CN109739213A - A kind of failure prediction system and prediction technique - Google Patents
A kind of failure prediction system and prediction technique Download PDFInfo
- Publication number
- CN109739213A CN109739213A CN201910012072.0A CN201910012072A CN109739213A CN 109739213 A CN109739213 A CN 109739213A CN 201910012072 A CN201910012072 A CN 201910012072A CN 109739213 A CN109739213 A CN 109739213A
- Authority
- CN
- China
- Prior art keywords
- inkle loom
- failure
- data
- module
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Looms (AREA)
Abstract
The present invention relates to ribbon equipment technical field, a kind of failure prediction system and prediction technique, including data acquisition module, data analysis module, prediction module, device control module and multi-section inkle loom are referred in particular to；Data acquisition module is for acquiring the operational data of multi-section inkle loom and by the working data transport to data analysis module；Data analysis module is used to carry out analytical calculation to operational data, obtains stream time of the inkle loom next time before failure, and the stream time is transferred to prediction module；Prediction module is used to carry out prediction and matching calculating according to the stream time, predicts the optimal inkle loom for executing current task；Device control module is used to control multi-section inkle loom according to the prediction result of prediction module and is worked or stopped working.A kind of failure prediction system provided by the invention and prediction technique can effectively avoid the fault interrupting in inkle loom operation process, save cost depletions, improved production efficiency.
Description
Technical field
The present invention relates to ribbon equipment technical field, a kind of failure prediction system and prediction technique are referred in particular to.
Background technique
In current inkle loom job shop, the assignment of machine operation task be substantially based on artificial subjective desire into
What row arranged.In the application scenarios opened simultaneously without whole machinery equipments, due to personal use inertia, portion may be such that
The frequency of usage of point machinery equipment (closer to station or the position in aisle) is significantly larger than other machines.In this case, one
The excessive use of part machinery equipment, insufficient utilization of another part machinery equipment result in the machine that same batch is introduced and set
Standby depreciation speed is unbalanced, to reduce the overall utilization rate of equipment.In addition, the machine that frequent uses possesses higher event
Barrier rate, due to lacking the plant maintenance of reasonable task schedule and science, failure occurs also will in the ongoing probability of production
It greatly increases, and production efficiency will be seriously affected by producing ongoing disorderly closedown.
Summary of the invention
The present invention provides a kind of failure prediction system and prediction technique for problem of the prior art, by inkle loom
Equipment carries out data acquisition using time, frequency and failure, predicts what inkle loom broke down next time after being analyzed
Time makes the health status of inkle loom and prejudges in advance, gives birth to reasonably be assigned according to the health status of inkle loom for it
Production task also can effectively avoid the event in inkle loom operation process so that the abrasion depreciation speed of multi-section inkle loom is more balanced
Barrier interrupts.
In order to solve the above-mentioned technical problem, the present invention adopts the following technical scheme: a kind of failure prediction system and prediction
Method, including data acquisition module, data analysis module, prediction module, device control module and multi-section inkle loom；
The data acquisition module is for acquiring the operational data of multi-section inkle loom and by the working data transport to data
Analysis module；
The data analysis module is used to carry out analytical calculation to operational data, obtains company of the inkle loom next time before failure
The continuous working time, and the stream time is transferred to prediction module；
The prediction module is used to carry out prediction and matching calculating according to the stream time, predicts execution as predecessor
The optimal inkle loom of business；
The device control module be used for according to the prediction result of prediction module control multi-section inkle loom carry out work or
It stops working.
Preferably, the data acquisition module includes Wireless data acquisition unit and data terminal, the data acquisition
The time data of switching on and shutting down of the device for acquiring inkle loom in real time are simultaneously transferred to data analysis module, and the data terminal is for remembering
It records the failure logging of inkle loom and is transferred to data analysis module.
Preferably, the failure prediction system further includes for showing data acquisition module, data analysis module and pre-
Survey the human-machine operation module of the operational data of module.
