CN109689470A - Apparatus for diagnosis of abnormality, abnormality diagnostic method and computer program - Google Patents
Apparatus for diagnosis of abnormality, abnormality diagnostic method and computer program Download PDFInfo
- Publication number
- CN109689470A CN109689470A CN201880003331.7A CN201880003331A CN109689470A CN 109689470 A CN109689470 A CN 109689470A CN 201880003331 A CN201880003331 A CN 201880003331A CN 109689470 A CN109689470 A CN 109689470A
- Authority
- CN
- China
- Prior art keywords
- brake
- abnormality detection
- abnormality
- vehicle
- diagnosis
- 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.)
- Granted
Links
- 230000005856 abnormality Effects 0.000 title claims abstract description 227
- 238000003745 diagnosis Methods 0.000 title claims abstract description 98
- 238000002405 diagnostic procedure Methods 0.000 title claims description 7
- 238000004590 computer program Methods 0.000 title claims description 6
- 238000001514 detection method Methods 0.000 claims abstract description 174
- 230000002159 abnormal effect Effects 0.000 claims description 43
- 230000002547 anomalous effect Effects 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 description 24
- 238000000034 method Methods 0.000 description 23
- 238000012423 maintenance Methods 0.000 description 17
- 238000012360 testing method Methods 0.000 description 14
- 238000009826 distribution Methods 0.000 description 11
- 238000009434 installation Methods 0.000 description 10
- 230000004044 response Effects 0.000 description 10
- 238000005096 rolling process Methods 0.000 description 9
- 230000006866 deterioration Effects 0.000 description 8
- 235000013399 edible fruits Nutrition 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 6
- 230000015654 memory Effects 0.000 description 6
- 238000005259 measurement Methods 0.000 description 5
- 238000013016 damping Methods 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000008929 regeneration Effects 0.000 description 3
- 238000011069 regeneration method Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- RTZKZFJDLAIYFH-UHFFFAOYSA-N Diethyl ether Chemical compound CCOCC RTZKZFJDLAIYFH-UHFFFAOYSA-N 0.000 description 2
- 230000003247 decreasing effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005401 electroluminescence Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000001502 supplementing effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B61—RAILWAYS
- B61L—GUIDING RAILWAY TRAFFIC; ENSURING THE SAFETY OF RAILWAY TRAFFIC
- B61L15/00—Indicators provided on the vehicle or train for signalling purposes
- B61L15/0081—On-board diagnosis or maintenance
-
- 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
- B61L27/50—Trackside diagnosis or maintenance, e.g. software upgrades
- B61L27/57—Trackside diagnosis or maintenance, e.g. software upgrades for vehicles or trains, e.g. trackside supervision of train conditions
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Mechanical Engineering (AREA)
- Valves And Accessory Devices For Braking Systems (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
- Regulating Braking Force (AREA)
Abstract
The embodiment of the present invention realizes high accuracy diagnosis.According to one embodiment, apparatus for diagnosis of abnormality includes anomaly detector and diagnostor.Anomaly detector executes the abnormality detection of the decelerability of vehicle based on the prediction model of the control command value for brake apparatus and the deceleration for vehicle.Anomaly detector executes the abnormality detection of brake apparatus based on the prediction model of control command value and the brake force of brake apparatus.Diagnostor diagnoses vehicle based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus.
Description
Technical field
The embodiment of the present invention is related to apparatus for diagnosis of abnormality, abnormality diagnostic method and computer program.
Background technique
The maintenance management and inspection of rolling stock need to carry out daily, to maintain the safety and reliability of railway operation.
For example, the brake force that brake apparatus obtains reduces, prevent vehicle is from the phase when the brake apparatus of rolling stock breaks down
The position of prestige stops, this, which may cause convenience, reduces or cause in the worst case accident.Therefore, vehicle maintenance management
It is particularly important for railway operator.
Traditional maintenance management lays particular emphasis on always inspecting periodically for rolling stock.But recently, it has developed for leading to
It crosses and collects and realize using the information of vehicles of the sensor and controlling value that such as obtain from each rolling stock brake exception
Early detection technology.
But for rolling stock, due to such as the railroad track gradient, Changes in weather, passenger getting on/off and drive
The operation for the person of sailing, driving conditions dynamically change in a time-sequential manner.Accordingly, it is difficult to accurately be diagnosed to this vehicle.
The embodiment provides apparatus for diagnosis of abnormality, abnormality diagnostic method and the calculating of realizing pin-point accuracy diagnosis
Machine program.
Summary of the invention
According to one embodiment, apparatus for diagnosis of abnormality includes anomaly detector and diagnostor.
Prediction model of the anomaly detector based on the control command value for brake apparatus and the deceleration for vehicle come
Execute the abnormality detection of the decelerability of vehicle.The prediction of brake force of the anomaly detector based on control command value and brake apparatus
Model executes the abnormality detection of brake apparatus.
Diagnostor diagnoses vehicle based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus.
Detailed description of the invention
Fig. 1 is the block diagram of the abnormity diagnostic system of embodiment according to the present invention.
Fig. 2 is the figure of the representative configuration of the breaker slot for showing rolling stock, brake and air spring.
Fig. 3 is the figure for showing the representative configuration of generation braker and regeneration brake of rolling stock.
Fig. 4 is the figure for showing exemplary table relevant to metrical information and environmental information.
Fig. 5 is the figure for showing exemplary table relevant to metrical information.
Fig. 6 is the figure for showing illustrative translation table.
Fig. 7 is the figure for showing exemplary model database.
Fig. 8 is the figure for showing the illustrative methods by using normal distribution threshold value.
Fig. 9 is the figure for showing the exemplary operation of anomaly detector.
Figure 10 is the figure for showing exemplary detection result database.
Figure 11 is the figure for showing exemplary diagnostics rule database.
Figure 12 is the figure for showing the exemplary display screen of diagnosis output information.
Figure 13 is the figure for showing another exemplary diagnostics rule database.
Figure 14 is the figure for showing another exemplary diagnostics rule database.
Figure 15 is the figure for showing the hardware configuration of apparatus for diagnosis of abnormality of the present embodiment according to the present invention.
Figure 16 is the flow chart of the diagnostic process of embodiment according to the present invention.
Specific embodiment
The embodiment of the present invention is described below with reference to the accompanying drawings.Any same parts between attached drawing are by identical attached drawing mark
Note indicates, and will suitably the descriptions thereof are omitted.
Fig. 1 is the block diagram for showing the exemplary exceptions diagnostic system of embodiment according to the present invention.
Abnormity diagnostic system shown in FIG. 1 include apparatus for diagnosis of abnormality 100, Vehicular system 200, environmental information system 300,
Terminal 400, input unit 500 and screen display device 600.The summary of abnormity diagnostic system will be described below.
Apparatus for diagnosis of abnormality 100 has mode of learning and operation mode.In mode of learning, model generator 140 is based on
In the metrical information of the vehicle obtained from Vehicular system 200 and the environmental information of the vehicle from the acquisition of environmental information system 300
At least one abnormality detection model to generate the decelerability of rolling stock (hereinafter, vehicle).Measurement including vehicle
The information of at least one of information and the environmental information of vehicle is referred to as driving information.Vehicle, which can be included in, to be connected
The vehicle Che Zuzhong of the multiple vehicles connect.
Model generator 140 also generates the abnormality detection model of the brake apparatus of vehicle based on driving information.The present embodiment
Assuming that aor brake generates abnormality detection model as brake apparatus, and for aor brake.The vehicle vehicle group the case where
Under, it can be the abnormality detection model for being provided with each vehicle generation brake apparatus of brake apparatus.Common application can be generated
In the abnormality detection model of multiple brake apparatus, rather than abnormality detection model is generated for each individual brake apparatus.
Model generator 140 also generates the abnormality detection model for braking system (hereinafter referred to vehicle group brake).
Vehicle group brake includes the brake apparatus being arranged in multiple vehicles, and the abnormality detection model for vehicle group brake is to use
In the abnormality detection model of all brake apparatus.
It is stored in model database 102 by the various abnormality detection models that model generator 140 generates.
In operation mode, anomaly detector 150 executes decelerability by using decelerability abnormality detection model
Abnormality detection.Anomaly detector 150 also executes each air by using for the abnormality detection model of aor brake
The abnormality detection of brake.Anomaly detector 150 also executes vehicle by using the abnormality detection model for vehicle group brake
The abnormality detection of group brake.Abnormality detection determines whether there is exception.Abnormality detection is also referred to as abnormal determining.Decelerability,
The abnormality detection result of aor brake and vehicle group brake is stored in testing result database 103.
