CN109581995A - A kind of intelligent diagnosis system and method - Google Patents
A kind of intelligent diagnosis system and method Download PDFInfo
- Publication number
- CN109581995A CN109581995A CN201710900078.2A CN201710900078A CN109581995A CN 109581995 A CN109581995 A CN 109581995A CN 201710900078 A CN201710900078 A CN 201710900078A CN 109581995 A CN109581995 A CN 109581995A
- Authority
- CN
- China
- Prior art keywords
- board
- data
- diagnosis
- failure
- intelligent
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
- G05B23/0213—Modular or universal configuration of the monitoring system, e.g. monitoring system having modules that may be combined to build monitoring program; monitoring system that can be applied to legacy systems; adaptable monitoring system; using different communication protocols
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a kind of intelligent diagnosis system and methods, the system includes: main system, including system master board, main hinge board and several data boards and the data/address bus connecting with system master board, main hinge board and data board and control bus；Subsystem, including system master board, from hinge board and several data boards and with system master board, the data/address bus being connect from hinge board and data board and control bus；It main hinge board and is connect from hinge board with sensor；Diagnosis prediction board is connect with system master board.The present invention can be achieved the data detected to sensor and carry out real-time operation, avoids the time-consuming operation of the frequency I/O of batch data, using configurable data of interest obtaining mode, avoids the processing of mass of redundancy data；" online " processing and analysis are carried out to the operation intermediate data of data board in equipment simultaneously, failure can be predicted, accelerate manufacturing schedule, improve the yield of equipment.
Description
Technical field
The present invention relates to Diagnosis Technique fields, and in particular to a kind of intelligent diagnosis system and method.
Background technique
With advances in technology with the development of computer science, various equipment have begun to enlargement, complicate, intelligence
Change and develop, so that machine or equipment are more nearly or meet the operating habit and functional requirement of natural person.However likewise, machine
Intelligence with equipment is the intelligentized equipment operation using machine design, manufacture and the complexity of operation as precondition
In the process usually along with various problem.Showing for problem often has " hysteresis quality ", once that is, machine shows
People may find that the problem of, machine is " being eaten up with illness " already, relatively very numerous of positioning problems, reparation, maintenance etc. at this time
It is trivial, usually rely on engineer experience carry out problem investigation, can not quick positioning question, rapid restorer work normally, and
It is unable to complete the succession of technology.
The conventional means that industry solves problem above is to increase various sensing equipments, to temperature, humidity, the pressure inside equipment
The factors such as power, electric current are detected in real time, since the generation of failure has the characteristics that " randomness " and " contingency ", are needed
The data that will test are stored in real time, are considerably increased the time of storage in this way, are more increased the space of storage, i.e.,
Space complexity and time complexity can linear increase or even exponentially increases；And when an error occurs, often it is only necessary to events
The data of former seconds or first few minutes before barrier generation, a large amount of data are rubbish or redundant data, are not only occupied a large amount of
Hardware space increases hardware cost, same because of the I/O operation of computer high-frequency, considerably increases runing time, affects
The overall performance of equipment.
In addition, industry is typically all shutdown processing after the failure occurred to the detection and processing of failure, not using some events
Hinder forecasting mechanism, simple or double faults generations is avoided to will lead to MTTR although failure problems may finally be solved
(Mean Time To Repair, mean time to restoration) is larger, affects manufacturing schedule, reduces the yield of equipment.
Summary of the invention
The present invention aiming at the problems existing in the prior art, provides a kind of achievable failure anticipation and failure pretreatment,
Accelerate manufacturing schedule, improves the intelligent diagnosis system and method for equipment yield.
In order to solve the above-mentioned technical problem, the technical scheme is that a kind of intelligent diagnosis system, comprising:
Main system, including system master board, main hinge board and several data boards and with the systematic master control board
The data/address bus and control bus that card, main hinge board are connected with data board；
Subsystem, including system master board, from hinge board and several data boards and with the systematic master control board
Card, the data/address bus connected from hinge board with data board and control bus；
It the main hinge board and is connect from hinge board with sensor；
Diagnosis prediction board is connect with the system master board.
Further, it is connected between the diagnosis prediction board and the system master board by industry ethernet.
Further, the system master board uses powerpc board card.
Intelligent diagnosis system according to claim 1, which is characterized in that the diagnosis prediction board uses host computer
Or powerpc board card.
Further, the diagnosis prediction board includes failure predication module, database and failure reception and processing module.
Further, the diagnosis prediction board also with the main hinge board and between hinge board pass through HSSL light
Fine transfer bus connects bus connection with serial ports.
Further, the main system and subsystem further include fault diagnosis board, and the fault diagnosis board is connected to
The data/address bus and control bus are transmitted between the fault diagnosis board and the diagnosis prediction board by HSSL optical fiber
Bus connects bus connection with serial ports.