A kind of failure prediction method, includes the following steps
A. the operational data for passing through data collecting module collected inkle loom, when operational data includes the switching on and shutting down of inkle loom
Between, the work total duration of inkle loom and failure logging, failure logging include the title of trouble unit and the time of origin of failure；
B. initial neural network model is established, using the operational data of collected inkle loom as input data, to initial mind
It is trained through network model, obtains the target nerve network model that can carry out failure predication；
C. target nerve network model obtained in the operational data input step B by collected inkle loom, predicts
All maximum continuous operating time of the idle inkle loom before break down next time；
D. the maximum continuous work duration of acquisition is compared with the use duration of multi-section inkle loom and is matched, obtained and complete
The most suitable inkle loom of current task.
Preferably, in step A, multi-section inkle loom is numbered, the number of i-th inkle loom is set as Ri, inkle loom Ri's
The time of jth time switching on and shutting down is expressed as Tij_sAnd Tij_e, the representation for fault that k-th of component occurs is Fk, opened collected
Unused time Tij_sAnd Tij_eAnd the failure F occurredkNeural network model is conveyed on by way of wireless data transmission.
Preferably, by inkle loom RiOperational data be divided into 5 dimensions comprising work total duration Ti, total start-stop time
Numi_se, total failare times N umi_F, use duration T between current failure and last similar failureikAnd equipment is in TikPhase
Between start-stop time Numik_se, whereinInkle loom R is obtained by 5 dimensionsiOccur
Failure FkHistory vectors information Vik=[TiNumi_seNumi_FTikNumik_se], with history vectors information VikAs neural network
Input data, with inkle loom RiLast time break down before continuous operating time Tik_last_segFor initial neural network
Training objective, initial neural network is trained.
Preferably, to history vectors information VikIt is normalized, obtains normalization history vectors informationWherein N is general headquarters' number of inkle loom,
By normalization history vectors information Vik_stdInitial neural network model is inputted, trouble unit F is predictedkEvent occurs next time
Maximum continuous work duration t before barrierik_next。
Preferably, in step D, task to be allocated and inkle loom RiBetween matching factor formula are as follows:
Wherein, TiFor inkle loom RiWork total duration, Ti_nextFor inkle loom RiMaximum before failure next time occurs
Continuous work duration, Ti_taskFor the operating time of inkle loom needed for current task,Maximum work in multi-section inkle loom
Total duration；Work as PiBetween [0,1] when value, make PiObtain the inkle loom R of minimum valueiThe optimal allocation of as current task is knitted
Band machine
Beneficial effects of the present invention:
A kind of failure prediction system provided by the invention and prediction technique, by acquiring the operational data of inkle loom, packet
Switching on and shutting down time, the work total duration of inkle loom and failure logging are included, and data are analyzed, establishes neural network mould
Type is trained neural network model, can predict the time that inkle loom breaks down next time, be good for inkle loom
Health situation is made to be prejudged in advance, to reasonably assign production task according to the health status of inkle loom for it.The failure predication
System and prediction technique make the abrasion depreciation speed of multi-section inkle loom more balanced, can effectively avoid inkle loom operation process
In fault interrupting improve production efficiency to save the cost depletions of equipment；In addition, the failure prediction system and
Prediction technique has highlighted the importance of creation data statistic record, also provides reference for other production fields.
Detailed description of the invention
Fig. 1 is failure prediction system schematic diagram of the invention.
Fig. 2 is failure prediction method flow chart of the invention.
Include: in appended drawing reference of the Fig. 1 into Fig. 2
1- inkle loom, 2- data acquisition module, 21- data collector, 22- data terminal, 3- data analysis module, 4- are pre-
Survey module, 5- device control module, 6- human-machine operation module, 61- display, 62- operation keyboard, 63- operation mouse.