Abnormality detection result, the abnormality detection result of aor brake and vehicle group system of the diagnostor 160 based on decelerability
The abnormality detection result of device is moved to execute vehicle diagnostics.For example, aor brake is abnormal, and vehicle group when decelerability is normal
When brake exception, diagnostor 160 is diagnosed to be the sign that all aor brakes have deterioration.When decelerability exception, vehicle group
Brake is normal, and when one of aor brake exception, diagnostor 160 be diagnosed to be decelerability have be attributable to one
The exception of a brake apparatus deterioration.This diagnosis uses the various diagnostic rules being stored in diagnostic roles database 104.
Diagnostor 160 generates diagnosis output information according to diagnostic result, and generation is shown on screen display device 600
Diagnose output information.The supervision of display support such as railroad maintenance personnel or driver.
Mode of learning and operation mode can automatically switch perhaps by maintenance personnel etc. by instruction switching or can be simultaneously
Row executes.
Aor brake, its peripheral configuration and the breaker slot of vehicle according to the present embodiment are described below.Fig. 2 shows systems
The representative configuration of the aor brake and air spring of dynamic slot and the specific wheel for vehicle.Breaker slot is actually upper
In in the driver's cabin of vehicle group.
Brake bar 10 as example controller provides the equipment for operator brake operation.Driver passes through upward
Mobile brake bar carrys out abrupt deceleration vehicle.The digital representation breaker slot (retrostage) of 1 to 8 indicated on brake bar 10, and number is got over
Mean that the brake force for being applied to vehicle is stronger greatly.The slot number is exemplary and uses more or more without limitation on vehicle
The slot of small number.Each breaker slot is the exemplary control command value for vehicle or aor brake.
The operation executed by driver is not limited to the brake operating of vehicle.For example, stopping being equipped with automatic vehicle group
(ATS), it in the vehicle of automatic vehicle group control (ATC) or automatic vehicle group operation (ATO), in some cases, replaces driving by device
The person of sailing executes brake operating.In this case, for example, the brake command exported from device is corresponding with control command value.
Fig. 2 shows the wheels 30 of the vehicle travelled on railway 20.For being set by one kind of brake vehicle
Standby is the tyre surface brake 42 as a kind of aor brake.In this example, to simplify the description, a vehicle is illustrated only
Wheel, but in fact, provide multipair left and right wheels.For example, including an aor brake in each vehicle.But Yi Xieche
It may not include aor brake.
Tyre surface brake 42 uses cylinder as power source.When the brake-cylinder pressure increase as the pressure in cylinder 43
When, brake block 41 is pressed against on the tyre surface with the surface of rail contact as wheel 30.Between wheel 30 and brake block 41
Frictional force is used as the brake force of tyre surface brake 42.
Since tyre surface brake uses the frictional force of brake block in this way, brake block is ground due to continuous use
Damage, this may be decreased brake force.Tyre surface brake is the exemplary mechanical brake for vehicle, and another scheme uses
Disk brake, the disk brake by with such as brake block (pad) against wheel be push-fitted into the disk on axletree come
Obtain brake force.The brake force of brake changes with the state of wear of such as brake block or brake block.When pass through this exception
When diagnostic device detects the exception in aor brake, the brake block or disk that worker etc. can check such as aor brake are simultaneously
Check abnormal physical presence.
Other than the abrasion of the component of aor brake, the brake force of aor brake also as the load on vehicle and
Variation.Load response device 50 is mounted on vehicle shown in Fig. 2.Load response device 50 includes air spring 51, and is passed through
The air spring pressure of air spring 51 is detected to measure the load on vehicle.In order to control the braking of vehicle, in addition to brake bar
Except 10 operation, the air spring pressure that may also respond to be detected by load response device 50 is come the system of regulating brake
Power.Therefore, no matter how the load on vehicle changes, and can realize desired deceleration.
In order to supplement the brake force of mechanical brake, electric brake can be additionally used.Alternatively, mechanical brake can
With the brake force for supplementing electric brake.Electric brake will be described with reference to Fig. 3.Fig. 3 show vehicle generation braker and
The representative configuration of regeneration brake.The vehicle is a vehicle of Che Zuzhong, and before and after the vehicle with it is other
Vehicle connection.
Main motor 60a and 60b are mounted on vehicle shown in Fig. 3.When using generation braker, main motor 60a
Closed circuit is formed with 60b and resistor 70, by the electrical power conversion of main motor at thermal energy.
When using regeneration brake, the electric power generated by main motor 60a and 60b is transferred to route by bow collector 80
90.When secondary cell is installed on vehicle, generated electric power can be used for charging to secondary cell.In this way, then
Raw brake converts kinetic energy into electric power as generator to obtain brake force by using main motor 60a and 60b.
Mechanical brake and generation braker are exemplary, and apparatus for diagnosis of abnormality 100 can be to used
What the brake of its scheme executes abnormality detection and diagnosis.
Since there is brake the characteristic of relative complex construction and brake to change with Multiple factors and condition, because
This is difficult to execute accurate abnormality detection to the brake of vehicle.
For example, as described above, in the car using the brake of multiple schemes with different characteristics.In addition, institute as above
It states, the brake force of vehicle brake changes with load.For example, the quantity of passenger is with time slot and operation room for passenger vehicle
Every and change very big, therefore, the brake force of brake changes very big in short duration.For goods stock, load with goods
Object amount and change very big.In addition, the deceleration when executing brake operating may be in the vehicle with different gradient and gradient trend
Travel route and interval between change.In addition, any difference of the weather condition of rainfall, atmospheric temperature etc. may all change
The physical property of brake components, to influence the characteristic of brake.As other factors, driver executes system in different ways
Dynamic operation, and vehicle is manufactured with different braking characteristics.
In the present embodiment, for the abnormality detection model of decelerability (the abnormality detection model for braking ability) and
Abnormality detection model for vehicle group brake is used together with the abnormality detection model for aor brake to reduce mistake
The risk of diagnosis, to realize abnormal in early stage detection and safe and reliable railway operation.
Apparatus for diagnosis of abnormality 100 shown in Fig. 1 are described more fully.In the following description, it is assumed that abnormal
Diagnosing target vehicle is rolling stock, but not limited to this.Abnormity diagnosis target vehicle can be the optional vehicle including wheel, all
Such as automobile, building machinery or aircraft.
Apparatus for diagnosis of abnormality 100 shown in FIG. 1 includes information of vehicles collector 110, environmental information collector 120, data
Processor 130, model generator 140, anomaly detector 150, diagnostor 160 and alarm device 170.
Information of vehicles collector 110 obtains relevant to vehicle from the various sensors of the Vehicular system 200 of vehicle interior
Metrical information (also referred to as measurement data).The example of sensor includes: brake operating for being configured as sensing vehicle etc. as control
The sensor of bid value processed；It is configured as the sensor of the deceleration of detection vehicle；It is configured as the sensing of detection travel speed
Device；And it is configured as the sensor that measurement is applied to the load of vehicle.It may include various other sensors.Metrical information packet
The sensing value (for example, control command value) of sensor and the measured value of sensor are included (for example, drive speed, load or deceleration
Degree).When Vehicular system 200 calculates deceleration according to the value of velocity sensor, available calculated deceleration, which is used as, to be subtracted
The measured value of speed.
The type (type of sensor or the type of control command value) for the metrical information to be obtained can be optionally set.
Metrical information can be obtained in the period being optionally arranged.For example, metrical information relevant to the travel speed of vehicle with
Millisecond is to obtain in the short sampling period of unit.It is configured as measuring the value of the sensor for the load for being applied to vehicle with minute
To be obtained in the sampling period of unit.
Environmental information collector 120 obtains the environmental information of vehicle from environmental information system 300.The example packet of environmental information
It includes and the operation relevant information of route and information relevant with weather.Between the example of information relevant to operation route includes each
Every the gradient and gradient (difference in height between the interior outside track of railway).The example of information relevant to weather includes weather, big
Temperature degree, precipitation, wind speed and atmospheric pressure.The acquisition of environmental information, which can be in the database obtained in ground system, tires out
Long-pending information, or obtain the information distributed from external server.The type for the environmental information to be obtained can be optionally set
With the frequency of acquisition.
Apparatus for diagnosis of abnormality 100 can be installed as the ground installation of outside vehicle, for example, it is public to be mounted on railway operation management
The facility of department or operational order center, or can be used as car-mounted device and be installed on vehicle.Apparatus for diagnosis of abnormality 100 is not limited to
Specific mounting means.
