residual error portion later, β are linear coefficient, βjFor β's
The vector of jth column, j=1,2 ..., m, PrFor ekObtained load matrix, t are modeled by static pivot analysis algorithmr,kFor ek
Dynamic latent variable, er,kFor ekStatic residual error portion after the modeling of static pivot analysis algorithm;
V is used in the dynamic part of real-time data collectionkTo replace pairMonitoring, to vkCarry out static pivot analysis algorithm
Modeling monitoring, uses overall target in monitoring processFault detection, overall target are carried out to real-time data collectionTable
Up to formula are as follows:
Wherein, QvWithRespectively vkT2Statistic and Q statistical magnitude,WithRespectively vkT2Statistic and Q statistical magnitude
Control limit, ΦvFor symmetrical positively definite matrix, I is unit matrix, ΛvPredict error v's for the dynamic latent variable of real-time data collection
Pivot covariance matrix, PvPredict that error component static state pivot analysis algorithm models resulting load matrix for dynamic latent variable;
Overall targetControl limitExpression formula are as follows:
Wherein,gvForWithBetween coefficient of relationship, SvFor the covariance matrix of v;Be freedom degree be hv, confidence level be 0.95 chi square distribution critical value;WhenWhen, then it is assumed that it is had occurred in dynamic part
Failure;
Step 2.2: using the monitoring of the real-time bearing temperature data inactivity part acquired to temperature sensor, to real-time acquisition
After data carry out the extraction of dynamic latent variable, ekFor static residual error portion, to ekStatic pivot analysis algorithm modeling is carried out, mould is established
Type are as follows:
ek=Prtr,k+er,k
Overall target is used in static residual error portionThe static part of real-time data collection is monitored, overall target's
Expression formula are as follows:
Wherein, QrWithRespectively ekT2Statistic and Q statistical magnitude, ΦrFor symmetrical positively definite matrix,WithRespectively ekT2
The control of statistic and Q statistical magnitude limits, and I is unit matrix, ΛrFor the pivot association of the static residual error portion e of real-time data collection
Variance matrix;Overall targetControl limit be written as:
Wherein,SrFor the covariance matrix of e;WhenWhen, then it is assumed that in static state
Failure has occurred in part;It is identical when breaking down with the dynamic part in step 2.1, when vehicle-mounted TCMS system acquires number from real-time
Detect that static part failure starts in, vehicle-mounted TCMS system storage temperature sensor saves high-speed rail by sample rate of T2
The bearing temperature data of failure operation, until failure terminates;
Step 3: in malfunction monitoring area, determining failure signature using multi-direction reconstructing method;After the failure occurred, using more
Direction reconstructing method determines failure variable, and adds failure while vehicle-mounted TCMS system saves in data to data buffer area
Signature;
Step 4: vehicle-mounted TCMS system sends data to ground system in real time;Data buffer storage is stored data into vehicle-mounted TCMS system
While in area, the data of storage are sent to high-speed rail train radio transmitting device, pass through GPRS, 3G/4G public network platform and iron
Road unifies MQ transmission platform and is sent to ground system in real time, passes through WLAN local network transport data to ground when high-speed rail train enters the station
Plane system.