Specific embodiment
For the ease of the understanding of those skilled in the art, the present invention is made further below with reference to embodiment and attached drawing
Bright, the content that embodiment refers to not is limitation of the invention.The present invention is described in detail below in conjunction with attached drawing.
A kind of failure prediction system provided in this embodiment, such as Fig. 1, including data acquisition module 2, data analysis module 3,
Prediction module 4, device control module 5 and multi-section inkle loom 1；The data acquisition module 2 is for acquiring multi-section inkle loom 1
Operational data and by the working data transport to data analysis module 3；The data analysis module 3 is used to carry out operational data
Analytical calculation obtains stream time of the inkle loom 1 next time before failure, and the stream time is transferred to prediction mould
Block 4；The prediction module 4 is used to carry out prediction and matching calculating according to the stream time, predicts execution current task
Optimal inkle loom 1；The device control module 5 be used for according to the prediction result of prediction module 4 control multi-section inkle loom 1 into
Row work stops working.
Specifically, the working principle of failure prediction system are as follows: when multi-section inkle loom 1 is started to work, data acquisition module 2 is real
When to multi-section inkle loom 1 carry out operational data acquisition, operational data includes the work of the switching on and shutting down time of inkle loom 1, inkle loom 1
Make total duration and failure logging, failure logging includes the title of trouble unit and the time of origin of failure, data acquisition module 2
Collected data are transmitted through the network to data analysis module 3 again, the storage of data is carried out by data analysis module 3, and
Calculating analysis is carried out, neural network model is established, neural network model is trained with the operational data of inkle loom 1, thus
Go out the continuous work duration idle inkle loom 1 breaks down next time before by trained Neural Network model predictive, in advance
Module 4 is surveyed according to the continuous work duration, the operating time in conjunction with needed for current task carries out the matching of inkle loom 1, predicts
It is most suitable for completing the inkle loom 1 of current task, device control module 5 is knitted according to what prediction module 4 was predicted as a result, controlling accordingly
Current production task is completed with machine 1；The failure prediction system can predict the time that inkle loom 1 breaks down next time, right
The health status of inkle loom 1 is made to be prejudged in advance, is appointed to reasonably assign production according to the health status of inkle loom 1 for it
Business, so that the abrasion depreciation speed of multi-section inkle loom 1 is more balanced, can effectively avoid in the failure in 1 operation process of inkle loom
It is disconnected, to save the cost depletions of equipment, improve production efficiency.
A kind of failure prediction system provided in this embodiment, such as Fig. 1, the data acquisition module 2 are adopted including wireless data
Storage 21 and data terminal 22, the time data of switching on and shutting down of the Wireless data acquisition unit 21 for acquiring inkle loom 1 in real time
And it is transferred to data analysis module 3, the data terminal 22 is used to record the failure logging of multi-section inkle loom 1 and is transferred to data
Analysis module 3.
Specifically, by the data acquisition of Wireless data acquisition unit 21 and the data record of data terminal 22, by wireless
Network sends data to other appliance arrangements, so that the acquisition process of data becomes simpler, it is preferred that data terminal 22 is
Handheld data terminal 22 is observed data record case at any time convenient for staff and is handled.
A kind of failure prediction system provided in this embodiment, such as Fig. 1, the failure prediction system further include for showing number
According to the human-machine operation module 6 of the operational data of acquisition module 2, data analysis module 3 and prediction module 4.
Specifically, human-machine operation module 6 includes display 61, operation keyboard 62 and operation mouse 63, failure prediction system
When work, data transmission scenarios are shown by display 61, system working condition is understood convenient for staff, needs
When, staff can carry out manual intervention to failure prediction system by operation keyboard 62 and operation mouse 63.