When apparatus for diagnosis of abnormality 100 is installed as the ground installation of outside vehicle, the Vehicular system 200 of vehicle interior
Metrical information etc. is received for example, by vehicle-mounted element, transponder ground component and terrestrial information network.Specifically, Vehicular system
200 transmit data to terrestrial information network by ground component etc., and apparatus for diagnosis of abnormality 100 passes through terrestrial information network
Receive data.Such as wire rope, coaxial cable, optical cable, telephone wire, wireless device or ether can be used in terrestrial information network
Net (registered trademark), but it is not limited to specified scheme.Apparatus for diagnosis of abnormality 100 passes through terrestrial information network from environmental information system
300 receive data.
When apparatus for diagnosis of abnormality 100 is car-mounted device, apparatus for diagnosis of abnormality 100 by the information network of vehicle interior from
Vehicular system 200 obtains data.The information network of vehicle interior is, for example, Ethernet or WLAN (LAN), but can be with
With the realization of any other scheme.Vehicle-mounted element can be used in apparatus for diagnosis of abnormality 100 and transponder ground component is believed from ground
The environmental information system 300 of breath network connection obtains data.
Input unit 500 provides the interface of the operation for maintenance personnel.Input unit 500 includes mouse, keyboard, voice
Identifying system, image identification system, touch tablet or combinations thereof.Maintenance personnel can be filled by input unit 500 to abnormity diagnosis
The various order or data of 100 inputs are set to execute operation.
Screen display device 600 shows the data exported from apparatus for diagnosis of abnormality 100 or information as static image or shifting
Motion video.Screen display device 600 is, for example, liquid crystal display (LCD), display of organic electroluminescence or is vacuum fluorescence
Display (VFD) is but it is also possible to be the display device in any other scheme.
Each of input unit 500 and screen display device 600 can be one in the device of multiple installations.Example
Such as, input unit 500 and screen display device 600 may be mounted at each of the station of operational order center and vehicle
Place.
Input unit 500 and screen display device 600 can be an integrating device.For example, there is touch tablet when using
When the display of function, single device may be used as input unit 500 and screen display device 600.
Apparatus for diagnosis of abnormality 100 includes information database 101, model database 102, testing result database 103 and examines
The database of disconnected rule database 104.
Database 101,102,103 and 104 is all disposed within the inside of apparatus for diagnosis of abnormality 100 shown in FIG. 1.But number
Ad hoc approach is not limited to according to the arrangement in library.For example, partial database can be externally arranged in server or storage device.Each
Database can be realized by relational database management system and various NoSQL systems, but can be realized in any other scheme.
Each database can using XML, JSON or CSV storage format or any other format, such as binary format.Not
All databases in apparatus for diagnosis of abnormality 100 are required through identical Database Systems and identical storage format come real
It is existing, but database can be realized with the mixing of kinds of schemes.
Information database 101 stores the metrical information obtained by information of vehicles collector 110 and by environmental information collector
120 environmental informations obtained.Driving information according to the present embodiment includes at least one of metrical information and environmental information.It is all
The storage medium for such as storing the memory device of driving information is inserted into apparatus for diagnosis of abnormality 100 and is used as Information Number
According to library 101.
Fig. 4 and Fig. 5 shows exemplary information database 101.For example, with table 101a shown in Fig. 4 and table shown in fig. 5
The form of 101b stores driving information (metrical information and environmental information).In this illustration, driving information is according to sample frequency
It is stored in two individual tables.
The generation time of the Time Column storage entry of table 101a shown in Fig. 4.In this example, it constant is adopted each
The sample time generates entry.However, it is possible to preset the interval of railroad track or generate entry with any other standard.
The value of " vehicle group/vehicle " column storage identification the vehicle group and vehicle of table 101a.Present embodiment assumes that wherein in each vehicle
A case where brake apparatus is arranged in.But some vehicles may not include brake apparatus.Multiple brake apparatus can be by
It is included in a vehicle.
" weather " column of table 101a store information relevant to the weather obtained from environmental information system 300.
" atmospheric temperature " column of table 101a store information relevant to the atmospheric temperature obtained from environmental information system 300.
The label that information relevant to atmospheric temperature can be actual value or classify to actual value.In the example depicted in fig. 4,
The classification that the storage of " atmospheric temperature " column is converted into as the atmospheric temperature of real number by using conversion table 101c as shown in FIG. 6
The label of any one of T1, T2, T3, T4, T5, T6 and T7.For example, -11 DEG C of atmospheric temperature is converted into classification T1,15
DEG C atmospheric temperature be converted into classification T4, and 33 DEG C of atmospheric temperature is converted into classification T6.
Column in addition to " atmospheric temperature " column, which also can store, to be calculated or turns by executing to metrical information or environmental information
The processing information changed and obtained.
" seating rate " column of table 101a store seating rate as the index for the load for being applied to vehicle using percents.Separately
One index can serve to indicate that load.Seating rate is defined by the ratio of passengers quantity in such as vehicle capacity and vehicle.It is logical
Seating rate is often estimated based on the air spring pressure of load response device.In this case, air spring pressure can be straight
It connects and is used as index.
Air spring pressure is the actual value of sensor, different from seating rate, be not by using such as conversion table or
The value that formula is estimated indirectly, and therefore can be used for reducing the error in model generation.But the value of air spring pressure takes
Certainly in the model of manufacturer and the load response device being installed on vehicle, and therefore lack versatility.Therefore, when using logical
When the index for such as seating rate being often used, in some cases it may absorb since its different load response device causes
Vehicle between difference.
" gradient " column of table 101a are to be that the value that unit indicates stores the gradient of route with permillage (permil).Permillage
It (permil) is the value obtained and the difference in height with the horizontal distance for indicating 1000 meters.Permillage (permil) is exemplary
, " gradient " arranges the value that can store another unit.
The gradient of " gradient " column storage of table 101a in millimeters, but can store the value of another unit.
Table 101a further includes " wind speed " column and " atmospheric pressure " column.Table 101a may include that storage such as railroad track is taken in
The column of the other information of front position and current interval.
It is set in table 101b storage time information shown in fig. 5, the actual value of the deceleration of breaker slot, vehicle and vehicle
The aor brake 1 set to N air brake pressure actual value.Entry in table 101b is short with the time interval than table 101a
Time interval generate.The generation interval (sampling interval) of entry in table 101b can be identical with table 101a.Table 101b and
Table 101a can be integrated with each other.Deceleration can be the value of acceleration transducer, or can be according to velocity sensor
Value calculate.
It can handle the data being stored in information database 101.For example, data processor 130 makes screen display device
The content for each table that 600 displays are stored in information database 101.Maintenance personnel or driver are by using input unit
500 pairs of data execute processing operation.Data processor 130 executes data processing according to processing operation.
Adjustable information of vehicles collector 110 or environmental information collector 120 obtain the interval of information or data.Example
Such as, data processor 130 is received by input unit 500 from maintenance personnel or the specified operation for obtaining interval of driver, and root
Interval is obtained according to the content adjustment of operation.
Model generator 140 is generated by using the data being stored in information database 101 for the different of decelerability
Normal detection model, the abnormality detection mould for the abnormality detection model of brake apparatus (aor brake) and for vehicle group brake
Type.
Abnormality detection model for decelerability includes deceleration prediction model (hereinafter referred to deceleration model), with
And threshold value (hereinafter referred to deceleration threshold) relevant to the residual error of the predicted value from deceleration model.Deceleration threshold is used
In anomaly detector 150 to the abnormality detection of the deceleration of vehicle.Specifically, deceleration threshold be used for as deceleration mould
The predicted value of type is compared with the residual error of the difference between the actual value of deceleration.
Abnormality detection model for brake apparatus includes brake force prediction model, and with the prediction from prediction model
The relevant threshold value of the residual error of value (hereinafter referred to individual braking threshold).The brake force and air brake pressure pair of aor brake
It answers.In this example, it is assumed that brake force prediction model is air brake pressure prediction model (hereinafter referred to air damping pressure
Power model).Individual braking threshold is for anomaly detector 150 to the abnormality detection of each brake apparatus.Specifically, a system
Dynamic threshold value is for the residual of the difference between the actual value of predicted value and air brake pressure as air brake pressure model
The comparison of difference.