4. a kind of method that the high-speed rail train data based on operating condition adaptively acquires according to claim 3, feature
Be, the step 3 the following steps are included:
Step 3.1: determining real-time data collection in the failure variable of dynamic part;For the dynamic part of real-time data collection, vk
The bearing temperature data sample x acquired in real time for the k momentkDynamic latent variable prediction error, because of vkIn principal component subspace
In, so needing fault moment f by vfIt is projected back in in original variable space, vfFor f moment sample xfDynamic latent variable
Predict error;The dynamic part of real-time data collection breaks down the original variable space x of moment fv,fExpression formula it is as follows:
xv,f=(w (pTw)-1)vf
Calculate the contribution degree Cont of the dynamic part Candidate Fault variable of real-time data collectionjIt is written as:
Dynamic part failure variable number can be determined by the size of contribution degree and determines failure signature, it is real in storage
When acquisition data procedures in failure signature is added in the data of storage;Due to the static part in real-time data collection
Also it can break down, so needing the failure variable of the static part of the determining real-time data collection of step 3.2;
Step 3.2: determining real-time data collection in the failure variable of static part;Real-time data collection is extracted dynamic latent variable
Static residual error portion afterwards is ef, calculate the contribution degree Cont of the static part Candidate Fault variable of real-time data collectioniExpression
Formula is as follows:
Static part failure variable number can be determined by contribution degree size and determines failure signature, it is real-time in storage
Failure signature is added in the data of storage in acquisition data procedures.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910070738.8A CN110111446A (en) | 2019-01-25 | 2019-01-25 | A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910070738.8A CN110111446A (en) | 2019-01-25 | 2019-01-25 | A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110111446A true CN110111446A (en) | 2019-08-09 |
Family
ID=67483693
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910070738.8A Pending CN110111446A (en) | 2019-01-25 | 2019-01-25 | A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110111446A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110782550A (en) * | 2019-09-20 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Data acquisition method, device and equipment |
CN110827446A (en) * | 2019-11-13 | 2020-02-21 | 北京理工大学 | Method for predicting running state of electric automobile |
CN111539374A (en) * | 2020-05-07 | 2020-08-14 | 上海工程技术大学 | Rail train bearing fault diagnosis system and method based on multidimensional data space |
CN112801320A (en) * | 2021-02-05 | 2021-05-14 | 苏州捷杰传感技术有限公司 | Data acquisition system, monitoring system and data acquisition method for rail train bearing |
CN114170705A (en) * | 2021-11-17 | 2022-03-11 | 华人运通(江苏)技术有限公司 | Vehicle data uploading method, device and equipment |
CN114333103A (en) * | 2021-12-24 | 2022-04-12 | 中国铁道科学研究院集团有限公司 | Locomotive-mounted data vehicle transfer and storage system and method |
CN114333103B (en) * | 2021-12-24 | 2024-04-19 | 中国铁道科学研究院集团有限公司 | Locomotive-mounted data train-ground dumping system and method |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103762688A (en) * | 2014-01-24 | 2014-04-30 | 湖南大学 | Grid frequency dynamic response device of electric car charging equipment |
CN104331947A (en) * | 2014-10-14 | 2015-02-04 | 深圳创维数字技术有限公司 | Vehicle monitoring method and related equipment and system |
CN105635241A (en) * | 2014-11-24 | 2016-06-01 | 现代自动车株式会社 | Method, system and computer-readable recording medium for managing abnormal state of vehicle |
CN105700517A (en) * | 2016-03-09 | 2016-06-22 | 中国石油大学(北京) | Adaptive data-driven early fault monitoring method and device during refining process |
CN106200518A (en) * | 2015-04-29 | 2016-12-07 | 中国科学院电工研究所 | A kind of frequency self-adaption method of electric-vehicle remote monitoring system |
US20160358722A1 (en) * | 2015-02-05 | 2016-12-08 | Ramasamy Lakshmanan | Intelligent wireless and wired control of devices |
CN106535253A (en) * | 2016-11-23 | 2017-03-22 | 北京必创科技股份有限公司 | Method for dynamic acquisition and transmission of wireless data |
CN107199914A (en) * | 2017-07-21 | 2017-09-26 | 广东电网有限责任公司信息中心 | A kind of charging device of electric automobile and method |
CN109141945A (en) * | 2018-08-16 | 2019-01-04 | 东北大学 | A kind of train bearing method for diagnosing faults based on multi-direction reconstruct |
-
2019
- 2019-01-25 CN CN201910070738.8A patent/CN110111446A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103762688A (en) * | 2014-01-24 | 2014-04-30 | 湖南大学 | Grid frequency dynamic response device of electric car charging equipment |
CN104331947A (en) * | 2014-10-14 | 2015-02-04 | 深圳创维数字技术有限公司 | Vehicle monitoring method and related equipment and system |