A kind of failure prediction method provided in this embodiment, such as Fig. 2,
A. the operational data for passing through data collecting module collected inkle loom 1, when operational data includes the switching on and shutting down of inkle loom
Between, the work total duration of inkle loom 1 and failure logging, failure logging include trouble unit title and failure generation when
Between；
B. initial neural network model is established, using the operational data of collected inkle loom 1 as input data, to initial mind
It is trained through network model, obtains the target nerve network model that can carry out failure predication；
C. by target nerve network model obtained in the operational data input step B of newest collected inkle loom 1, in advance
Measure maximum continuous operating time of all idle inkle loom 1 before break down next time；
D. the maximum continuous work duration of acquisition is compared with the use duration of multi-section inkle loom 1 and is matched, obtained
At the most suitable inkle loom 1 of current task.
Specifically, by carrying out data acquisition using time, frequency and failure to 1 equipment of inkle loom, after being analyzed
The time that inkle loom breaks down next time is predicted, the health status of inkle loom 1 is made and is prejudged in advance, thus according to ribbon
The health status of machine reasonably also can for its assignment production task so that the abrasion depreciation speed of multi-section inkle loom 1 is more balanced
Effectively avoid the fault interrupting in 1 operation process of inkle loom.
A kind of failure prediction method provided in this embodiment, is numbered multi-section inkle loom 1, the volume of i-th inkle loom 1
Number it is set as Ri, inkle loom RiTime of jth time switching on and shutting down be expressed as Tij_sAnd Tij_e, the bug list of k-th of component generation
It is shown as Fk, by collected switching on and shutting down time Tij_sAnd Tij_eAnd the failure F occurredkOn by way of wireless data transmission
It is conveyed into neural network model.
A kind of failure prediction method provided in this embodiment, by inkle loom RiOperational data be divided into 5 dimensions comprising
Work total duration Ti, total start-stop time Numi_se, total failare times N umi_F, make between current failure and last similar failure
With duration TikAnd equipment is in TikThe start-stop time Num of periodik_se, whereintikm=Tij_e-Tij_s, pass through described 5
A dimension obtains inkle loom RiBreak down FkHistory vectors information Vik=[TiNumi_seNumi_FTikNumik_se], with history
Vector information VikAs the input data of neural network, with inkle loom RiLast time break down before continuous operating time
Tik_last_segFor the training objective of initial neural network, initial neural network is trained.
A kind of failure prediction method provided in this embodiment, to history vectors information VikIt is normalized, is returned
One changes history vectors informationWherein N is
General headquarters' number of inkle loom, by normalization history vectors information Vik_stdInitial neural network model is inputted, trouble unit is predicted
FkMaximum continuous work duration t break down next time beforeik_next。
A kind of failure prediction method provided in this embodiment, task to be allocated and inkle loom RiBetween matching factor formula
Are as follows:
Wherein, TiFor inkle loom RiWork total duration, Ti_nextFor inkle loom RiMaximum before failure next time occurs
Continuous work duration, Ti_taskFor the operating time of inkle loom needed for current task,Maximum work in multi-section inkle loom 1
Make total duration；Work as PiBetween [0,1] when value, make PiObtain the inkle loom R of minimum valueiThe as optimal allocation of current task
Inkle loom
Specifically, the principle of this failure prediction method are as follows:
(1) data acquire
For the training of later period neural network model, need to unite to the related data in 1 operational process of multi-section inkle loom
Meter, related data here mainly includes the work total duration and failure logging of 1 switching on and shutting down time of inkle loom, inkle loom 1, therefore
Barrier record includes the title of trouble unit and the time of origin of failure.It can be real-time by the data acquisition module 2 on inkle loom 1
The switching on and shutting down time data of machine are acquired, and collected runing time data are uploaded to by data point by wireless data transmission
It analyses in module 3.Manual operation can be carried out by data terminal 22 for the record of mechanical disorder, for the correlation of equipment fault
Component is numbered, can be by the number of trouble unit and the time broken down (with working as inkle loom 1 by data terminal 22
Secondary downtime is identical) it is sent in the database of background server.Assuming that the sum of whole inkle looms is N, i-th inkle loom
Number is Ri, wherein 1≤i≤N, the time of i-th inkle loom jth time switching on and shutting down is respectively to be expressed as Tij_sAnd Tij_e, wherein 1
≤ i≤J, J are history switching on and shutting down total degree；Every inkle loom includes K trouble unit, will according to the component difference that failure occurs
Fault type is divided into K class, wherein the representation for fault that k-th of component occurs is Fk, wherein 1≤k≤K, number RiInkle loom hair
Raw failure FkTime be Tij_eIn a time point.