Abnormality detection model for vehicle group brake includes the prediction mould of the value of the brake force based on multiple brake apparatus
Type (hereinafter referred to vehicle group brake model), and threshold value relevant to the residual error of the predicted value from prediction model is (hereinafter
Referred to as vehicle group braking threshold).In this example, it is assumed that brake apparatus is aor brake, and assume to be used for air damping
The prediction model of pressure summation.Brake force summation can be replaced with statistical value, such as average value or intermediate value, or be based on multiple systems
Any other value of the brake force of dynamic device.Specifically, vehicle group braking threshold is used for and the prediction as vehicle group brake model
Value is compared with the residual error of the difference between the actual value summation of air brake pressure.
It is stored in model database 102 by these abnormality detection models that model generator 140 generates.
Fig. 7 shows the example of model database 102.Each abnormality detection model is identified by model ID.Prediction model column
Store the address (pointer) of the data of indication predicting model or the memory of Storage Estimation model.Prediction is arrived in threshold column storage setting
The threshold value of model.
The generation of abnormality detection model executes in mode of learning, for example, when starting apparatus for diagnosis of abnormality 100 or working as
When new addition vehicle is as diagnosis target.When there are multiple diagnosis targets, abnormality detection model is generated for each diagnosis target.
Model generator 140 can be either periodically or in response to the instruction abnormal replication detection model of such as maintenance personnel, and benefit
With the abnormality detection model that the abnormality detection model modification of duplication is previous.
The method described in detail below for generating abnormality detection model.By using the data extracted from information database 101
Sample (feature vector) generates abnormality detection model.
The example for wherein generating the abnormality detection model for being used for deceleration will be described below, but can be with similar
Mode generate the abnormality detection model for aor brake and the abnormality detection model for vehicle group brake.
Data sample (feature vector) includes one or more explanatory variables.The example of explanatory variable is the system in table 101b
The value (control command value) of dynamic slot.The driving information and specification (such as size and weight of vehicle) of vehicle can additionally be used
In any other value (such as speed) be used as explanatory variable.It can be by calculating the multiple project next life for including in driving information
At explanatory variable.In this example, the target variable of prediction model is deceleration.It can be each of braking information table 101b
Entry generates data sample, or can collect multiple continuous entries with lower time granularity to generate a data sample
This.
The method for generating prediction model (in this example, deceleration model) is described below.Assuming that wherein using recurrence mould
The case where type is as prediction model.Model generator 140 is obtained by using information database 101 has explanatory variable as member
Feature vector, X=(x of element1,x2,x3,...,xn)。
Then, model generator 140 executes multiple regression analysis to calculate prediction as the public affairs of the deceleration of target variable
Formula (1).
[expression formula 1]
Y=b0+b1x1+b2x2+b3x3+...+bnxn(1)
In the formula, y indicates target variable, xnIndicate explanatory variable, and bnIt indicates partial regression coefficient (parameter).It can
With by maximal possibility estimation etc. come calculating parameter.In order to absorb the difference of the measurement unit between explanatory variable, can pass through
Target variable and all explanatory variables are normalized to average value 0 and dispersion degree 1 uses standard partial regression coefficient as partial regression
Coefficient bn.The quantity of explanatory variable can be one or more.
Be exemplary by the generation of the model of multiple regression analysis, and can by such as support vector regression or from
Any other method of recurrence generates the prediction model of target variable.
When generating prediction model, cross validation can be used.For example, data sample can be divided into multiple set,
So that at least one set is used as the test data for verifying, and other set are generated for model.This, which allows to check, gives birth to
At model performance.
For example, cannot be obtained when control command value is used only as explanatory variable by using simple relation formula
Enough estimated accuracies.This is because the external factor of the transient response of such as deceleration or such as railroad track gradient etc.
Therefore, in order to establish more accurate prediction model, can by consider to slow down be in static in the transient response,
The influence of the switch mode of breaker slot and the railroad track gradient executes parameter Estimation.
Above description is the exemplary generation about the abnormality detection model for deceleration.Air damping is used for when generating
When the abnormality detection model of device, for example, control command value (breaker slot) may be used as explanatory variable, and air brake pressure can
For use as target variable.For example, control command value (breaker slot) can when generating the abnormality detection model for being used for vehicle group brake
Air brake pressure summation for use as explanatory variable, and multiple aor brakes may be used as target variable.It can add
Any explanatory variable in addition to control command value.
Assuming that the storage of information database 101 is obtained when brake apparatus (aor brake) is normal in mode of learning
Information.Therefore, normally assumed based on these aor brakes to be modeled in the various prediction models generated in mode of learning and (subtract
Rate pattern, air brake pressure model and vehicle group brake model).But prediction model is suitable for wherein information database 101
The case where storing the metrical information of some faulty aor brakes.
The method that the threshold value for determining setting to prediction model is described below.In the following description, prediction model means to subtract
Any one of rate pattern, air brake pressure model and vehicle group brake model.Similarly, threshold value means deceleration threshold
Value, individual any one of braking threshold and vehicle group braking threshold.
Threshold value is for determining the predicted value (for example, predicted value of deceleration) when the target variable calculated based on prediction model
When difference between the measured value (actual value) of deceleration is more than threshold value, there are exceptions.Abnormal determining also referred to as abnormality detection.
Difference between predicted value and actual value is referred to as residual error.Actual value can be more than or less than predicted value, therefore the value of residual error can
To be positive or negative.When symbol is inessential, residual error can be defined as absolute value of the difference, be the distance away from predicted value
Absolute value.
Fig. 8 is for describing to determine method using the example thresholds of normal distribution.Fig. 8 shows the normal distribution of residual error
401 curve graph.Horizontal axis indicates residual error, and the longitudinal axis indicates probability density.Assuming that these residual error Normal Distributions, obtain prediction mould
Multiple residual errors between the predicted value and actual value of type are to generate normal distribution 401.Data for obtaining residual error can be use
In the data sample of generation prediction model, test data, the driving information unrelated with the generation of prediction model or its optional group
It closes.When residual error has biggish variance, distribution has broader skirt section, as used 402 He of normal distribution shown in dotted line in figure
Shown in 403.
Normal distribution 401 is used to be arranged the threshold value of prediction model.For example, when standard deviation is indicated by σ, standard deviation
Multiple, such as 2 σ or 3 σ, are set to threshold value.When residual error is more than to be set to the threshold value of 2 σ, detected in abnormality detection
It is abnormal.When such threshold value is arranged, about 95% actual value is confirmed as without abnormal (normal).Property as another example
Threshold value setting, can will residual values corresponding with predetermined probability (for example, highest X percentage point or minimum X percentage point) or its absolutely
Value is set as threshold value.Above-mentioned Threshold is exemplary, and any other method can be used.For example, can lead to
It crosses and assumes that the distribution in addition to normal distribution carrys out threshold value, or its experience is based on by the people of such as maintenance personnel or driver
Carry out threshold value.
Any one in deceleration threshold, individual braking threshold and vehicle group braking threshold can be determined by the above method
It is a.
In operation mode, anomaly detector 150 is by using the various abnormality detections being stored in model database 102
Model and the driving information being stored in information database 101 execute abnormality detection.For example, can be by for shown in Fig. 5
Table 101b or table 101a in each entry generate feature vector come execute abnormality detection, can be by with constant time intervals
Selection entry simultaneously generates feature vector from selected entry to execute abnormality detection, or can be by out of constant
Entry generates feature vector to execute abnormality detection.Alternatively, the item from the time for example specified by maintenance personnel can be passed through
Mesh or the entry in the duration specified by maintenance personnel generate feature vector to execute abnormality detection.
Abnormality detection, the abnormality detection of each aor brake and the vehicle group of the execution decelerability of anomaly detector 150
The abnormality detection of brake.
In the abnormality detection of decelerability, the generation of the driving information according to used in abnormality detection feature vector (for example,
Control command value), and deceleration is predicted by using the feature vector of generation and deceleration model.Will prediction deceleration and
Residual error between actual deceleration degree (for example, obtaining from table 101b) is compared with deceleration threshold.When residual error is equal to or less than
When deceleration threshold, decelerability is confirmed as normally.When residual error is greater than deceleration threshold, decelerability is confirmed as different
Often.
In the abnormality detection of each aor brake, the driving information according to used in abnormality detection generates feature vector
(for example, control command value), and air damping is predicted by using the feature vector of generation and air brake pressure model
Pressure.It will predict the residual error between air brake pressure and actual air brake pressure (for example, obtaining from table 101b) and individual
Braking threshold is compared.When residual error is equal to or less than individual braking threshold, aor brake is confirmed as normally.Work as residual error
When greater than individual braking threshold, aor brake is confirmed as exception.