CN105635241A (en) * | 2014-11-24 | 2016-06-01 | 现代自动车株式会社 | Method, system and computer-readable recording medium for managing abnormal state of vehicle |
US20160358722A1 (en) * | 2015-02-05 | 2016-12-08 | Ramasamy Lakshmanan | Intelligent wireless and wired control of devices |
CN106200518A (en) * | 2015-04-29 | 2016-12-07 | 中国科学院电工研究所 | A kind of frequency self-adaption method of electric-vehicle remote monitoring system |
CN105700517A (en) * | 2016-03-09 | 2016-06-22 | 中国石油大学(北京) | Adaptive data-driven early fault monitoring method and device during refining process |
CN106535253A (en) * | 2016-11-23 | 2017-03-22 | 北京必创科技股份有限公司 | Method for dynamic acquisition and transmission of wireless data |
CN107199914A (en) * | 2017-07-21 | 2017-09-26 | 广东电网有限责任公司信息中心 | A kind of charging device of electric automobile and method |
CN109141945A (en) * | 2018-08-16 | 2019-01-04 | 东北大学 | A kind of train bearing method for diagnosing faults based on multi-direction reconstruct |
Non-Patent Citations (1)
Title |
---|
张斌,朱建涛,徐曌: "基于动态频率算法的远程监控系统数据采集优化策略", 《微电子学与计算机》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110782550A (en) * | 2019-09-20 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Data acquisition method, device and equipment |
CN110827446A (en) * | 2019-11-13 | 2020-02-21 | 北京理工大学 | Method for predicting running state of electric automobile |
CN111539374A (en) * | 2020-05-07 | 2020-08-14 | 上海工程技术大学 | Rail train bearing fault diagnosis system and method based on multidimensional data space |
CN111539374B (en) * | 2020-05-07 | 2022-05-20 | 上海工程技术大学 | Rail train bearing fault diagnosis method based on multidimensional data space |
CN112801320A (en) * | 2021-02-05 | 2021-05-14 | 苏州捷杰传感技术有限公司 | Data acquisition system, monitoring system and data acquisition method for rail train bearing |
CN112801320B (en) * | 2021-02-05 | 2024-04-02 | 苏州捷杰传感技术有限公司 | Data acquisition system, monitoring system and data acquisition method for rail train bearing |
CN114170705A (en) * | 2021-11-17 | 2022-03-11 | 华人运通(江苏)技术有限公司 | Vehicle data uploading method, device and equipment |
CN114333103A (en) * | 2021-12-24 | 2022-04-12 | 中国铁道科学研究院集团有限公司 | Locomotive-mounted data vehicle transfer and storage system and method |
CN114333103B (en) * | 2021-12-24 | 2024-04-19 | 中国铁道科学研究院集团有限公司 | Locomotive-mounted data train-ground dumping system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110111446A (en) | A kind of adaptive acquisition system of high-speed rail train data and method based on operating condition | |
CN103258427B (en) | Urban expressway traffic real-time monitoring system and method based on information physical network | |
CN100538761C (en) | Built-in intelligent fault diagnosing device and method based on the data fusion pattern-recognition | |
CN104091070B (en) | Rail transit fault diagnosis method and system based on time series analysis | |
CN109581871B (en) | Industrial control system intrusion detection method of immune countermeasure sample | |
CN102139700B (en) | Vehicle working condition online monitoring system for rail transit | |
CN110059631A (en) | The contactless monitoring defect identification method of contact net | |
CN110135273A (en) | Contact net video image cloud intellectual monitoring and fault recognition method | |
JP2018084854A (en) | Sensor data processing method | |
CN113465920B (en) | Cloud, fog and edge end cooperative bearing state monitoring and management method and system | |
KR102168218B1 (en) | Apparatus and method for predicting train fault | |
CN103493019A (en) | Collaborative multi-agent vehicle fault diagnostic system & associated methodology | |
CN102176050B (en) | Earthquake monitoring and pre-warning device and working method thereof | |
CN108683642B (en) | The detector and detection method of smart grid line status wrong data injection attacks | |
CN101958046A (en) | Vehicle track recognition system and method | |
CN108398934A (en) | The system that a kind of equipment fault for rail traffic monitors | |
JP2021501079A (en) | Data fusion concept | |
CN111717753A (en) | Self-adaptive elevator fault early warning system and method based on multi-dimensional fault characteristics | |
CN112162545A (en) | Automobile fault diagnosis method and system | |
CN106502248A (en) | A kind of method and device for starting intelligent vehicle | |
CN107991552A (en) | A kind of Reusable Launch Vehicles life appraisal and forecasting system and method | |
CN113093797A (en) | Unmanned aerial vehicle-based intelligent power transmission line inspection method and system | |
Chellaswamy et al. | An IoT based rail track condition monitoring and derailment prevention system | |
CN104504713B (en) | A kind of EMUs running status picture control failure automatic identifying method | |
CN113850138B (en) | Unmanned aerial vehicle sensor fault detection method, system and device based on data driving |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190809 |
|
RJ01 | Rejection of invention patent application after publication |