(2) neural net model establishing
According to the collected data, the operational data record of available every inkle loom, sends out each trouble unit
5 dimensions of equipment history run information point when raw failure are described, each failure portion of every equipment available in this way
The historical information vector V of partik, corresponding label is then the continuous use duration that certain failure occurs for last time
Tik_last_seg。
Three dimensions of the overall operation situation of ribbon machine equipment point are described first, are work total duration T respectivelyi, always
Start-stop time Numi_se, total failare times N umi_F, k-th of trouble unit broken down times N umik_F, wherein
tikm=Tij_e-Tij_s；
Then it is directed to specific fault type, 2 dimensions is further divided into describe, is that current failure is similar with the last time respectively
Time T is used between failureikWith the start-stop time Num of equipment during this periodik_se, it should be noted that do not include among these
Stream time section T before current failure generationik_last_seg, therefore can be by Tik_last_segAs model training prediction
As a result.
Comprehensively consider this historical information vector Vik:
Vik=[TiNumi_seNumi_FTikNumik_se]；
Although its each dimension can describe the historgraphic data recording of failure generation, the information of each dimension is not right
Claim, namely there are Sparse Problems, it is necessary into once doing normalized to it.Add in each dimension multiplied by an overall situation
Weight factor can obtain a normalization historical information vector:
There is normalization historical information vector Vik_stdContinuous operating time before breaking down with equipment last time
Tik_last_segLater, this is a typical Supervised machine learning problem, and wherein the label of supervised learning is one continuous
Value.It is therefore possible to use Neural Network Based Nonlinear regression model models this problem.Neural network model it is defeated
Entering data is normalization historical information vector Vik_std, corresponding label is Tik_last_seg, by the data that have acquired to nerve
Network model is trained, an available convergent machine learning model.Newest normalization historical information vector is defeated
Enter trained neural network model, so that it may predict maximum of the trouble unit currently investigated before break down next time
Stream time length tik_next。
(3) optimizing scheduling
Every ribbon machine equipment includes K trouble unit, and in each booting, each trouble unit can be by
Trained model prediction obtains a continuous work maximum time tik_next, shortest continuous work duration is whole and sets
Standby maximum continuous work duration Ti_next, i.e. Tik_last_seg=[Ti_next, tik_next]。
Assignment for each inkle loom job task, in addition to the maximum continuous work duration T of equipmenti_nextIt wants and task
Except flux matched, the factor equally to be considered is current inkle loom RiUse total duration Ti, it should preferential selection uses total duration
Shorter equipment.
Task amount can describe T with the inkle loom activity durationi_task, can be obtained by task to be allocated and preferential in this way
The inkle loom R of distribution taskiBetween matching factor:
Wherein TiFor inkle loom RiWork total duration,It is then maximum work total duration in all inkle looms.When
PiBetween [0,1] when value, work as PiBetween [0,1] when value, make PiObtain the inkle loom R of minimum valueiAs current task
Optimal allocation inkle loom Ri_best:
The failure predication of inkle loom is completed as a result,.
The above is only present pre-ferred embodiments, is not intended to limit the present invention in any form, although
The present invention is disclosed as above with preferred embodiment, and however, it is not intended to limit the invention, any person skilled in the art,
It does not depart within the scope of technical solution of the present invention, when the technology contents using the disclosure above make a little change or are modified to equivalent change
The equivalent embodiment of change, but without departing from the technical solutions of the present invention, technology refers to above embodiments according to the present invention
Made any simple modification, equivalent change and modification, belong in the range of technical solution of the present invention.