In the abnormality detection of vehicle group brake, the driving information according to used in abnormality detection generates feature vector (example
Such as, control command value), and multiple aor brakes are predicted by using the feature vector of generation and vehicle group brake model
Air brake pressure summation.By the residual error and vehicle between the air brake pressure summation of prediction and actual air brake pressure summation
Group braking threshold is compared.When residual error is equal to or less than vehicle group braking threshold, vehicle group brake is confirmed as normally.When residual
When difference is greater than vehicle group braking threshold, vehicle group brake is confirmed as exception.
The specific illustrative operation of the decelerability abnormality detection carried out by anomaly detector 150 is described below.
Fig. 9 is the figure for describing the exemplary operation of anomaly detector 150.The top of Fig. 9 shows breaker slot.Wherein
Between partially illustrate braking deceleration.Its underpart shows the residual error between actual value and the predicted value of prediction model.Slowing down
It spends in model, breaker slot is corresponding with explanatory variable, and deceleration is corresponding with target variable.
In time t1, the operation that breaker slot is set as to slot " 4 " is executed.After receiving the operation, each aor brake
Apply brake force to vehicle.Therefore, the deceleration of vehicle increases and hereafter keeps approximately constant value for a period of time.Deceleration
There is small difference between predicted value and its measured value (actual value), but substantially with identical value transition, and predict
Residual error between value and actual value is less than deceleration threshold.
Hereafter, residual error is more than deceleration threshold at three moment of time t2, t3 and t4, and anomaly detector 150 exists
Each moment detects exception.
In time t5, the operation that breaker slot is changed into slot " 2 " from slot " 4 " is executed.After receiving the operation, Mei Gekong
Pneumatic brake reduces the brake force for being applied to vehicle, and therefore, the deceleration of vehicle reduces.
In time t6, the operation for cancelling braking is executed.After receiving the operation, each aor brake is further decreased
It is applied to the brake force of vehicle, therefore, the deceleration of vehicle further decreases.
After above-mentioned time t4 detects exception, residual error does not detect different in the range of deceleration threshold
Often.
Although above description is the exemplary operation about decelerability abnormality detection, can hold in a similar way
The abnormality detection of the aor brake of each vehicle of row and the abnormality detection of vehicle group brake.
Abnormality detection result, the abnormality detection result of aor brake and vehicle of the anomaly detector 150 based on decelerability
The abnormality detection result of group brake stores information in testing result database 103.
Figure 10 shows exemplary detection result database 103.The testing result database 103 is specific in a time-sequential manner
Vehicle group store breaker slot, the abnormality detection result of decelerability, each vehicle aor brake abnormality detection result, and
The abnormality detection result of vehicle group brake.In this example, the abnormality detection knot in first to any one of Article 6 mesh
Fruit all indicates without abnormal, but detects exception at the aor brake 1 in the 7th and Article 8 mesh.
When detecting predetermined exception, alarm device 170 by abnormal notification message be sent to by railroad operator, driver or
The terminal 400 that maintenance personnel uses.Predetermined exception can optionally be defined as the exception for example detected in decelerability,
The exception detected in a certain number of aor brakes, or the exception detected in vehicle group brake.
Message informing can be by sending Email, showing Pop-up message, Huo Zhetong in the operation display of terminal 400
It crosses scheduled tool management agreement to be notified to execute, or can be executed by any other means.Notice includes different
Normal details (for example, the identifier in the place (current location) or the vehicle occurred extremely that are abnormal on map).Behaviour
Work person or maintenance personnel can understand abnormal detection and its details by receiving notice.
Diagnostor 160 is based on testing result database 103 and diagnostic roles database 104 executes vehicle diagnostics.
Diagnostic roles database 104 keeps diagnostic rule data, each aor brake of the diagnostic rule data definition (system
Dynamic device) abnormality detection result, every kind of the abnormality detection result of the abnormality detection result of decelerability and vehicle group brake
Combined diagnostic result.
Figure 11 shows exemplary diagnostics rule database 104.In this example, there are eight diagnostic rules, difference
With diagnostic rule number 1 to 8.Each circle in Figure 11 means abnormality detection result instruction without abnormal.In Figure 11
Each intersection means that abnormality detection result instruction is abnormal.
The definition of diagnostic rule 1 is when all results instruction of the abnormality detection of aor brake 1 to N is without abnormal, vehicle group system
The abnormality detection result instruction of dynamic device without abnormal, and the abnormality detection result instruction of decelerability without it is abnormal when diagnosis
As a result.Specifically, the diagnostic result of diagnostic rule 1 indicates normal condition.
The definition of diagnostic rule 2 is when all results instruction of the abnormality detection of aor brake 1 to N is without exception, vehicle group system
The abnormality detection result instruction of device is moved without abnormal, and the diagnosis knot when abnormality detection result instruction exception of decelerability
Fruit.Specifically, the exception of the diagnostic result instruction brake block of diagnostic rule 2, wheel or pavement state.
The definition of diagnostic rule 3 is when all results instruction of the abnormality detection of aor brake 1 to N is without abnormal, vehicle group system
The abnormality detection result of dynamic device indicate it is abnormal, and the abnormality detection result instruction of decelerability without it is abnormal when diagnosis knot
Fruit.Specifically, the sign that the diagnostic result instruction aor brake of diagnostic rule 3 deteriorates as a whole.
The definition of diagnostic rule 4 is when all results instruction of the abnormality detection of aor brake 1 to N is without abnormal, vehicle group system
The abnormality detection result of dynamic device indicates exception, and the diagnostic result when abnormality detection result instruction exception of decelerability.Tool
For body, decelerability is different caused by the diagnostic result instruction of diagnostic rule 4 is deteriorated as a whole due to aor brake
Often.
The definition of diagnostic rule 5 is when the instruction of at least one of aor brake 1 to the abnormality detection result of N is abnormal, vehicle group
The abnormality detection result of brake is indicated without abnormal, and the abnormality detection result of decelerability indicates examining when no exception
Disconnected result.Specifically, the diagnostic result instruction of diagnostic rule 5 has deterioration or its sign of abnormal aor brake.
The definition of diagnostic rule 6 is when the instruction of at least one of aor brake 1 to the abnormality detection result of N is abnormal, vehicle group
The abnormality detection result instruction of brake is without abnormal, and the diagnosis knot when abnormality detection result instruction exception of decelerability
Fruit.Specifically, the exception of the diagnostic result instruction decelerability of diagnostic rule 6, is attributable to abnormal air system
Dynamic device.
The definition of diagnostic rule 7 is when all results instruction of the abnormality detection of aor brake 1 to N is abnormal, vehicle group brake
Abnormality detection result indicate it is abnormal, and the abnormality detection result instruction of decelerability without it is abnormal when diagnostic result.Tool
For body, decelerability is abnormal caused by the diagnostic result instruction of diagnostic rule 7 is deteriorated as a whole due to aor brake
Sign.
The definition of diagnostic rule 8 is when all results instruction of the abnormality detection of aor brake 1 to N is abnormal, vehicle group brake
Abnormality detection result indicate abnormal, and the diagnostic result when abnormality detection result of decelerability indicates abnormal.It is specific and
Speech, the exception of decelerability caused by the diagnostic result instruction of diagnostic rule 8 is deteriorated as a whole due to aor brake.
The diagnostic rule in addition to diagnostic rule 1 to 8 can be defined.For example, the modification as diagnostic rule 7, diagnosis knot
Fruit can be defined as instruction when the quantity of the abnormal aor brake abnormality detection result of instruction is equal to or more than predetermined quantity
When, it is attributable to determine the sign of the decelerability exception of the deterioration with abnormal aor brake.As diagnostic rule 8
Modification, diagnostic result can be defined as instruction when the quantity of the abnormal aor brake abnormality detection result of instruction is equal to or greatly
When predetermined quantity, it is attributable to determine the exception of the decelerability of the deterioration with abnormal aor brake.It can define
Any other diagnostic rule.
Diagnostor 160 determines in each entries match diagnostic rule 1 to 8 being stored in testing result database 103
Which, and diagnosis output information is generated according to the diagnostic result indicated by matched diagnostic rule.It can use multiple diagnosis
Rule is matched.Diagnostor 160 makes screen display device 600 show diagnosis output information generated.When instruction is normally examined
When the diagnostic rule 1 of disconnected result matches, it will not generate and also not show diagnosis output information.