The present invention also provides a kind of diagnostic methods of intelligent diagnosis system as described above, comprising the following steps:
S1: the main hinge board and the detection data for obtaining sensor in real time from hinge board, and by the detection data
Database card is sent to be calculated；
S2: the intermediate operation data of data board is periodically acquired, and is sent to diagnosis prediction board；
S3: intermediate operation data carries out failure predication to the diagnosis prediction board based on the received, and prediction result is anti-
It is fed to system master board.
Further, in the step S2, database is periodically acquired by the main hinge board and from hinge board
The intermediate operation data of card, and it is sent to the diagnosis prediction board.
Further, in the step S2, the intermediate of data board is periodically acquired by fault diagnosis board and runs number
According to, progress failure predication, and predictive information and intermediate operation data are sent to the diagnosis prediction board.
Further, in the step S2, failure predication the following steps are included:
S21: the fault diagnosis board is periodically obtained according to the time parameter and data of interest in configuration file
The intermediate operation data of data board, and be put into memory buffer；
S22: the fault diagnosis board carries out real-time monitoring to the intermediate operation data of acquisition, judges whether data value is located
In the safe range of configuration file setting；
S23: when data value exceeds safe range (0, m%), warning message is reported to give failure predication board, and anti-in real time
Driving assembly in feeding system master control board card, is adjusted correspondingly, and avoids the deterioration of operation conditions；
S24: when data value exceeds safe range > m%, reporting fault information gives the processing of failure predication board, while straight
Driving assembly in reversed feeding system master control board card, initializes, and wherein m is real number, is set by configuration file.
Further, in the step S24, when data value exceeds safe range > m%, fault diagnosis board is to failure
Type carries out coded treatment, and by serial down trigger, notifies diagnosis prediction board.
Further, the step S3 further includes sending instruction first after diagnosis prediction board receives fault message
To suspend all subsystem movements, and failure is handled, judges failure mechanism, if run action retries, if allowing,
It sends " retrying " and orders the subsystem acted to participation, retry this movement, if retrying same failure, report to server
End；If not allowing, " system mistake " is sent to server end, waits manual intervention.
Further, the step S3 the following steps are included:
S31: the diagnosis prediction board is configured according to the sampling time in configuration file, data of interest；
S32: the fixed n servo period sampling primary fault of the diagnosis prediction board diagnoses the data of board and storage is arrived
In database, wherein n is natural number, is arranged by configuration file；
S33: the history in data and database that failure predication module in the diagnosis prediction board samples this
Data carry out integrated treatment, fitting data change curve, and find corresponding rule in database, obtain failure predication information；
S34: the diagnosis prediction board feeds back to failure predication information accordingly by the main hinge board in main system
Subsystem, corresponding adjustment and operation are done by the system master board in subsystem.
Further, in the step S33, it is described rule in the form of fault tree save, i.e., a kind of data trend with
A kind of failure mode type.
Further, in the step S33, if not finding rule, the failure predication module carries out failure training, and
It is stored as new rule.
Further, in the step S33, song is changed by least square method or the method fitting data for being averaging trend
Line.
Further, further include step S4, direct fault location is carried out to the diagnosis prediction board or systematic master control board card, with
The diagnosis effect of checking system.
There is following advantage compared with prior art in intelligent diagnosis system provided by the invention and method:
(1) real-time operation is carried out to the data of sensor detection, avoids the time-consuming behaviour of the frequency I/O of batch data
Make；
(2) using configurable data of interest obtaining mode, the processing of mass of redundancy data is avoided；
(3) " online " processing and analysis are carried out to the operation intermediate data of data board in equipment, failure can be carried out pre-
It surveys；
(4) Intelligent fault judgement and online processing, avoid equipment downtime wait artificial interference, accelerate produce into
Degree, improves the yield of equipment；
(5) " down trigger " and " pause " mode is used, avoids the extension of failure, and be easy to fault location；
(6) using distributed diagnostics and processing model, the failure of main system and subsystem can be carried out quickly, in time
Processing；
(7) it using " failure training " and " rule process " mode and deposits, quickly, in real time, accurately predicts the generation of failure
And handling failure.
(8) pass through the generation of direct fault location simulated failure, it is reliable to improve system for the troubleshooting capability of test macro
Property.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of intelligent diagnosis system in the embodiment of the present invention 1；
Fig. 2 is the structural schematic diagram of diagnosis prediction board in the embodiment of the present invention 1；
Fig. 3 is the structural schematic diagram of diagnosis prediction board in the embodiment of the present invention 2；
Fig. 4 be in the embodiment of the present invention 2 fault diagnosis board to the judgement schematic diagram of three kinds of states.
It is as shown in the figure: 100, main system；200, subsystem；300, diagnosis prediction board；400, server end；1, system master
Control board；2, main hinge board；3, data board；4, data/address bus；5, control bus；6, from hinge board；7, Ethernet is total
Line；8, HSSL optical fiber transfer bus；9, serial ports connects bus；10, failure predication module；11, database；12, failure receive with
Processing module；13, fault diagnosis board.
Specific embodiment
The present invention is described in detail with reference to the accompanying drawing.
Embodiment 1
As shown in Figure 1, the present invention provides a kind of intelligent diagnosis system, comprising:
Main system 100, including system master board 1, main hinge board (Master Hub Board, MHB) 2 and several numbers
According to board 3 and with 3 data/address bus connecting of the system master board 1, main hinge board 2 and data board (DATA board)
4 and control bus 5.The main hinge board 2 sensor connection corresponding with the main system 100, the survey for receiving sensor
Data are measured, and is issued to each data board 3 and is calculated.
Subsystem 200, including system master board 1, from hinge board 6 and several data boards 3 and with the system
Master control board card 1, the data/address bus 4 connected from hinge board 6 and data board 3 and control bus 5, it is described from hinge board 6 with
The corresponding sensor connection of the subsystem, for the measurement data of receiving sensor, and is issued to each data board 3 of the system
It is calculated.
Wherein, system master board 1 uses PowerPC (Performance Optimization With Enhanced
RISC-Performance Computing, the central processing unit of reduced instruction set computer RISC Architecture) board, also referred to as PPC board, it is main
It is responsible for receiving the order of diagnosis prediction board 300, and other boards will be sent in the system after the command interpretation.Specifically
, system master board 1 receives order (such as initialization, machine ginseng that diagnosis prediction board 300 issues by industry ethernet 7
Number issues, runs the distribution etc. of firmware), and by after command interpretation, order is issued to by each plate in system by data/address bus 4
SRIO, SDB, MDB, PCIe etc. can be used in card, including main hinge board 2 and data board 3, the data/address bus 4；System was run
Cheng Zhong, some other auxiliary informations (such as Trace) will be sent to PPC board by control bus 5, complete to the information
Processing or storage, the control bus 5 be VME64x or VPX bus, or be Ethernet.Meanwhile the driving journey of the system
Sort run carries out parameter to DATA board by control bus 5 and issues and control on PPC board.
Data board 3 is mainly used for the operation and processing of data during the realizing and controlling of control algolithm.System is initial
After change, data board 3 is constantly in ready state, waits for the arrival of sensor detection data；It, will when there is data arrival
Data are quickly moved into the RAM of the board, complete this calculating with most fast speed, by calculated result according to arranging in advance
Sequence carries out data broadcasting by data/address bus 4, and all boards in system can obtain the calculating knot from data/address bus 4
Fruit data are simultaneously stored, and the interaction of data between plate is realized；Intermediate operation data in data handling procedure can be according to configuration
Demand is written in the external memory or DPRAM of data board 3, to be supplied to main hinge board 2 or be grabbed from hinge board 6
It takes, data is directly abandoned after the completion of avoiding operation from executing.
Diagnosis prediction board (Master Diagnosis Trigger Board, MDT) 300, using host computer or
Powerpc board card is connect with the system master board 1 by industry ethernet 7, at the same diagnosis prediction board 300 also with institute
It states main hinge board 2 and connects the connection of bus 9 with serial ports from HSSL optical fiber transfer bus 8 is passed through between hinge board 6.Wherein go here and there
Row bus 5 can be RS232, RS485, USB, IEEE1394 etc., as shown in Fig. 2, the diagnosis prediction board 300 includes that failure is pre-
Module 10, database 11 and failure reception and processing module 12 are surveyed, wherein failure predication module 10 is using " rule process " and " event
Barrier training " two ways carries out the prediction processing of failure, and improves the diagnosis rule of database 11 in real time；Database 11 is mainly used
In storage processing rule, rule is saved in the form of " fault tree ", i.e., a kind of data trend corresponds to a kind of fault type；Failure
It receives and is mainly responsible for troubleshooting with processing module 12.Specifically, main hinge board 2 or from the periodically crawl of hinge board 6
DATA board is written to the data in external memory or DPRAM, and every n servo period, n is natural number, is arranged by configuration file,
The operation data of this servo period is uploaded to diagnosis prediction board 300, diagnosis prediction board by HSSL optical fiber transfer bus 8
It stores after 300 receptions into database 11；Failure predication module 10 is by the operation data for grabbing this and going through in database 11
History data carry out integrated treatment, fitting data change curve, and find corresponding rule in database 11, obtain failure predication letter
Breath, and corresponding subsystem 200 is fed back to by the main hinge board 2 in main system 100, by the system master in subsystem 200
Control board 1 does corresponding adjustment and operation；If not finding rule, " failure training " is carried out, and be stored as new rule.
Wherein the mode of supplemental characteristic fitting includes two kinds:
Least square method: due to parameter values such as electric current, voltage, temperature, it should keep as far as possible stable in equipment operation, increase
Long trend is slow, can be set as a simple regression linear equation yi=f (ti)=ati+b+ξi, for different supplemental characteristics, if
The function item set is different, such as velocity and acceleration, is multistage multinomial, according to f (yi,ti)=∑ [yi-ati-b]2, to f
(yi,ti) ask local derviation that the value of a and b can be obtained, so that it is determined that relation function formula.
MA method: i.e. averaging trend, settable function are yi=f (ti)=ati+ b, by the average value of point-to-point transmission, by
Step determines and repairs the formula, passes through the formula predictions fault trend.
The diagnostic method of above-mentioned intelligent diagnosis system is also provided in the present embodiment, comprising the following steps:
S1: the main hinge board 2 and the detection data for obtaining sensor in real time from hinge board 6, and by the testing number
It is calculated according to data board 3 is sent to, the intermediate operation data of data board 3 in data processing can be according to configuration need
It asks and is written in its external memory or DPRAM.
S2: the main hinge board 2 and the intermediate operation data that data board 3 is periodically acquired from hinge board 6, and will
It is sent to diagnosis prediction board 300；Specifically, the main hinge board 2 or from the periodically crawl DATA board of hinge board 6
The data being written in external memory or DPRAM, and every n servo period, n is natural number, is arranged by configuration file, and HSSL is passed through
Optical fiber transfer bus 8 uploads the operation data of this servo period to diagnosis prediction board 300, and diagnosis prediction board 300 receives
After store into database 11.
S3: intermediate operation data carries out failure predication to the diagnosis prediction board 300 based on the received, and by prediction result
Feed back to system master board 1.Specifically, failure predication module 10 is by the operation data for grabbing this and going through in database 11
History data carry out integrated treatment, fitting data change curve, and find corresponding rule in database 11, obtain failure predication letter
Breath, and corresponding subsystem 200 is fed back to by the main hinge board 2 in main system 100, by the system master in subsystem 200
Control board 1 does corresponding adjustment and operation；If not finding rule, " failure training " is carried out, and be stored as new rule.
S4: direct fault location is carried out to the diagnosis prediction board 300 or system master board 1, with the diagnosis of checking system
Effect.I.e. direct fault location can be injected by diagnosis prediction board 300, and diagnosis prediction board 300 issues fault message packing
To main system 100 and subsystem 200, and main system 100 and the corresponding failure of subsystem 200 are handled in real time；Or it is directed to principal series
System 100 or the system master board 1 of subsystem 200 are injected separately into, and diagnosis prediction board 300 receives and processes 100 He of main system
The failure of subsystem 200.
Embodiment 2
As shown in figure 3, unlike the first embodiment, main system 100 described in the present embodiment is washed system 200 with point and is also wrapped
Fault diagnosis board (Slave Diagnosis Trigger Board, SDT) 13 is included, the fault diagnosis board 13 is connected to
The data/address bus 4 and control bus 5 pass through HSSL between the fault diagnosis board 13 and the diagnosis prediction board 300
Optical fiber transfer bus 8 connects the connection of bus 9 with serial ports.
The function of fault diagnosis board 13 mainly includes following aspect:
1, the configurability of time and data, can be according to the time parameter and data of interest of configuration file DTS.cfg
It is configured, if movement subsystem, and data can use the voltage or current data of motor；If illumination subsystem, then data
The parameters such as desirable laser intensity, laser dosage；If environment subsystem, then data can use the parameters such as temperature, pressure, humidity.When
So it is not limited in the above parameter, design parameter is by actual scene or engineer's self-defining.
2, in-between operation data is grabbed in every DPRAM or external memory for removing each data board 3 at regular intervals, and be put into
In the memory buffer that fault diagnosis board 13 is opened up by every block number according to board 3；Time as unit of the number of servo period, just
The beginningization moment reads from configuration file DTS.cfg, and the same parameter can be arranged in real time by user interface；Since failure is sent out
For raw probability often in initialization and highest when machine startup, therefore, time interval at this time is small as far as possible, may be configured as 1 and watches
Take the period；It, can according to demand or actual conditions are adjusted in real time after stable equipment operation；
3, intermediate operation data is monitored in real time, judges whether data value is in safe range, the safe model
It encloses and is arranged by configuration file DTS.cfg；As shown in figure 4, safe range may be configured as three kinds of states: health, failure and inferior health；
Health status is the data in safe range；Sub-health state be beyond secure threshold m% within；Malfunction is then super
Secure threshold m% or more out, wherein m is real number, is set by configuration file；
4, when the parameter value of monitoring exceeds safe range (0, m%), i.e., system is in sub-health state at this time, and failure is examined
Disconnected board 13 reports warning message to failure predication board 300 by series connection bus 9, and Real-time Feedback is to systematic master control board
Driving assembly in card 1, is adjusted correspondingly, avoids the deterioration of operation conditions；
5, when the parameter value of monitoring exceeds An full Fan Wei≤m%, system is defined as malfunction, fault diagnosis at this time
Board 13 carries out coded treatment to fault type first, and gives failure predication plate by 9 reporting fault information of series connection bus
300 processing of card, while the driving assembly being fed directly in system master board 1, initialize, avoid in failure
Wait state sends " Retry " or other requests convenient for upper layer, and wherein m is real number, is set by configuration file.
Corresponding, the failure in diagnosis prediction board 300 receives the failure for receiving system with processing module 12 and believes
After breath, transmission event is suspended all subsystems 200 and is acted first, and handles failure, judges failure mechanism, if fortune
Action retries, if allowing, sends " Retry " and orders the subsystem 200 acted to participation, retry this movement；If not permitting
Perhaps, then " system mistake " is sent to server end, waits manual intervention.
The diagnostic method of intelligent diagnosis system described in the present embodiment, comprising the following steps:
S1: the main hinge board 2 and the detection data for obtaining sensor in real time from hinge board 3, and by the testing number
It is calculated according to data board 3 is sent to；The intermediate operation data of data board 3 in data processing can be according to configuration need
It asks and is written in its external memory or DPRAM.
S2: fault diagnosis board 13 periodically acquires the intermediate operation data of data board 3, and it is pre- to be sent to diagnosis
Drafting board card 300；Fault diagnosis board 13 periodically acquires the intermediate operation data of data board 3, carries out failure predication, and will be pre-
Measurement information and intermediate operation data are sent to the diagnosis prediction board 300.Wherein failure predication the following steps are included:
S21: the fault diagnosis board 13 is according to the time parameter and data of interest in configuration file DTS.cfg, week
The intermediate operation data of the crawl data board 3 of phase property, naturally it is also possible to actively obtain, and be put into memory buffer；It needs
Bright, if movement subsystem, and data of interest can use the voltage or current data of motor；If illumination subsystem, then
Data can use the parameters such as laser intensity, laser dosage；If environment subsystem, then data of interest can use temperature, pressure, humidity
Etc. parameters.For the above parameter, it is not limited in the above parameter, design parameter is by actual scene or engineer's self-defining.
S22: 300 pairs of the fault diagnosis board intermediate operation datas obtained carry out real-time monitoring, judge that data value is
In the no safe range in configuration file setting；The safe range is arranged by configuration file DTS.cfg；Safe range is settable
For three kinds of states: health, failure and inferior health；Health status is the data in safe range；Sub-health state be beyond
Within secure threshold m%；Malfunction is then beyond secure threshold m% or more, and wherein m is real number, is set by configuration file；
S23: when data value exceeds safe range (0, m%), fault diagnosis board 13 is reported by series connection bus 9
Warning message is to failure predication board 300, and Real-time Feedback is adjusted accordingly to the driving assembly in system master board 1
It is whole, avoid the deterioration of operation conditions；
S24: when data value exceeds safe range > m%, system is defined as malfunction, fault diagnosis board at this time
13 carry out coded treatment to fault type first, and by serial down trigger, notify diagnosis prediction board 300, while directly
The driving assembly in system master board 1 is fed back to, is initialized, wherein m is real number, is set by configuration file.
S3: intermediate operation data carries out failure predication to the diagnosis prediction board 300 based on the received, and by prediction result
Feed back to system master board 1.The following steps are included:
S31: the diagnosis prediction board 300 is configured according to the sampling time in configuration file, data of interest；
S32: the fixed n servo period sampling of the diagnosis prediction board 300 is primary (can actively obtain, or passive receiving)
The data of fault diagnosis board 13 are simultaneously stored into database 11, and wherein n is natural number, is arranged by configuration file；
S33: in the data and database 11 that failure predication module 10 in the diagnosis prediction board 300 samples this
Historical data carry out integrated treatment, it is pre- to obtain failure for fitting data change curve, and find corresponding rule in database 11
Measurement information；
S34: the diagnosis prediction board 300 feeds back failure predication information by the main hinge board 1 in main system 100
To corresponding subsystem 200, corresponding adjustment and operation are done by subsystem 200 system master board 1.
In addition, it is temporary to send instruction first after diagnosis prediction board 300 receives the fault message of fault diagnosis board 13
To stop all subsystems 200 to act, and failure is handled, judges failure mechanism, if run action retries, if allowing,
It then sends " retrying " and orders the subsystem acted to participation, retry this movement, if retrying same failure, report to server
End 400 waits manual intervention；If not allowing, " system mistake " is sent to server end 400, waits manual intervention.
Compared to embodiment 1, by increasing fault diagnosis board 13 in the technical solution that is provided in the present embodiment, using distribution
The pre-detection and online processing of failure, further mention in formula fault diagnosis and dealing model realization main system 100 and subsystem 200
High efficiency.
In conclusion, there is following advantage compared with prior art in intelligent diagnosis system provided by the invention and method:
(1) real-time operation is carried out to the data of sensor detection, avoids the time-consuming behaviour of the frequency I/O of batch data
Make；
(2) using configurable data of interest obtaining mode, the processing of mass of redundancy data is avoided；
(3) " online " processing and analysis are carried out to the operation intermediate data of data board 3 in equipment, failure can be carried out pre-
It surveys；
(4) Intelligent fault judgement and online processing avoid equipment downtime and wait artificial interference, saved the time, mentioned
High efficiency；
(5) " down trigger " and " pause " mode is used, avoids the extension of failure, and be easy to fault location；
(6) using distributed diagnostics and processing model, the failure of main system 100 and subsystem 200 can be carried out fast
Speed, in time processing；
(7) it using " failure training " and " rule process " mode and deposits, quickly, in real time, accurately predicts the generation of failure
And handling failure.
(8) pass through the generation of direct fault location simulated failure, it is reliable to improve system for the troubleshooting capability of test macro
Property.
Although embodiments of the present invention are illustrated in specification, these embodiments are intended only as prompting,
It should not limit protection scope of the present invention.It is equal that various omission, substitution, and alteration are carried out without departing from the spirit and scope of the present invention
It should be included within the scope of the present invention.
Claims (18)
1. a kind of intelligent diagnosis system characterized by comprising
Main system, including system master board, main hinge board and several data boards and with the system master board, master
The data/address bus and control bus that hinge board is connected with data board；
Subsystem, including system master board, from hinge board and several data boards and with the system master board, from
The data/address bus and control bus that hinge board is connected with data board；
It the main hinge board and is connect from hinge board with sensor；
Diagnosis prediction board is connect with the system master board.
2. intelligent diagnosis system according to claim 1, which is characterized in that the diagnosis prediction board and the system master
It controls and is connected between board by industry ethernet.
3. intelligent diagnosis system according to claim 1, which is characterized in that the system master board uses PowerPC
Board.
4. intelligent diagnosis system according to claim 1, which is characterized in that the diagnosis prediction board using host computer or
Powerpc board card.
5. intelligent diagnosis system according to claim 1, which is characterized in that the diagnosis prediction board includes failure predication
Module, database and failure receive and processing module.
6. intelligent diagnosis system according to claim 1, which is characterized in that the diagnosis prediction board also with the main pivot
It knob board and is connected from passing through HSSL optical fiber transfer bus between hinge board and connect with serial ports bus.
7. intelligent diagnosis system according to claim 1, which is characterized in that the main system and subsystem further include failure
Diagnose board, the fault diagnosis board is connected to the data/address bus and control bus, the fault diagnosis board with it is described
Bus connection is connected with serial ports by HSSL optical fiber transfer bus between diagnosis prediction board.
8. a kind of diagnostic method of intelligent diagnosis system as described in claim 1, which comprises the following steps:
S1: the main hinge board and the detection data for obtaining sensor in real time from hinge board, and the detection data is sent
It is calculated to database card；
S2: the intermediate operation data of data board is periodically acquired, and is sent to diagnosis prediction board；
S3: intermediate operation data carries out failure predication to the diagnosis prediction board based on the received, and prediction result is fed back to
System master board.
9. intelligent diagnosing method according to claim 8, which is characterized in that in the step S2, pass through the main hinge
Board and the intermediate operation data that data board is periodically acquired from hinge board, and it is sent to the diagnosis prediction plate
Card.
10. intelligent diagnosing method according to claim 8, which is characterized in that in the step S2, pass through fault diagnosis plate
Card periodically acquires the intermediate operation data of data board, carries out failure predication, and predictive information and intermediate operation data are passed
It send to the diagnosis prediction board.
11. intelligent diagnosing method according to claim 10, which is characterized in that in the step S2, failure predication includes
Following steps:
S21: the fault diagnosis board is according to the time parameter and data of interest in configuration file, periodically acquisition data
The intermediate operation data of board, and be put into memory buffer；
S22: the fault diagnosis board carries out real-time monitoring to the intermediate operation data of acquisition, judges whether data value is in and matches
In the safe range for setting file setting；
S23: when data value exceeds safe range (0, m%), warning message is reported to give failure predication board, and Real-time Feedback is given
Driving assembly in system master board, is adjusted correspondingly, and avoids the deterioration of operation conditions；
S24: when data value exceeds safe range > m%, reporting fault information gives the processing of failure predication board, while directly anti-
Driving assembly in feeding system master control board card, initializes, and wherein m is real number, is set by configuration file.