Claims (8)
1. a kind of failure prediction system, it is characterised in that: including data acquisition module, data analysis module, prediction module, equipment
Control module and multi-section inkle loom；
The data acquisition module is used to acquire the operational data of multi-section inkle loom and analyzes the working data transport to data
Module；
The data analysis module is used to carry out analytical calculation to operational data, obtains continuous work of the inkle loom next time before failure
Make the time, and the stream time is transferred to prediction module；
The prediction module is used to carry out prediction and matching calculating according to stream time, predicts and executes the best of current task
Inkle loom；
The device control module is used to control multi-section inkle loom according to the prediction result of prediction module and is worked or stopped
Work.
2. a kind of failure prediction system according to claim 1, it is characterised in that: the data acquisition module includes no line number
According to collector and data terminal, the time data of switching on and shutting down of the Wireless data acquisition unit for acquiring inkle loom in real time are simultaneously passed
Defeated to arrive data analysis module, the data terminal is used to record the failure logging of inkle loom and is transferred to data analysis module.
3. a kind of failure prediction system according to claim 1, it is characterised in that: the failure prediction system further includes being used for
Show the human-machine operation module of the operational data of data acquisition module, data analysis module and prediction module.
4. a kind of failure prediction method based on failure prediction system described in claim 1, it is characterised in that: including following step
Suddenly
A. pass through the operational data of data collecting module collected inkle loom, operational data includes the switching on and shutting down time of inkle loom, knits
Work total duration and failure logging with machine, failure logging include the title of trouble unit and the time of origin of failure；
B. initial neural network model is established, using the operational data of collected inkle loom as input data, to initial nerve net
Network model is trained, and obtains the target nerve network model that can carry out failure predication；
C. target nerve network model obtained in the operational data input step B by collected inkle loom, predicts whole
Maximum continuous operating time of the idle inkle loom before break down next time；
D. the maximum continuous work duration of acquisition is compared with the use duration of multi-section inkle loom and is matched, obtained and complete currently
The most suitable inkle loom of task.
5. a kind of failure prediction method according to claim 4, it is characterised in that: in step A, compiled to multi-section inkle loom
Number, the number of i-th inkle loom is set as Ri, inkle loom RiTime of jth time switching on and shutting down be expressed as Tij_sAnd Tij_e, kth
The representation for fault that a component occurs is Fk, by collected switching on and shutting down time Tij_sAnd Tij_eAnd the failure F occurredkPass through nothing
Line number is according to being conveyed into neural network model in the mode of transmission.
6. a kind of failure prediction method according to claim 5, it is characterised in that: by inkle loom RiOperational data be divided into 5
Dimension comprising work total duration Ti, total start-stop time Numi_se, total failare times N umi_F, current failure it is similar with the last time
Use duration T between failureikAnd equipment is in TikThe start-stop time Num of periodik_se, whereintikm=Tij_e-
Tij_s, inkle loom R is obtained by 5 dimensionsiBreak down FkHistory vectors information Vik=[TiNumi_seNumi_ FTikNumik_se], with history vectors information VikAs the input data of neural network, with inkle loom RiLast time occur therefore
Continuous operating time T before barrierik_last_segFor the training objective of initial neural network, initial neural network is trained.