When detecting the exception of decelerability, diagnostor 160 can be according to the predicted value and deceleration of deceleration model
Measured value between difference calculate control information, and export calculated control information as diagnosis output information.Accidentally
In the example calculation of poor information, the braking distance when deceleration is in predicted value can be calculated and measured with when deceleration is in
The difference between braking distance when value.Braking distance is since braking starts to stop or until vehicle until reaching desired deceleration
Distance until degree or speed.Compared with normal condition, this helps to understand aor brake deterioration or the deterioration of vehicle group brake
To the influence degree of deceleration.Therefore, above-mentioned braking distance difference is anomalous effects degree information.
Figure 12 shows exemplary display screen (the diagnosis knot of the diagnosis output information shown by screen display device 600
Fruit screen).The screen shows the diagnostic result executed to vehicle group A.In this example, screen display device 600 is mounted on management
It is indoor with the order of monitoring vehicle group.
The topmost portion display diagnosis target carriage group of diagnostic result screen is vehicle group A.Second part shows multiple diagnosis rule
The number of matched diagnostic rule in then.Part III shows the diagnostic result indicated by matched diagnostic rule.In the example
In, diagnostic rule 6 matches, therefore shows that decelerability has the exception for being attributable to the deterioration of individual air brake.4th
Divide the information for showing the attributable aor brake of exception determined in specified decelerability.The aor brake can pass through
Such as specified abnormality detection result indicates the abnormal aor brake in testing result database to determine.Part V is shown
The difference of braking distance compared with normal condition is as anomalous effects degree.
Above-mentioned diagnosis output information is exemplary, and can be shown in various formats depending on the application.Diagnosis output letter
Breath can be sequentially recorded in the database individually prepared.In this case, selection is recorded and is executed from the database
Maintenance personnel of idsplay order operation etc. can show diagnosis output information on screen display device 600.
Above-mentioned diagnostic rule shown in Figure 11 is by using aor brake, deceleration and vehicle group brake abnormality detection
As a result these three are defined, but can be defined by using two kinds in these kind of result.For example, can be by using subtracting
Velocity anomaly testing result and aor brake abnormality detection result define diagnostic rule.
For example, aor brake abnormality detection result and deceleration abnormality detection result can be combined to define such as Figure 13 institute
The diagnostic rule shown.It alternatively, can be with combination vehicle group brake abnormality detection result and deceleration abnormality detection result to define
Diagnostic rule as shown in figure 14.
It can be different by combination aor brake, deceleration and vehicle group brake according to the apparatus for diagnosis of abnormality of the present embodiment
Two kinds in normal testing result these three or these kind of result execute the vehicle diagnostics of high accuracy.For example, can be based on examining
Rule 3, diagnostic rule 5 or diagnostic rule 7 break to detect the sign of aor brake exception.In addition, when being based on 2 nothing of diagnostic rule
When method obtains desired decelerability (braking ability), the external environment of vehicle can be appointed as factor.
Figure 15 shows the hardware configuration of the apparatus for diagnosis of abnormality according to the present embodiment.It is realized by computer installation 100
According to the apparatus for diagnosis of abnormality of the present embodiment.Computer installation 100 include CPU 151, input interface 152, display device 153,
Communication device 154, main storage means 155 and external memory 156.These components are connected with each other by bus 157.
Central processing unit (CPU) 151 executes abnormality diagnosis procedure as computer program in main storage means 155.
Abnormality diagnosis procedure is the computer program for realizing the above-mentioned functional component of each of apparatus for diagnosis of abnormality.It is executed by CPU 151
Abnormality diagnosis procedure realizes each functional component.
Input interface 152 is for inputting the input dress from such as keyboard, mouse or touch tablet to apparatus for diagnosis of abnormality
The circuit for the operation signal set.
Display device 153 shows the data or information exported from apparatus for diagnosis of abnormality.Display device 153 is such as liquid crystal
Show device (LCD), cathode-ray tube (CRT) or plasma display (PDP), but not limited to this.It is exported from computer installation 100
Data or information can be shown by display device 153.
Communication device 154 be allow apparatus for diagnosis of abnormality by wireless or cable with the circuit of communication with external apparatus.It can
To pass through communication device 154 from external device (ED) input measurement information.The metrical information inputted from external device (ED), which can store, to be believed
It ceases in database 101.
Data needed for main storage means 155 store such as abnormality diagnosis procedure, execute abnormality diagnosis procedure, and pass through
The data for executing abnormality diagnosis procedure and generating.Abnormality diagnosis procedure is loaded into main storage means 155 and is performed.It is main
Storage device 155 is such as RAM, DRAM or SRAM, but not limited to this.Information data can be constructed in main storage means 155
Library 101, model database 102, testing result database 103 and diagnostic roles database 104.
Data needed for external memory 156 stores such as abnormality diagnosis procedure, executes abnormality diagnosis procedure, Yi Jitong
Cross the data for executing abnormality diagnosis procedure and generating.When executing abnormality diagnosis procedure, these programs and data are read into master
On storage device 155.External memory 156 is such as hard disk, CD, flash memory or tape, but not limited to this.It can be in outside
Information database 101, model database 102, testing result database 103 and diagnostic rule data are constructed on storage device 156
Library 104.
Abnormality diagnosis procedure can be pre-installed on computer installation 100, or can store such as CD-ROM's
In storage medium.Alternatively, abnormality diagnosis procedure can be uploaded on the internet.
Computer installation 100 may include one or more CPU 151, one or more input interfaces 152, one or more
A display device 153, one or more communication device 154, and one or more main storage means 155, and can with it is all
As the peripheral equipment of printer or scanner etc connects.
Apparatus for diagnosis of abnormality can be realized by single computer installation 100, or can be configured as including being connected to each other
Multiple computer installations 100 system.
Figure 16 is the flow chart for the diagnostic process of embodiment according to the present invention executed in operation mode.Shown in Figure 16
Flow chart processing can diagnose target vehicle specific operation when execute, can be executed within the constant cycle, Ke Yi
It executes or can be executed in any other time when receiving the instruction of the user from such as maintenance personnel.In the example
In, it is assumed that diagnosis target vehicle is vehicle vehicle group.
In step S101, anomaly detector 150 obtains row relevant to diagnosis target vehicle vehicle group from information database 101
Sail information (with reference to Fig. 5 and 4).
In step S102, anomaly detector 150 obtains relevant different to diagnosis target vehicle vehicle group from model database 102
Normal detection model.Specifically, obtaining the abnormality detection model for deceleration, for installing air system on each vehicle
The abnormality detection model of dynamic device, and the abnormality detection model for vehicle group brake.It can be provided for each aor brake
For the abnormality detection model of aor brake, but in this example, common abnormal inspection is provided for multiple aor brakes
Survey model.
Abnormality detection model for deceleration includes deceleration model and deceleration threshold.For the different of aor brake
Normal detection model includes air brake pressure model and individual braking threshold.Abnormality detection model for vehicle group brake includes
Vehicle group brake model and vehicle group braking threshold.
In step S103, anomaly detector 150 is by using information database 101 and is stored in model database 102
Various abnormality detection models execute abnormality detection.Specifically, anomaly detector 150 is generated according to driving information for subtracting
The feature vector (for example, control command value) of velocity anomaly detection, and by using feature vector generated and deceleration mould
Type predicts deceleration.Residual error between the deceleration of prediction and actual deceleration is compared with deceleration threshold.When
When residual error is equal to or less than deceleration threshold, deceleration is confirmed as normally.When residual error is greater than deceleration threshold, deceleration quilt
It is determined as exception.
Similarly, anomaly detector 150 also according to driving information generate for aor brake abnormality detection feature to
It measures (for example, control command value), and air is predicted by using feature vector generated and air brake pressure model
Brake pressure.By the residual error and individual braking threshold progress between the air brake pressure of prediction and actual air brake pressure
Compare.When residual error is equal to or less than individual braking threshold, air brake pressure is confirmed as normally.When residual error is greater than a system
When dynamic threshold value, air brake pressure is confirmed as exception.
Anomaly detector 150 also according to driving information generate for vehicle group brake abnormality detection feature vector (for example,
Control command value), and multiple aor brakes are predicted by using feature vector generated and vehicle group brake model
Air brake pressure summation.By between the air brake pressure summation of prediction and actual air brake pressure summation residual error with
Vehicle group braking threshold is compared.When residual error is equal to or less than vehicle group braking threshold, summation is confirmed as normally.When residual error is big
When vehicle group braking threshold, summation is confirmed as exception.
In step S104, diagnostor 160 is by using deceleration abnormality detection result, aor brake abnormality detection knot
Fruit, vehicle group brake abnormality detection result and multiple diagnostic rules in diagnostic roles database are stored in diagnose diagnosis target
Vehicle vehicle group.Diagnostor 160 is specified with the matched diagnostic rule of these abnormality detection results, and determines and indicated by diagnostic rule
Diagnostic result.