12. intelligent diagnosing method according to claim 11, which is characterized in that in the step S24, when data value exceeds
When safe range > m%, fault diagnosis board carries out coded treatment to fault type, and passes through serial down trigger, notice diagnosis
Predict board.
13. intelligent diagnosing method according to claim 12, which is characterized in that the step S3 further includes working as diagnosis prediction
After board receives fault message, instruction is sent first and suspends all subsystem movements, and failure is handled, judges event
Hinder mechanism, if run action retries, if allowing, sends " retrying " and orders the subsystem acted to participation, it is dynamic to retry this
Make, if retrying same failure, reports to server end；If not allowing, " system mistake " is sent to server end, waits people
Work intervention.
14. intelligent diagnosing method according to claim 10, which is characterized in that the step S3 the following steps are included:
S31: the diagnosis prediction board is configured according to the sampling time in configuration file, data of interest；
S32: the data of the fixed n servo period sampling primary fault diagnosis board of the diagnosis prediction board are simultaneously stored to data
In library, wherein n is natural number, is arranged by configuration file；
S33: the historical data in data and database that failure predication module in the diagnosis prediction board samples this
Integrated treatment, fitting data change curve are carried out, and finds corresponding rule in database, obtains failure predication information；
S34: failure predication information is fed back to corresponding point by the main hinge board in main system by the diagnosis prediction board
System does corresponding adjustment and operation by the system master board in subsystem.
15. intelligent diagnosing method according to claim 12, which is characterized in that in the step S33, it is described rule with
The form of fault tree saves, i.e., a kind of data trend corresponds to a kind of fault type.
16. intelligent diagnosing method according to claim 12, which is characterized in that in the step S33, if not finding rule
Then, then the failure predication module carries out failure training, and is stored as new rule.
17. intelligent diagnosing method according to claim 12, which is characterized in that in the step S33, pass through least square
Method or the method fitting data change curve for being averaging trend.
18. intelligent diagnosing method according to claim 8, which is characterized in that further include step S4, to the diagnosis prediction
Board or systematic master control board card carry out direct fault location, with the diagnosis effect of checking system.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710900078.2A CN109581995B (en) | 2017-09-28 | 2017-09-28 | Intelligent diagnosis system and method |
| TW107134561A TW201925942A (en) | 2017-09-28 | 2018-09-28 | Intelligent diagnosis system and method |
| PCT/CN2018/108250 WO2019062833A1 (en) | 2017-09-28 | 2018-09-28 | Intelligent diagnosis system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710900078.2A CN109581995B (en) | 2017-09-28 | 2017-09-28 | Intelligent diagnosis system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109581995A true CN109581995A (en) | 2019-04-05 |
| CN109581995B CN109581995B (en) | 2021-09-17 |
Family
ID=65900686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710900078.2A Active CN109581995B (en) | 2017-09-28 | 2017-09-28 | Intelligent diagnosis system and method |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN109581995B (en) |
| TW (1) | TW201925942A (en) |
| WO (1) | WO2019062833A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112838944A (en) * | 2020-07-29 | 2021-05-25 | 中兴通讯股份有限公司 | Diagnosis and management, rule determination and deployment method, distributed device, and medium |
| CN114726712A (en) * | 2022-03-31 | 2022-07-08 | 湖南宇诺辰电子科技有限公司 | Control method and system for reinforcing board card |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113051137B (en) * | 2021-04-22 | 2024-03-26 | 北京计算机技术及应用研究所 | Design method of extensible server remote health management system |
| CN113655773B (en) * | 2021-07-19 | 2023-02-28 | 东风汽车集团股份有限公司 | Vehicle machine system communication serial port pressure testing system and method |
| CN116880400B (en) * | 2023-07-27 | 2023-12-22 | 小黑(广州)智能科技有限公司 | Intelligent production flow management system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393543A (en) * | 2007-09-18 | 2009-03-25 | 西门子公司 | Failure analysis and diagnosis method and system |
| CN102662788A (en) * | 2012-04-28 | 2012-09-12 | 浪潮电子信息产业股份有限公司 | Computer system fault diagnosis decision and processing method |
| CN103197231A (en) * | 2013-04-03 | 2013-07-10 | 湖南大学 | Field programmable gate array (FPGA) device for diagnosing and predicting artificial circuit faults |
| CN103389723A (en) * | 2012-05-11 | 2013-11-13 | 北汽福田汽车股份有限公司 | Motor controller fault detecting system and method |
| CN107054410A (en) * | 2017-04-01 | 2017-08-18 | 广州地铁集团有限公司 | The intelligent diagnosis system and diagnostic method of point machine |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130039692A (en) * | 2011-10-12 | 2013-04-22 | 쥬키 가부시키가이샤 | Electronic component mounting apparatus, electronic component mounting system and electronic component mounting method |
| KR101970794B1 (en) * | 2014-04-07 | 2019-04-22 | 한화정밀기계 주식회사 | Method for controlling operation of component mounter |
-
2017
- 2017-09-28 CN CN201710900078.2A patent/CN109581995B/en active Active
-
2018
- 2018-09-28 WO PCT/CN2018/108250 patent/WO2019062833A1/en active Application Filing
- 2018-09-28 TW TW107134561A patent/TW201925942A/en unknown
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101393543A (en) * | 2007-09-18 | 2009-03-25 | 西门子公司 | Failure analysis and diagnosis method and system |
| CN102662788A (en) * | 2012-04-28 | 2012-09-12 | 浪潮电子信息产业股份有限公司 | Computer system fault diagnosis decision and processing method |
| CN103389723A (en) * | 2012-05-11 | 2013-11-13 | 北汽福田汽车股份有限公司 | Motor controller fault detecting system and method |
| CN103197231A (en) * | 2013-04-03 | 2013-07-10 | 湖南大学 | Field programmable gate array (FPGA) device for diagnosing and predicting artificial circuit faults |
| CN107054410A (en) * | 2017-04-01 | 2017-08-18 | 广州地铁集团有限公司 | The intelligent diagnosis system and diagnostic method of point machine |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112838944A (en) * | 2020-07-29 | 2021-05-25 | 中兴通讯股份有限公司 | Diagnosis and management, rule determination and deployment method, distributed device, and medium |
| WO2022022556A1 (en) * | 2020-07-29 | 2022-02-03 | 中兴通讯股份有限公司 | Fault diagnosis methods, fault diagnosis rule determining method, diagnosis function deployment method, related device, and storage medium |
| CN114726712A (en) * | 2022-03-31 | 2022-07-08 | 湖南宇诺辰电子科技有限公司 | Control method and system for reinforcing board card |
| CN114726712B (en) * | 2022-03-31 | 2022-11-08 | 湖南宇诺辰电子科技有限公司 | Control method and system for reinforcing board card |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109581995B (en) | 2021-09-17 |
| WO2019062833A1 (en) | 2019-04-04 |
| TW201925942A (en) | 2019-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109581995A (en) | A kind of intelligent diagnosis system and method | |
| JP7012871B2 (en) | Devices and methods for controlling the system | |
| US10591886B2 (en) | Control system, control program, and control method for device switching responsive to abnormality detection | |
| JP7017861B2 (en) | Anomaly detection system and anomaly detection method | |
| US8126679B2 (en) | Automatic remote monitoring and diagnostics system | |
| CN108628263B (en) | Control system, control device, computer-readable storage medium, and control method | |
| US7809529B2 (en) | Remote monitoring and diagnostic system | |
| DE112012007224T5 (en) | System and method for operation monitoring and / or diagnosis of a production line of an industrial plant | |
| RU2686257C1 (en) | Method and system for remote identification and prediction of development of emerging defects of objects | |
| CN102637019A (en) | Intelligent integrated fault diagnosis method and device in industrial production process | |
| CN117474357B (en) | Power distribution room operation and maintenance management method and system based on deep learning | |
| JP7119978B2 (en) | controller and program | |
| WO2022030041A1 (en) | Prediction system, information processing device, and information processing program | |
| CN109084971A (en) | A kind of pneumatic control valve method for diagnosing faults based on particle group optimizing | |
| US20220414555A1 (en) | Prediction system, information processing apparatus, and information processing program | |
| GB2607652A (en) | Improved faulty variable identification technique for data-driven fault detection within a process plant | |
| US11339763B2 (en) | Method for windmill farm monitoring | |
| Hong et al. | Intelligent anomaly detection of robot manipulator based on energy consumption auditing | |
| CN109556861A (en) | A kind of bearing real-time fault diagnosis system of case-based reasioning | |
| EP4136515B1 (en) | Method, apparatus and computer program product for monitoring of industrial devices | |
| Filios et al. | Realizing predictive maintenance in production machinery through low-cost IIoT framework and anomaly detection: A case study in a real-world manufacturing environment | |
| Suzuki et al. | An anomaly detection system for advanced maintenance services | |
| RU2818858C1 (en) | Method for diagnosing a complex of on-board equipment of aircraft based on unsupervised machine learning with automatic determination of model training parameters | |
| JP2023100544A (en) | Anomaly diagnosis device, anomaly diagnosis method, and anomaly diagnosis model generation method | |
| CN118408583A (en) | Encoder fault diagnosis method and 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 |