7. a kind of failure prediction method according to claim 6, it is characterised in that: to history vectors information VikIt is normalized
Processing obtains normalization history vectors informationWherein N is general headquarters' number of inkle loom,
By normalization history vectors information Vik_stdInitial neural network model is inputted, trouble unit F is predictedkEvent occurs next time
Maximum continuous work duration t before barrierik_next。
8. a kind of failure prediction method according to claim 7, it is characterised in that: in step D, task to be allocated and inkle loom
RiBetween matching factor formula are as follows:
Wherein, TiFor inkle loom RiWork total duration, Ti_nextFor inkle loom RiMaximum before failure next time occurs is continuous
Operating time, Ti_taskFor the operating time of inkle loom needed for current task,When maximum work is total in multi-section inkle loom
It is long；Work as PiBetween [0,1] when value, make PiObtain the inkle loom R of minimum valueiThe as optimal allocation inkle loom of current task
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012072.0A CN109739213A (en) | 2019-01-07 | 2019-01-07 | A kind of failure prediction system and prediction technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012072.0A CN109739213A (en) | 2019-01-07 | 2019-01-07 | A kind of failure prediction system and prediction technique |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109739213A true CN109739213A (en) | 2019-05-10 |
Family
ID=66363643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910012072.0A Pending CN109739213A (en) | 2019-01-07 | 2019-01-07 | A kind of failure prediction system and prediction technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109739213A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210105A (en) * | 2019-12-17 | 2020-05-29 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Continuous operation performance verification device, computer equipment and storage medium |
| CN111553476A (en) * | 2020-04-30 | 2020-08-18 | 汪科道 | Neural network training method and device based on memory score and storage medium |
| CN112232370A (en) * | 2020-09-16 | 2021-01-15 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Fault analysis and prediction method for engine |
| CN112903001A (en) * | 2019-12-03 | 2021-06-04 | 财团法人纺织产业综合研究所 | Operation method of fabric setting machine |
| CN112988843A (en) * | 2021-03-26 | 2021-06-18 | 桂林电子科技大学 | SMT chip mounter fault management and diagnosis system based on SQL Server database |
| CN114461704A (en) * | 2022-04-14 | 2022-05-10 | 浙江理工大学 | Method and system for predicting loom availability based on loom productivity |
| CN116260235A (en) * | 2023-03-11 | 2023-06-13 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105825271A (en) * | 2016-03-21 | 2016-08-03 | 南京邮电大学 | Satellite fault diagnosis and prediction method based on evidential reasoning (ER) |
| CN106951984A (en) * | 2017-02-28 | 2017-07-14 | 深圳市华傲数据技术有限公司 | A kind of dynamic analyzing and predicting method of system health degree and device |
| US20170249551A1 (en) * | 2016-02-29 | 2017-08-31 | International Business Machines Corporation | Developing an accurate dispersed storage network memory performance model through training |
| CN107679649A (en) * | 2017-09-13 | 2018-02-09 | 珠海格力电器股份有限公司 | A kind of failure prediction method of electrical equipment, device, storage medium and electrical equipment |
| CN109034413A (en) * | 2018-07-11 | 2018-12-18 | 广东人励智能工程有限公司 | Intelligence manufacture equipment fault prediction technique and system based on neural network model |
-
2019
- 2019-01-07 CN CN201910012072.0A patent/CN109739213A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170249551A1 (en) * | 2016-02-29 | 2017-08-31 | International Business Machines Corporation | Developing an accurate dispersed storage network memory performance model through training |
| CN105825271A (en) * | 2016-03-21 | 2016-08-03 | 南京邮电大学 | Satellite fault diagnosis and prediction method based on evidential reasoning (ER) |
| CN106951984A (en) * | 2017-02-28 | 2017-07-14 | 深圳市华傲数据技术有限公司 | A kind of dynamic analyzing and predicting method of system health degree and device |
| CN107679649A (en) * | 2017-09-13 | 2018-02-09 | 珠海格力电器股份有限公司 | A kind of failure prediction method of electrical equipment, device, storage medium and electrical equipment |
| CN109034413A (en) * | 2018-07-11 | 2018-12-18 | 广东人励智能工程有限公司 | Intelligence manufacture equipment fault prediction technique and system based on neural network model |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112903001A (en) * | 2019-12-03 | 2021-06-04 | 财团法人纺织产业综合研究所 | Operation method of fabric setting machine |