In step S105, diagnostor 160 generates diagnosis output information according to diagnostic result, and in screen display device 600
Screen on show diagnosis output information.
In the processing of this flow chart, by using aor brake, deceleration and vehicle group brake abnormality detection result
These three execute diagnosis, but can execute diagnosis by using two kinds in these kinds.
Although some embodiments have been described, but these embodiments are merely given as examples, it is no intended to which limitation is originally
The range of invention.In fact, novel embodiment described herein can be implemented in the form of various other；In addition, not departing from this
In the case where the spirit of invention, can to embodiment described herein form carry out various omissions, substitutions and changes.
[label declaration]
10 brake bars
20 railways
30 wheels
41 brake block
42 tyre surface brakes
43 cylinders
50 load response devices
51 air springs
60a, 60b main motor
70 resistors
80 bow collectors
90 routes
100 apparatus for diagnosis of abnormality
101 information databases
102 model databases
(deceleration model, threshold value)
(air brake pressure model, threshold value)
(vehicle group brake model, threshold value)
103 testing result databases
110 information of vehicles collectors
120 environmental information collectors
130 data processors
140 model generators
150 anomaly detectors
160 diagnostors
170 alarm devices
151 CPU
152 input interfaces
153 display devices
154 communication devices
155 main storage means
156 external memories
157 buses
200 Vehicular systems
300 environmental information systems
400 terminals
401,402,403 normal distributions
500 input units
600 screen display devices
Claims (14)
1. a kind of apparatus for diagnosis of abnormality, comprising:
Anomaly detector is configured as the prediction mould based on the control command value for brake apparatus and the deceleration for vehicle
Type executes the abnormality detection of the decelerability of vehicle, and is configured as based on the control command value and for brake apparatus
The prediction model of brake force execute the abnormality detection of brake apparatus；And
Diagnostor is configured as diagnosing vehicle based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus
?.
2. apparatus for diagnosis of abnormality as described in claim 1, wherein
The control command value is used for the multiple brake apparatus being arranged in connected multiple vehicles,
Anomaly detector executes the abnormality detection of brake apparatus, and
Diagnostor diagnoses vehicle based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus.
3. apparatus for diagnosis of abnormality as claimed in claim 2, wherein diagnostor is based on diagnostic rule data diagnosis vehicle, described
Every kind of combination of the abnormality detection result of the abnormality detection result and brake apparatus of diagnostic rule data definition decelerability is examined
Disconnected result.
4. apparatus for diagnosis of abnormality as claimed in claim 2 or claim 3, wherein
Anomaly detector executes the system including brake apparatus based on the prediction model of the value for the brake force based on brake apparatus
The abnormality detection of dynamic system, and
Diagnostor diagnoses vehicle by using the abnormality detection result of braking system.
5. apparatus for diagnosis of abnormality as claimed in claim 4, wherein the value of the brake force based on brake apparatus is brake apparatus
The summation of brake force.
6. apparatus for diagnosis of abnormality as described in claim 4 or 5, wherein diagnostor diagnoses vehicle based on diagnostic rule data,
Abnormality detection result, the abnormality detection result of brake apparatus and the braking system of the diagnostic rule data definition decelerability
The diagnostic result of every kind of combination of the abnormality detection result of system.
7. such as apparatus for diagnosis of abnormality described in any one of claims 1 to 6, wherein anomaly detector is based on being used for deceleration
The predicted value of prediction model and the measured value of deceleration between difference execute the abnormality detection of decelerability.
8. the apparatus for diagnosis of abnormality as described in any one of claims 1 to 7, wherein anomaly detector is based on being used for brake force
The predicted value of prediction model and the measured value of brake force between difference execute the abnormality detection of brake apparatus.
9. apparatus for diagnosis of abnormality as claimed in claim 8, wherein when detecting abnormal in brake apparatus, diagnostor root
Anomalous effects degree letter is calculated according to the difference between the predicted value of the prediction model for deceleration and the measured value of deceleration
Breath.
10. apparatus for diagnosis of abnormality as claimed in claim 9, wherein diagnostor calculate predicted value at deceleration braking away from
The difference between braking distance from the deceleration at measured value is as anomalous effects degree information.
11. the apparatus for diagnosis of abnormality as described in any one of claims 1 to 10, wherein control command value is and brake force phase
The bid value of pass.
12. the apparatus for diagnosis of abnormality as described in any one of claims 1 to 11, wherein
Brake apparatus is aor brake, and
Brake force is air brake pressure.
13. a kind of abnormality diagnostic method, comprising:
The deceleration of vehicle is executed based on the prediction model of the control command value for brake apparatus and the deceleration for vehicle
The abnormality detection of energy:
The prediction model of brake force based on the control command value and for brake apparatus executes the abnormality detection of brake apparatus；
And
Vehicle is diagnosed based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus.
14. a kind of make computer execute the following computer program handled:
The deceleration of vehicle is executed based on the prediction model of the control command value for brake apparatus and the deceleration for vehicle
The abnormality detection of energy:
The prediction model of brake force based on control command value and for brake apparatus executes the abnormality detection of brake apparatus；And
Vehicle is diagnosed based on the abnormality detection result of decelerability and the abnormality detection result of brake apparatus.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2017-138257 | 2017-07-14 | ||
| JP2017138257A JP6889057B2 (en) | 2017-07-14 | 2017-07-14 | Information processing equipment, information processing methods and computer programs |
| PCT/JP2018/007060 WO2019012730A1 (en) | 2017-07-14 | 2018-02-26 | Anomaly diagnosis device, anomaly diagnosis method, and computer program |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109689470A true CN109689470A (en) | 2019-04-26 |
| CN109689470B CN109689470B (en) | 2021-06-29 |
Family
ID=61656283
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201880003331.7A Active CN109689470B (en) | 2017-07-14 | 2018-02-26 | Abnormality diagnosis device, abnormality diagnosis method, and computer program |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP6889057B2 (en) |
| CN (1) | CN109689470B (en) |
| TW (1) | TWI691420B (en) |
| WO (1) | WO2019012730A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112158237A (en) * | 2020-09-24 | 2021-01-01 | 交控科技股份有限公司 | Deep fusion system integrating TCMS and ATO functions and train |
| CN112440746A (en) * | 2019-08-29 | 2021-03-05 | 北京新能源汽车股份有限公司 | Vehicle-mounted terminal communication fault control method, device and system and vehicle |
| CN112793557A (en) * | 2019-11-13 | 2021-05-14 | 纳博特斯克有限公司 | Brake control device, abnormality determination method, and computer-readable storage medium |
| CN113110399A (en) * | 2021-05-20 | 2021-07-13 | 三一重机有限公司 | Method and system for diagnosing faults of working machine |
| CN115946673A (en) * | 2022-12-28 | 2023-04-11 | 成都赛力斯科技有限公司 | Fault diagnosis method, system, equipment and medium for automobile braking |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6714626B2 (en) | 2018-02-26 | 2020-06-24 | 株式会社京三製作所 | Railway equipment condition determination device and railway equipment condition determination method |
| WO2020144939A1 (en) * | 2019-01-11 | 2020-07-16 | パナソニックＩｐマネジメント株式会社 | Diagnosis system, diagnosis method, program, and recording medium |
| JP2020132006A (en) * | 2019-02-21 | 2020-08-31 | 株式会社京三製作所 | Railway facility state determination device and railway facility state determination method |
| JP7230691B2 (en) * | 2019-05-30 | 2023-03-01 | 株式会社デンソー | Abnormality detection method, abnormality detection device, and abnormality detection system |
| US20220234635A1 (en) * | 2019-06-27 | 2022-07-28 | Mitsubishi Electric Corporation | Deterioration diagnosis apparatus, deterioration diagnosis system, and deterioration diagnosis method |
| CN110751633A (en) * | 2019-10-11 | 2020-02-04 | 上海眼控科技股份有限公司 | Multi-axis cart braking detection method, device and system based on deep learning |
| JP7273755B2 (en) * | 2020-03-16 | 2023-05-15 | 株式会社東芝 | Information processing device, information processing method and program |
| WO2021199173A1 (en) * | 2020-03-30 | 2021-10-07 | 三菱電機株式会社 | Monitoring system |
| JP7046148B1 (en) * | 2020-11-26 | 2022-04-01 | 三菱電機株式会社 | Diagnostic system, diagnostic program and diagnostic method |
| WO2022244051A1 (en) * | 2021-05-17 | 2022-11-24 | 三菱電機株式会社 | Data accumulation device, equipment monitoring system, and data accumulation method |
| JP7475553B2 (en) | 2021-09-06 | 2024-04-26 | 三菱電機株式会社 | Health diagnostic device and health diagnostic method |
| WO2023243077A1 (en) * | 2022-06-17 | 2023-12-21 | 三菱電機株式会社 | Soundness evaluating device, soundness evaluating method, and soundness evaluating program |
| WO2023248130A1 (en) * | 2022-06-21 | 2023-12-28 | Faiveley Transport Italia S.P.A. | Methods for verifying the operation of at least one braking means of at least one vehicle and corresponding verification systems |
| WO2023248378A1 (en) * | 2022-06-22 | 2023-12-28 | 三菱電機株式会社 | Actual deceleration acquisition device, deterioration discrimination device, brake control device, actual deceleration acquisition method, deterioration discrimination method, and brake control method |
| CN117078687B (en) * | 2023-10-17 | 2023-12-15 | 常州海图信息科技股份有限公司 | Track inspection system and method based on machine vision |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5744707A (en) * | 1996-02-15 | 1998-04-28 | Westinghouse Air Brake Company | Train brake performance monitor |
| WO1999006809A1 (en) * | 1997-07-29 | 1999-02-11 | Tom Lalor | Method and apparatus for determining vehicle brake effectiveness |
| WO2007095401A3 (en) * | 2006-02-13 | 2008-04-17 | New York Air Brake Corp | Distributed train intelligence system and method |
| JP2011162046A (en) * | 2010-02-09 | 2011-08-25 | Hitachi Ltd | Onboard control device |
| CN104024070A (en) * | 2011-12-22 | 2014-09-03 | 西门子公司 | Method and arrangement for monitoring a brake system of a brake arrangement of a rail vehicle |
| CN104875772A (en) * | 2015-05-29 | 2015-09-02 | 南京南车浦镇城轨车辆有限责任公司 | Test train fixed-point parking early warning braking device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0891206A (en) * | 1994-09-21 | 1996-04-09 | Hitachi Ltd | Brake abnormality detecting device |
| TW200949225A (en) * | 2008-05-21 | 2009-12-01 | Univ Kun Shan | Method for inspecting abnormal conditions of an automatic transmission box |
| JP2012205332A (en) * | 2011-03-24 | 2012-10-22 | Hokkaido Railway Co | Vehicle monitoring device and vehicle monitoring system using the same |
| DE102012216218A1 (en) * | 2011-09-12 | 2013-03-14 | Continental Teves Ag & Co. Ohg | Sensor system comprising a fusion filter for common signal processing |
| TWI631033B (en) * | 2012-06-19 | 2018-08-01 | 張福齡 | Vehicle idle speed stop and start and control method for displaying vehicle position |
| JP2013100111A (en) * | 2013-03-07 | 2013-05-23 | Mitsubishi Electric Corp | Abnormal cause specifying device, abnormal cause specifying system, and abnormal cause specifying method |
| US9361650B2 (en) * | 2013-10-18 | 2016-06-07 | State Farm Mutual Automobile Insurance Company | Synchronization of vehicle sensor information |
| JP6588814B2 (en) * | 2015-12-17 | 2019-10-09 | 株式会社東芝 | Abnormality diagnosis apparatus and method |
-
2017
- 2017-07-14 JP JP2017138257A patent/JP6889057B2/en active Active
-
2018
- 2018-02-26 CN CN201880003331.7A patent/CN109689470B/en active Active
- 2018-02-26 WO PCT/JP2018/007060 patent/WO2019012730A1/en active Application Filing
- 2018-03-13 TW TW107108344A patent/TWI691420B/en active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5744707A (en) * | 1996-02-15 | 1998-04-28 | Westinghouse Air Brake Company | Train brake performance monitor |
| WO1999006809A1 (en) * | 1997-07-29 | 1999-02-11 | Tom Lalor | Method and apparatus for determining vehicle brake effectiveness |
| WO2007095401A3 (en) * | 2006-02-13 | 2008-04-17 | New York Air Brake Corp | Distributed train intelligence system and method |
| JP2011162046A (en) * | 2010-02-09 | 2011-08-25 | Hitachi Ltd | Onboard control device |
| CN104024070A (en) * | 2011-12-22 | 2014-09-03 | 西门子公司 | Method and arrangement for monitoring a brake system of a brake arrangement of a rail vehicle |
| CN104875772A (en) * | 2015-05-29 | 2015-09-02 | 南京南车浦镇城轨车辆有限责任公司 | Test train fixed-point parking early warning braking device |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112440746A (en) * | 2019-08-29 | 2021-03-05 | 北京新能源汽车股份有限公司 | Vehicle-mounted terminal communication fault control method, device and system and vehicle |
| CN112793557A (en) * | 2019-11-13 | 2021-05-14 | 纳博特斯克有限公司 | Brake control device, abnormality determination method, and computer-readable storage medium |
| CN112793557B (en) * | 2019-11-13 | 2023-03-21 | 纳博特斯克有限公司 | Brake control device, abnormality determination method, and computer-readable storage medium |
| CN112158237A (en) * | 2020-09-24 | 2021-01-01 | 交控科技股份有限公司 | Deep fusion system integrating TCMS and ATO functions and train |
| CN112158237B (en) * | 2020-09-24 | 2022-07-19 | 交控科技股份有限公司 | Deep fusion system integrating TCMS and ATO functions and train |
| CN113110399A (en) * | 2021-05-20 | 2021-07-13 | 三一重机有限公司 | Method and system for diagnosing faults of working machine |
| CN115946673A (en) * | 2022-12-28 | 2023-04-11 | 成都赛力斯科技有限公司 | Fault diagnosis method, system, equipment and medium for automobile braking |
| CN115946673B (en) * | 2022-12-28 | 2024-04-19 | 重庆赛力斯凤凰智创科技有限公司 | Fault diagnosis method, system, equipment and medium for automobile brake |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201908181A (en) | 2019-03-01 |
| JP2019022306A (en) | 2019-02-07 |
| JP6889057B2 (en) | 2021-06-18 |
| WO2019012730A1 (en) | 2019-01-17 |
| TWI691420B (en) | 2020-04-21 |
| CN109689470B (en) | 2021-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109689470A (en) | Apparatus for diagnosis of abnormality, abnormality diagnostic method and computer program | |
| CN109641603B (en) | Abnormality detection device, abnormality detection method, and computer program | |
| EP1900597B1 (en) | Diagnostic system and method for monitoring a rail system | |
| JP5416630B2 (en) | Moving object abnormality judgment support system | |
| KR102527319B1 (en) | System and method for machine learning prognostics and health management(phm) based on feature vector data of rolling stock parts | |
| SA515360150B1 (en) | Method for diagnosing railroad components of a railroad network for rail transport | |
| CN109641602A (en) | Abnormality detecting apparatus, method for detecting abnormality and non-transitory computer-readable medium | |
| US20210217256A1 (en) | Method and Apparatus for Diagnosing and Monitoring Vehicles, Vehicle Components and Routes | |
| KR101944070B1 (en) | Platform screen door management system and method with big data | |
| CN107045332A (en) | ECU earth fault xegregating units for delay system | |
| KR20200058132A (en) | Railway vehicle major component and system diagnosis apparatus | |
| CN112541430A (en) | Fault diagnosis method for fusion of temperature signal and noise signal | |
| US20160148447A1 (en) | Method, system and computer-readable recording medium for managing abnormal state of vehicle | |
| EP3246223B1 (en) | State monitoring device, state monitoring system, and train | |
| Gregori et al. | Assessment of catenary condition monitoring by means of pantograph head acceleration and Artificial Neural Networks | |
| JP6619083B2 (en) | Data integration analysis system | |
| CN109883598B (en) | Method for detecting traction force of rail locomotive | |
| CN112685827B (en) | Data recording device and data recording method | |
| US20240169772A1 (en) | Vehicle abnormality detection device and vehicle abnormality detection method | |
| KR20220089797A (en) | Railway vehicle safety diagnosis apparatus | |
| KR20230078596A (en) | A prediction system for failure of hydrogen charging stations using big data platforms and machine learning | |
| KR20230120751A (en) | Method and system of diagnosing condition of display for railway vehicle | |
| KR20210132352A (en) | Method and system of diagnosing defect and calculating remaining life of components of railway vehicle | |
| CN117131957A (en) | Machine learning-based engine oil low pressure fault prediction method and system | |
| CN117944742A (en) | Novel tram operation management system |
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 |