| CN111210105A (en) * | 2019-12-17 | 2020-05-29 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Continuous operation performance verification device, computer equipment and storage medium |
| CN111210105B (en) * | 2019-12-17 | 2023-11-07 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Continuous operation performance verification, apparatus, computer device, and storage medium |
| CN111553476A (en) * | 2020-04-30 | 2020-08-18 | 汪科道 | Neural network training method and device based on memory score and storage medium |
| CN111553476B (en) * | 2020-04-30 | 2023-12-01 | 上海个元科技有限公司 | Neural network training method, device and storage medium based on memory score |
| CN112232370A (en) * | 2020-09-16 | 2021-01-15 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Fault analysis and prediction method for engine |
| CN112988843A (en) * | 2021-03-26 | 2021-06-18 | 桂林电子科技大学 | SMT chip mounter fault management and diagnosis system based on SQL Server database |
| CN114461704A (en) * | 2022-04-14 | 2022-05-10 | 浙江理工大学 | Method and system for predicting loom availability based on loom productivity |
| CN114461704B (en) * | 2022-04-14 | 2022-06-24 | 浙江理工大学 | Method and system for predicting loom availability based on loom productivity |
| CN116260235A (en) * | 2023-03-11 | 2023-06-13 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
| CN116260235B (en) * | 2023-03-11 | 2023-08-22 | 傲视恒安科技（北京）有限公司 | Power supply switching method and device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109739213A (en) | A kind of failure prediction system and prediction technique | |
| CN106209482A (en) | A kind of data center monitoring method and system | |
| JP6298214B2 (en) | System and method for maximizing the expected utility of signal injection test patterns within a utility grid | |
| CN109684053A (en) | The method for scheduling task and system of big data | |
| Xia et al. | Modeling and optimizing maintenance schedule for energy systems subject to degradation | |
| CN109559027A (en) | Planning scheduling system applied to iron and steel enterprise's high-speed tool steel smelting process | |
| CN104102212A (en) | Dispatching method, apparatus and system for gas and steam system in iron and steel enterprises | |
| Xia et al. | Optimal selective maintenance scheduling for series–parallel systems based on energy efficiency optimization | |
| CN107037794B (en) | A kind of nonlinear restriction scheduling optimization system based on electroplating assembly line | |
| CN117709617A (en) | MES-based intelligent scheduling system for production workshop | |
| Liu et al. | Dynamic job-shop scheduling problems using graph neural network and deep reinforcement learning | |
| Borangiu et al. | Smart manufacturing control with cloud-embedded digital twins | |
| Zhao et al. | A cooperative whale optimization algorithm for energy-efficient scheduling of the distributed blocking flow-shop with sequence-dependent setup time | |
| Liu et al. | Scheduling optimization of task allocation in integrated manufacturing system based on task decomposition | |
| Brochado et al. | A data-driven model with minimal information for bottleneck detection-application at Bosch thermotechnology | |
| Shirazi et al. | Intelligent decision support system for the adaptive control of a flexible manufacturing system with machine and tool flexibility | |
| Zhang et al. | Counterfactual-attention multi-agent reinforcement learning for joint condition-based maintenance and production scheduling | |
| CN101776909A (en) | Gas supply and demand dynamic prediction system for steel enterprises and method thereof | |
| CN117114381A (en) | Dynamic optimization configuration method and system for intelligent manufacturing service chain | |
| Liu et al. | Manufacturing Process Optimization in the Process Industry | |
| CN116090702A (en) | ERP data intelligent supervision system and method based on Internet of things | |
| CN113407945A (en) | Man-machine cooperation based large-scale Fuzzing optimization system and method | |
| Xu et al. | Parallel machine scheduling with due date-to-deadline window, order sharing and time value of money | |
| Fu et al. | A sewing line planning model based on RFID technology | |
| Yadav et al. | Probabilistic Modeling of Workload Patterns for Capacity Planning in Data Center Environments |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |