CN109087008A - A kind of equipment fault early-warning method and device - Google Patents
A kind of equipment fault early-warning method and device Download PDFInfo
- Publication number
- CN109087008A CN109087008A CN201810890568.3A CN201810890568A CN109087008A CN 109087008 A CN109087008 A CN 109087008A CN 201810890568 A CN201810890568 A CN 201810890568A CN 109087008 A CN109087008 A CN 109087008A
- Authority
- CN
- China
- Prior art keywords
- numerical value
- variation tendency
- time point
- secular variation
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Analysis (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Marketing (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Life Sciences & Earth Sciences (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The present invention provides a kind of equipment fault early-warning method and devices, comprising: corresponding at least two historical values of at least one operating index are stored in advance, wherein each historical values corresponding time point;Determine at least one corresponding index to be detected of measurement equipment to be checked;The time series for corresponding to the index to be detected is generated according to the historical values of storage and the historical values corresponding time point for each index to be detected;At least two Secular Variation Tendency numerical value are decomposited from the time series;Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;According to the matched curve and preset early warning value, the index to be detected corresponding fault pre-alarming time point is determined;Before the fault pre-alarming time point, the fault pre-alarming time point is warned.This programme can formulate the more reasonable repair time.
Description
Technical field
The present invention relates to device management techniques field, in particular to a kind of equipment fault early-warning method and device.
Background technique
Equipment in enterprise belongs to weight assets, such as elevator, lathe, generating set etc., and investment is big, and maintenance is got up
Bothersome effort.Once failure normally results in large-scale accident, the property of life security and enterprise to personnel causes great
Loss.
With the use of equipment, critical component can continuous aging and abrasion, this variation would necessarily affect equipment and exists
Indices when working condition, such as motor temperature under working condition, vibration frequency etc..When a certain index is more than certain numerical value
When, then it may cause device fails.It is main by manually according to setting at present in order to reduce the probability of device fails
Time between overhauls(TBO) indices are detected, can be to the maintenance that equipment is responded when Indexes Abnormality.
But the time between overhauls(TBO) setting it is too short when, then will increase enterprise to the cost of overhaul of equipment.And the time between overhauls(TBO) is arranged
It is too long, may service personnel also not carry out inspection horological device broken down.
As can be seen from the above description, the repair time that prior art detection indices are formulated is unreasonable.
Summary of the invention
The embodiment of the invention provides a kind of equipment fault early-warning method and device, more reasonably maintenance week can be formulated
Phase.
In a first aspect, the embodiment of the invention provides a kind of equipment fault early-warning method,
Corresponding at least two historical values of at least one operating index are stored in advance, wherein each historical values
A corresponding time point;
Include:
Determine at least one corresponding index to be detected of measurement equipment to be checked;
It is corresponding described according to the historical values of storage and the historical values for each index to be detected
Time point generates the time series for corresponding to the index to be detected;
At least two Secular Variation Tendency numerical value are decomposited from the time series;
Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;
According to the matched curve and preset early warning value, the index to be detected corresponding fault pre-alarming time is determined
Point;
Before the fault pre-alarming time point, the fault pre-alarming time point is warned.
Preferably, described according to the historical values of storage and the historical values corresponding time point, it generates
Time series corresponding to the index to be detected, comprising:
From the historical values of storage, determining at least two originals numerical value corresponding with the index to be detected,
In, the corresponding operating index of the original numerical value is identical as the index to be detected;
According to the corresponding each former numerical value of time point arrangement of the original numerical value, generate corresponding to described to be detected
The time series of index.
It is preferably, described that at least two Secular Variation Tendency numerical value are decomposited from the time series, comprising:
According to preset smoothness period, the time series is divided at least two subsequences, wherein the smooth week
Phase is at least one described time point;
For each subsequence, by weighting local regression LOESS to the former numerical value in the subsequence into
Row smoothing processing, the former numerical value after obtaining smoothing processing;
According to preset sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, is slided
Average value;
Calculating is filtered to the sliding average, obtains low pass magnitude;
The former numerical value after smoothing processing is subtracted each other with the low pass magnitude, obtains smoothness period value;
The smoothness period value is removed from the subsequence, cycle value is removed in acquisition;
It goes cycle value to be smoothed to described in acquisition by the LOESS, obtains Secular Variation Tendency numerical value.
Preferably, the Secular Variation Tendency numerical value decomposited at described Dui carries out linear regression fit, obtains fitting
Before curve, further comprise:
Determine the trend numerical value quantity of the Secular Variation Tendency numerical value;
The Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity;
Obtain the average time point at the Secular Variation Tendency numerical value corresponding time point;
The described pair of Secular Variation Tendency numerical value decomposited carries out linear regression fit, obtains matched curve, comprising:
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biCharacterize i-th of the Secular Variation Tendency numerical value pair
The time point answered, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,It characterizes described flat
Equal time point, x characterize any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
Preferably, the Secular Variation Tendency that the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity
Average value, comprising:
To the Secular Variation Tendency numerical value sum acquisition trend numerical value summation, and by the trend numerical value summation with it is described
Trend numerical value quantity is divided by, and the Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained;
The average time point for obtaining the Secular Variation Tendency numerical value corresponding time point, comprising:
Sequentially in time, when the last one described Secular Variation Tendency numerical value of the time sequencing is corresponding described
Between point, the time point corresponding with first of the time sequencing Secular Variation Tendency numerical value subtracts each other, and obtains trend
Total duration;
The trend total duration and the trend numerical value quantity are divided by, average time point is obtained.
Preferably, described according to the matched curve and preset early warning value, determine the corresponding event of the index to be detected
Hinder pre-warning time point, comprising:
It is substituted into preset early warning value as the parameter in the matched curve, it is described corresponding to obtain the parameter
It anticipates time point;
Using the corresponding any point-in-time of the parameter as fault pre-alarming time point.
Second aspect, the embodiment of the invention provides a kind of equipment fault early-warning devices, comprising:
Numeric storage unit, for corresponding at least two historical values of at least one operating index to be stored in advance, wherein
The each historical values corresponding time point;
Information management unit, for determining at least one corresponding index to be detected of measurement equipment to be checked;For each described
Index to be detected, according to the historical values of numeric storage unit storage and the historical values corresponding time
Point generates the time series for corresponding to the index to be detected;
Numeric processing unit decomposites at least two length for generating from the information management unit in the time series
Phase variation tendency numerical value;Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;Root
According to the matched curve and preset early warning value, the index to be detected corresponding fault pre-alarming time point is determined;
Alarm unit, for the numeric processing unit determine the fault pre-alarming time point before, described in warning
Fault pre-alarming time point.
Preferably, the information management unit, for determining and the finger to be detected from the historical values of storage
Mark corresponding at least two former numerical value, wherein the corresponding operating index of the original numerical value and the index phase to be detected
Together;According to the corresponding each former numerical value of time point arrangement of the original numerical value, generates and correspond to the index to be detected
Time series.
Preferably, the numeric processing unit, for according to preset smoothness period, by the time series be divided into
Few two subsequences, wherein the smoothness period is at least one described time point;For each subsequence, by adding
Power local regression LOESS is smoothed the former numerical value in the subsequence, the original after obtaining smoothing processing
Numerical value;According to preset sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, obtains sliding average
Value;Calculating is filtered to the sliding average, obtains low pass magnitude;By after smoothing processing the former numerical value with it is described low
Amount of flux is subtracted each other, and smoothness period value is obtained;The smoothness period value is removed from the subsequence, cycle value is removed in acquisition;It is logical
Cross the LOESS goes cycle value to be smoothed to described in acquisition, obtains Secular Variation Tendency numerical value.
Preferably, the numeric processing unit is further used for determining the trend numerical value of the Secular Variation Tendency numerical value
Quantity;The Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity;Obtain institute
State the average time point at the Secular Variation Tendency numerical value corresponding time point;
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biCharacterize i-th of the Secular Variation Tendency numerical value pair
The time point answered, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,Described in characterization
Average time point, x characterize any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
Preferably, the numeric processing unit, for total to Secular Variation Tendency numerical value summation acquisition trend numerical value
With, and the trend numerical value summation and the trend numerical value quantity are divided by, obtain the long-term of the Secular Variation Tendency numerical value
Variation tendency average value;Sequentially in time, the last one described Secular Variation Tendency numerical value of the time sequencing is corresponding
The time point, the time point corresponding with first of the time sequencing Secular Variation Tendency numerical value subtracts each other,
Acquisition trend total duration;The trend total duration and the trend numerical value quantity are divided by, average time point is obtained.
Preferably, the numeric processing unit, it is bent for substituting into the fitting using preset early warning value as the parameter
In line, the corresponding any point-in-time of the parameter is obtained;Using the corresponding any point-in-time of the parameter as failure
Pre-warning time point.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.
Fig. 1 is a kind of flow chart for equipment fault early-warning method that one embodiment of the invention provides;
Fig. 2 is the flow chart for another equipment fault early-warning method that one embodiment of the invention provides;
Fig. 3 is a kind of structural schematic diagram for equipment fault early-warning device that one embodiment of the invention provides.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments, based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
As shown in Figure 1, the embodiment of the invention provides a kind of equipment fault early-warning methods, comprising:
Step 101: corresponding at least two historical values of at least one operating index are stored in advance, wherein each described
Historical values correspond to a time point;
Step 102: determining at least one corresponding index to be detected of measurement equipment to be checked;
Step 103: each index to be detected is directed to, according to the historical values and the historical values pair of storage
The time point answered generates the time series for corresponding to the index to be detected;
Step 104: at least two Secular Variation Tendency numerical value are decomposited from the time series;
Step 105: linear regression fit being carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;
Step 106: according to the matched curve and preset early warning value, determining that the corresponding failure of the index to be detected is pre-
Alert time point;
Step 107: before the fault pre-alarming time point, warning the fault pre-alarming time point.
In embodiments of the present invention, it according to the historical values of measurement equipment to be checked corresponding index to be detected and storage, can give birth to
At the time sequencing for corresponding to index to be detected, by the way that the quasi- of Secular Variation Tendency can be obtained to time series decomposition analysis
Fault pre-alarming time point can be determined in conjunction with matched curve and preset early warning value by closing curve, before fault pre-alarming time point
Early warning is carried out to warning failure pre-warning time point, time enough can be provided and take steps to prevent failure for service personnel
Occur, reduces enterprises' loss, and can resolutely increase enterprise's cost of overhaul to avoid the time between overhauls(TBO).
In an embodiment of the present invention, described corresponding described according to the historical values of storage and the historical values
Time point generates the time series for corresponding to the index to be detected, comprising:
From the historical values of storage, determining at least two originals numerical value corresponding with the index to be detected,
In, the corresponding operating index of the original numerical value is identical as the index to be detected;
According to the corresponding each former numerical value of time point arrangement of the original numerical value, generate corresponding to described to be detected
The time series of index.
In embodiments of the present invention, time series is made of multiple former numerical value according to the arrangement of corresponding time point, and
Former numerical value is then that determination is corresponding with index to be detected from the historical values of storage.According to historical values to future trend and
Data are predicted, and the suggestion to service personnel's repair time is given.
In an embodiment of the present invention, described that at least two Secular Variation Tendency numbers are decomposited from the time series
Value, comprising:
According to preset smoothness period, the time series is divided at least two subsequences, wherein the smooth week
Phase is at least one described time point;
For each subsequence, by weighting local regression LOESS to the former numerical value in the subsequence into
Row smoothing processing, the former numerical value after obtaining smoothing processing;
According to preset sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, is slided
Average value;
Calculating is filtered to the sliding average, obtains low pass magnitude;
The former numerical value after smoothing processing is subtracted each other with the low pass magnitude, obtains smoothness period value;
The smoothness period value is removed from the subsequence, cycle value is removed in acquisition;
It goes cycle value to be smoothed to described in acquisition by the LOESS, obtains Secular Variation Tendency numerical value.
In embodiments of the present invention, time series is mainly by Secular Variation Tendency numerical value, cyclically-varying numerical value and noise
Numerical value composition.Cyclically-varying numerical value and noise numerical value in time series are worth prediction reference little, it is therefore desirable to decompose
Secular Variation Tendency numerical value out.And during decomposition, it needs first to remove cyclically-varying numerical value, i.e., divides time series
For multiple subsequences, then each subsequence is smoothed, small throughput filtering, low pass magnitude is obtained, after smoothing processing
Former numerical value subtract each other with small throughput, obtain smoothness period value, smoothness period value removed from subsequence, can get includes long-term become
Change trend numerical value and noise numerical value remove cycle value, by can be obtained length to going cycle value to carry out LOESS smoothing processing
Phase variation tendency numerical value.
In an embodiment of the present invention, it is quasi- that the Secular Variation Tendency numerical value decomposited at described Dui carries out linear regression
It closes, before obtaining matched curve, further comprises:
Determine the trend numerical value quantity of the Secular Variation Tendency numerical value;
The Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity;
Obtain the average time point at the Secular Variation Tendency numerical value corresponding time point;
The described pair of Secular Variation Tendency numerical value decomposited carries out linear regression fit, obtains matched curve, comprising:
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biCharacterize i-th of the Secular Variation Tendency numerical value pair
The time point answered, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,It characterizes described flat
Equal time point, x characterize any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
In embodiments of the present invention, available long-term by obtaining the trend numerical value quantity of Secular Variation Tendency numerical value
Variation tendency average value and point of corresponding average time, according to Secular Variation Tendency average value, average time point, Ge Gechang
The fitting that phase variation tendency numerical value and Secular Variation Tendency numerical value corresponding time point obtain Secular Variation Tendency numerical value is bent
Line.
In an embodiment of the present invention, described that the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity
Secular Variation Tendency average value, comprising:
To the Secular Variation Tendency numerical value sum acquisition trend numerical value summation, and by the trend numerical value summation with it is described
Trend numerical value quantity is divided by, and the Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained;
The average time point for obtaining the Secular Variation Tendency numerical value corresponding time point, comprising:
Sequentially in time, when the last one described Secular Variation Tendency numerical value of the time sequencing is corresponding described
Between point, the time point corresponding with first of the time sequencing Secular Variation Tendency numerical value subtracts each other, and obtains trend
Total duration;
The trend total duration and the trend numerical value quantity are divided by, average time point is obtained.
In embodiments of the present invention, when obtaining Secular Variation Tendency average value, need to Secular Variation Tendency numerical value into
Row summation, then be divided by and can obtain with the trend numerical value quantity of Secular Variation Tendency.And obtaining average time point is by time sequence
The last one Secular Variation Tendency numerical value corresponding time point in column, time corresponding with first Secular Variation Tendency numerical value
Point subtracts each other acquisition trend total duration, and trend total duration and trend numerical value total quantity are divided by and can be obtained.
In an embodiment of the present invention, described according to the matched curve and preset early warning value, it determines described to be detected
Index corresponding fault pre-alarming time point, comprising:
It is substituted into preset early warning value as the parameter in the matched curve, it is described corresponding to obtain the parameter
It anticipates time point;
Using the corresponding any point-in-time of the parameter as fault pre-alarming time point.
In embodiments of the present invention, by being substituted into preset early warning value as dependent variable parameter in matched curve
The corresponding any point-in-time of parameter is obtained, when determining fault pre-alarming when the corresponding numerical value of index to be detected reaches early warning value
Between.
In order to more clearly illustrate technical solution of the present invention and advantage, below to one kind provided in an embodiment of the present invention
Equipment fault early-warning method is described in detail, and can specifically include following steps:
Step 201: corresponding at least two historical values of at least one operating index are stored in advance, wherein each history
Numerical value corresponds to a time point.
Specifically, it by storing historical values corresponding to the corresponding operating index of each equipment, can need to treat
When the index to be detected of detection device is predicted, the Secular Variation Tendency of index to be detected is determined according to historical values.
Step 202: determining at least one corresponding index to be detected of measurement equipment to be checked.
Step 203: being directed to each index to be detected, from the historical values of storage, determination is corresponding with index to be detected
At least two former numerical value, wherein the corresponding operating index of former numerical value is identical as index to be detected.
Specifically, it is predicted, is needed from the historical values of storage if you need to treat the index to be detected of detection device
With the corresponding former numerical value of target to be detected.
For example, storage equipment a is water pump for mining, and operating index is motor winding temperature, and time point is from 2017 1
The moon 1 to June 30, historical values are the corresponding temperature value of each time point.
Equipment b is motor, and operating index is vibration frequency, and time point is from March 1st, 2017 to June 30, history number
Value is the corresponding vibration number of each time point.
When measurement equipment to be checked is water pump for mining, then it can determine that the temperature value of equipment a is measurement equipment water pump for mining to be checked
Former numerical value.
Step 204: according to each former numerical value of former numerical value corresponding time point arrangement, generate correspond to index to be detected when
Between sequence.
Specifically, time series is ordered series of numbers made of the chronological order arrangement occurred according to numerical value, and therefore, it is necessary to will
Each original numerical value is arranged according to corresponding time point.
For example, by the corresponding temperature value of the motor winding temperature of equipment a, sequentially in time from January 1st, 2017
Time series is formed to arrangement on the 30th in June.
Step 205: according to preset smoothness period, time series being divided at least two subsequences, wherein smooth week
Phase is at least one time point.
Specifically, it when decompositing Secular Variation Tendency numerical value from time series, needs first to be divided into time series
Multiple subsequences, so as to handle respectively each subsequence.
For example, preset smoothness period is 1 month, and time series is divided into 6 subsequences.
Step 206: being directed to each subsequence, carried out by the former numerical value in weighting local regression LOESS sub-sequences flat
Sliding processing, the former numerical value after obtaining smoothing processing.
Specifically, it after time series is divided into each subsequence, needs to return each subsequence with LOESS,
And carry out sliding average according to preset sliding length and calculate acquisition sliding average, and its small throughput is filtered, that is, it is filtered
Wave calculates, after obtaining low pass magnitude, it is also necessary to subtract each other the former numerical value after smoothing processing with low pass magnitude, obtain smoothness period
Value, i.e. acquisition cyclically-varying data.Cyclically-varying data are removed from time series, that is, subtract smoothness period value, are obtained
Cycle value is removed, goes in cycle value to include Secular Variation Tendency numerical value and noise numerical value, to going cycle value to carry out again
LOESS is returned, and can be obtained Secular Variation Tendency numerical value.
Step 207: according to preset sliding length, sliding average calculating being carried out to the former numerical value after smoothing processing, is obtained
Sliding average.
Step 208: calculating being filtered to sliding average, obtains low pass magnitude.
Step 209: the former numerical value after smoothing processing being subtracted each other with low pass magnitude, obtains smoothness period value.
Step 210: smoothness period value is removed from subsequence, cycle value is removed in acquisition.
Step 211: going cycle value to be smoothed acquisition by LOESS, obtain Secular Variation Tendency numerical value.
Step 212: determining the trend numerical value quantity of Secular Variation Tendency numerical value.
Step 213: summing acquisition trend numerical value summation to Secular Variation Tendency numerical value, and by trend numerical value summation and trend
Numerical value quantity is divided by, and the Secular Variation Tendency average value of Secular Variation Tendency numerical value is obtained.
Specifically, after getting Secular Variation Tendency numerical value, need to Secular Variation Tendency numerical value sum, then with trend
Numerical value quantity is divided by, and can be obtained the Secular Variation Tendency average value of Secular Variation Tendency numerical value.
Step 214: sequentially in time, by the last one Secular Variation Tendency numerical value corresponding time of time sequencing
Point, time point corresponding with first Secular Variation Tendency numerical value of time sequencing subtract each other, and obtain trend total duration.
Specifically, when obtaining trend total duration, only need to by the time point of the last one in time series with first when
Between point subtract each other.
For example, first time point is May 1, the last one time point is May 26, and trend total duration is 26
It.
Step 215: trend total duration and trend numerical value quantity being divided by, average time point is obtained.
Step 216: linear regression fit being carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve.
Specifically, according to each Secular Variation Tendency numerical value of acquisition, average time point, Secular Variation Tendency average value,
To be that can obtain matched curve at Secular Variation Tendency numerical value corresponding time point.
Step 217: being substituted into preset early warning value as parameter in matched curve, get parms corresponding any time
Point.
Specifically, using early warning value as the corresponding independent variable that in dependent variable parameter dish such as matched curve, can get parms
Any point-in-time.
Step 218: using the corresponding any point-in-time of parameter as fault pre-alarming time point.
Specifically, after determining the corresponding independent variable any point-in-time of parameter, which is early warning value pair
The fault pre-alarming time point answered.
Step 219: before fault pre-alarming time point, warning failure pre-warning time point.
Specifically, it after determining fault pre-alarming time point, needs to warn before fault pre-alarming time point, so as to provide foot
The enough time takes steps to prevent the generation of failure, to avoid or reduce loss, reduce maintenance and maintenance expense.
To sum up, a kind of equipment fault early-warning method provided by the invention, can obtain to be detected from historical values
The corresponding historical values of index, and time series is formed sequentially in time.In the data of time series, contain prediction to
Testing index crosses the information of development and change, by decomposing and studying, can find the change in long term reflected in time series
Trend carries out equipment fault early-warning using Secular Variation Tendency information, formulates the more reasonable repair time, avoid time between overhauls(TBO) system
The too long equipment when not overhauling also that the cost of overhaul of equipment and period is arranged in fixed resolute increase enterprise has occurred
Failure.
As shown in figure 3, the embodiment of the invention provides a kind of equipment fault early-warning devices, comprising:
Numeric storage unit 301, for corresponding at least two historical values of at least one operating index to be stored in advance,
In, each historical values corresponding time point;
Information management unit 302, for determining at least one corresponding index to be detected of measurement equipment to be checked;For each institute
Index to be detected is stated, according to the historical values of the numeric storage unit 301 storage and the corresponding institute of the historical values
Time point is stated, the time series for corresponding to the index to be detected is generated;
Numeric processing unit 303 decomposites at least for generating in the time series from the information management unit 302
Two Secular Variation Tendency numerical value;Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains fitting
Curve;According to the matched curve and preset early warning value, the index to be detected corresponding fault pre-alarming time point is determined;
Alarm unit 304, for warning before the fault pre-alarming time point of the numeric processing unit 303 determination
Show the fault pre-alarming time point.
In embodiments of the present invention, information management unit is stored according to the corresponding index to be detected of measurement equipment to be checked and numerical value
The historical values of unit storage, produce the time sequencing corresponding to index to be detected, by numeric processing unit to message tube
The Time Series analysis for managing unit number stull, can obtain the matched curve of Secular Variation Tendency, in conjunction with matched curve and
Preset early warning value can determine fault pre-alarming time point, to warning failure pre-warning time point before fault pre-alarming time point
Early warning is carried out, the generation that time enough takes steps to prevent failure for service personnel can be provided, reduces enterprises' loss, again
Enterprise's cost of overhaul can resolutely be increased to avoid the time between overhauls(TBO).
In an embodiment of the present invention, the information management unit, for from the historical values of storage, determine with
The corresponding at least two former numerical value of the index to be detected, wherein the corresponding operating index of the original numerical value with it is described
Index to be detected is identical;According to the corresponding each former numerical value of time point arrangement of the original numerical value, generates and correspond to institute
State the time series of index to be detected.
In an embodiment of the present invention, the numeric processing unit is used for according to preset smoothness period, by the time
Sequence is divided at least two subsequences, wherein the smoothness period is at least one described time point;For each son
Sequence is smoothed the former numerical value in the subsequence by weighting local regression LOESS, obtains smoothing processing
The former numerical value afterwards;According to preset sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, is obtained
Obtain sliding average;Calculating is filtered to the sliding average, obtains low pass magnitude;By the former number after smoothing processing
Value is subtracted each other with the low pass magnitude, obtains smoothness period value;The smoothness period value is removed from the subsequence, week is gone in acquisition
Issue value;It goes cycle value to be smoothed to described in acquisition by the LOESS, obtains Secular Variation Tendency numerical value.
In an embodiment of the present invention, the numeric processing unit is further used for determining the Secular Variation Tendency number
The trend numerical value quantity of value;The Secular Variation Tendency for obtaining the Secular Variation Tendency numerical value according to the trend numerical value quantity is flat
Mean value;Obtain the average time point at the Secular Variation Tendency numerical value corresponding time point;
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biCharacterize i-th of the Secular Variation Tendency numerical value pair
The time point answered, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,It characterizes described flat
Equal time point, x characterize any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
In an embodiment of the present invention, the numeric processing unit, for being obtained to Secular Variation Tendency numerical value summation
Trend numerical value summation is obtained, and the trend numerical value summation and the trend numerical value quantity are divided by, the change in long term is obtained and becomes
The Secular Variation Tendency average value of gesture numerical value;Sequentially in time, by the last one described change in long term of the time sequencing
It is the trend numerical value corresponding time point, corresponding with the first Secular Variation Tendency numerical value of the time sequencing described
Time point subtracts each other, and obtains trend total duration;The trend total duration and the trend numerical value quantity are divided by, average time is obtained
Point.
In an embodiment of the present invention, the numeric processing unit, for using preset early warning value as the parameter generation
Enter in the matched curve, obtains the corresponding any point-in-time of the parameter;When the parameter is corresponding described any
Between point be used as fault pre-alarming time point.
The each embodiment of the present invention at least has the following beneficial effects:
It 1, can in embodiments of the present invention, according to the historical values of measurement equipment to be checked corresponding index to be detected and storage
The time sequencing for corresponding to index to be detected is generated, by the way that Secular Variation Tendency can be obtained to time series decomposition analysis
Matched curve can determine fault pre-alarming time point in conjunction with matched curve and preset early warning value, fault pre-alarming time point it
It is preceding that early warning is carried out to warning failure pre-warning time point, time enough can be provided and take steps to prevent failure for service personnel
Generation, reduce enterprises' loss, and can resolutely increase enterprise's cost of overhaul to avoid the time between overhauls(TBO).
2, in embodiments of the present invention, time series is made of multiple former numerical value according to the arrangement of corresponding time point,
And former numerical value is then that determination is corresponding with index to be detected from the historical values of storage.According to historical values to future trend
It is predicted with data, and the suggestion to service personnel's repair time is given.
3, in embodiments of the present invention, time series mainly by Secular Variation Tendency numerical value, cyclically-varying numerical value and is made an uproar
Sound numerical value composition.Cyclically-varying numerical value and noise numerical value in time series are worth prediction reference little, it is therefore desirable to point
Solve Secular Variation Tendency numerical value.And during decomposition, it needs first to remove cyclically-varying numerical value, i.e., draws time series
It is divided into multiple subsequences, then each subsequence is smoothed, small throughput filtering, low pass magnitude is obtained, by smoothing processing
Former numerical value afterwards subtracts each other with small throughput, obtains smoothness period value, and smoothness period value is removed from subsequence, and it includes long-term for can get
Variation tendency numerical value and noise numerical value remove cycle value, by can be obtained to going cycle value to carry out LOESS smoothing processing
Secular Variation Tendency numerical value.
4, in embodiments of the present invention, by obtaining the trend numerical value quantity of Secular Variation Tendency numerical value, available length
Phase variation tendency average value and point of corresponding average time, according to Secular Variation Tendency average value, average time point, each
The fitting that Secular Variation Tendency numerical value and Secular Variation Tendency numerical value corresponding time point obtain Secular Variation Tendency numerical value is bent
Line.
5, it in embodiments of the present invention, when obtaining Secular Variation Tendency average value, needs to Secular Variation Tendency numerical value
It sums, then is divided by and can obtain with the trend numerical value quantity of Secular Variation Tendency.And obtaining average time point is by the time
The last one Secular Variation Tendency numerical value corresponding time point in sequence, when corresponding with first Secular Variation Tendency numerical value
Between point subtract each other acquisition trend total duration, trend total duration and trend numerical value total quantity are divided by and can be obtained.
It should be noted that, in this document, such as first and second etc relational terms are used merely to an entity
Or operation is distinguished with another entity or operation, is existed without necessarily requiring or implying between these entities or operation
Any actual relationship or order.Moreover, the terms "include", "comprise" or its any other variant be intended to it is non-
It is exclusive to include, so that the process, method, article or equipment for including a series of elements not only includes those elements,
It but also including other elements that are not explicitly listed, or further include solid by this process, method, article or equipment
Some elements.In the absence of more restrictions, the element limited by sentence " including one ", is not arranged
Except there is also other identical factors in the process, method, article or apparatus that includes the element.
Finally, it should be noted that the foregoing is merely presently preferred embodiments of the present invention, it is merely to illustrate skill of the invention
Art scheme, is not intended to limit the scope of the present invention.Any modification for being made all within the spirits and principles of the present invention,
Equivalent replacement, improvement etc., are included within the scope of protection of the present invention.
Claims (10)
1. a kind of equipment fault early-warning method, which is characterized in that
Corresponding at least two historical values of at least one operating index are stored in advance, wherein each historical values are corresponding
One time point;
Include:
Determine at least one corresponding index to be detected of measurement equipment to be checked;
For each index to be detected, according to the historical values of storage and the historical values corresponding time
Point generates the time series for corresponding to the index to be detected;
At least two Secular Variation Tendency numerical value are decomposited from the time series;
Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;
According to the matched curve and preset early warning value, the index to be detected corresponding fault pre-alarming time point is determined;
Before the fault pre-alarming time point, the fault pre-alarming time point is warned.
2. the method according to claim 1, wherein
It is described according to the historical values of storage and the historical values corresponding time point, generate correspond to it is described to
The time series of Testing index, comprising:
From the historical values of storage, determining at least two originals numerical value corresponding with the index to be detected, wherein institute
It is identical as the index to be detected to state the corresponding operating index of former numerical value;
According to the corresponding each former numerical value of time point arrangement of the original numerical value, generates and correspond to the index to be detected
Time series.
3. according to the method described in claim 2, it is characterized in that,
It is described that at least two Secular Variation Tendency numerical value are decomposited from the time series, comprising:
According to preset smoothness period, the time series is divided at least two subsequences, wherein the smoothness period is
At least one described time point;
For each subsequence, the former numerical value in the subsequence is carried out by weighting local regression LOESS flat
Sliding processing, the former numerical value after obtaining smoothing processing;
According to preset sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, obtains sliding average
Value;
Calculating is filtered to the sliding average, obtains low pass magnitude;
The former numerical value after smoothing processing is subtracted each other with the low pass magnitude, obtains smoothness period value;
The smoothness period value is removed from the subsequence, cycle value is removed in acquisition;
It goes cycle value to be smoothed to described in acquisition by the LOESS, obtains Secular Variation Tendency numerical value.
4. method according to any one of claims 1 to 3, which is characterized in that
The Secular Variation Tendency numerical value decomposited at described Dui carries out linear regression fit, before obtaining matched curve, into
One step includes:
Determine the trend numerical value quantity of the Secular Variation Tendency numerical value;
The Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity;
Obtain the average time point at the Secular Variation Tendency numerical value corresponding time point;
The described pair of Secular Variation Tendency numerical value decomposited carries out linear regression fit, obtains matched curve, comprising:
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biIt is corresponding to characterize i-th of the Secular Variation Tendency numerical value
The time point, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,Characterize the mean time
Between point, x characterizes any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
5. according to the method described in claim 4, it is characterized in that,
The Secular Variation Tendency average value that the Secular Variation Tendency numerical value is obtained according to the trend numerical value quantity, packet
It includes:
It sums acquisition trend numerical value summation to the Secular Variation Tendency numerical value, and by the trend numerical value summation and the trend
Numerical value quantity is divided by, and the Secular Variation Tendency average value of the Secular Variation Tendency numerical value is obtained;
The average time point for obtaining the Secular Variation Tendency numerical value corresponding time point, comprising:
Sequentially in time, by the last one described Secular Variation Tendency numerical value corresponding time of the time sequencing
Point, the time point corresponding with first of the time sequencing Secular Variation Tendency numerical value subtract each other, and it is total to obtain trend
Duration;
The trend total duration and the trend numerical value quantity are divided by, average time point is obtained;
And/or
It is described according to the matched curve and preset early warning value, determine the index to be detected corresponding fault pre-alarming time
Point, comprising:
Substituted into preset early warning value as the parameter in the matched curve, obtain the parameter it is corresponding described any when
Between point;
Using the corresponding any point-in-time of the parameter as fault pre-alarming time point.
6. a kind of equipment fault early-warning device characterized by comprising
Numeric storage unit, for corresponding at least two historical values of at least one operating index to be stored in advance, wherein each
A historical values corresponding time point;
Information management unit, for determining at least one corresponding index to be detected of measurement equipment to be checked;For each described to be checked
Index is surveyed, the historical values stored according to the numeric storage unit and the historical values corresponding time point,
Generate the time series for corresponding to the index to be detected;
Numeric processing unit decomposites at least two long-term changes for generating from the information management unit in the time series
Change trend numerical value;Linear regression fit is carried out to the Secular Variation Tendency numerical value decomposited, obtains matched curve;According to institute
Matched curve and preset early warning value are stated, determines the index to be detected corresponding fault pre-alarming time point;
Alarm unit, for warning the failure before the fault pre-alarming time point that the numeric processing unit determines
Pre-warning time point.
7. device according to claim 6, which is characterized in that
The information management unit, for from the historical values of storage, determination to be corresponding with the index to be detected
At least two former numerical value, wherein the corresponding operating index of the original numerical value is identical as the index to be detected;According to described
The corresponding each former numerical value of time point arrangement of former numerical value, generates the time series for corresponding to the index to be detected.
8. device according to claim 7, which is characterized in that
The numeric processing unit, for according to preset smoothness period, the time series to be divided at least two sub- sequences
Column, wherein the smoothness period is at least one described time point;For each subsequence, by weighting local regression
LOESS is smoothed the former numerical value in the subsequence, the former numerical value after obtaining smoothing processing;According to pre-
If sliding length, sliding average calculating is carried out to the former numerical value after smoothing processing, obtains sliding average;To the cunning
Dynamic average value is filtered calculating, obtains low pass magnitude;The former numerical value after smoothing processing is subtracted each other with the low pass magnitude,
Obtain smoothness period value;The smoothness period value is removed from the subsequence, cycle value is removed in acquisition;Pass through the LOESS
It goes cycle value to be smoothed to described in acquisition, obtains Secular Variation Tendency numerical value.
9. according to the device any in claim 6 to 8, which is characterized in that
The numeric processing unit is further used for determining the trend numerical value quantity of the Secular Variation Tendency numerical value;According to institute
State the Secular Variation Tendency average value that trend numerical value quantity obtains the Secular Variation Tendency numerical value;The change in long term is obtained to become
The average time point at the gesture numerical value corresponding time point;
The matched curve meets following formula:
Wherein, aiCharacterize i-th of the Secular Variation Tendency numerical value, biIt is corresponding to characterize i-th of the Secular Variation Tendency numerical value
The time point, n characterize the trend numerical value quantity,The Secular Variation Tendency average value is characterized,Characterize the mean time
Between point, x characterizes any point-in-time, and y characterizes the corresponding parameter of any point-in-time.
10. device according to claim 9, which is characterized in that
The numeric processing unit is used for acquisition trend numerical value summation of summing to the Secular Variation Tendency numerical value, and will be described
Trend numerical value summation is divided by with the trend numerical value quantity, and the Secular Variation Tendency for obtaining the Secular Variation Tendency numerical value is average
Value;Sequentially in time, by the last one described Secular Variation Tendency numerical value corresponding time point of the time sequencing,
The time point corresponding with first of the time sequencing Secular Variation Tendency numerical value subtracts each other, obtain trend it is total when
It is long;The trend total duration and the trend numerical value quantity are divided by, average time point is obtained;
And/or
The numeric processing unit obtains institute for substituting into preset early warning value as the parameter in the matched curve
State the corresponding any point-in-time of parameter;Using the corresponding any point-in-time of the parameter as fault pre-alarming time point.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810890568.3A CN109087008A (en) | 2018-08-07 | 2018-08-07 | A kind of equipment fault early-warning method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810890568.3A CN109087008A (en) | 2018-08-07 | 2018-08-07 | A kind of equipment fault early-warning method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109087008A true CN109087008A (en) | 2018-12-25 |
Family
ID=64834020
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810890568.3A Pending CN109087008A (en) | 2018-08-07 | 2018-08-07 | A kind of equipment fault early-warning method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109087008A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110766054A (en) * | 2019-09-23 | 2020-02-07 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
CN110930108A (en) * | 2019-10-21 | 2020-03-27 | 深圳技术大学 | Government affair data processing method, system, device and storage medium |
CN112579391A (en) * | 2020-12-14 | 2021-03-30 | 浪潮云信息技术股份公司 | Distributed database automatic operation and maintenance method and system based on artificial intelligence |
CN112649153A (en) * | 2020-11-27 | 2021-04-13 | 国网冀北电力有限公司电力科学研究院 | Method, device and equipment for detecting air tightness of closed equipment |
US11740619B2 (en) | 2019-05-31 | 2023-08-29 | Nanjing University Of Aeronautics And Astronautics | Malfunction early-warning method for production logistics delivery equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103793601A (en) * | 2014-01-20 | 2014-05-14 | 广东电网公司电力科学研究院 | Turbine set online fault early warning method based on abnormality searching and combination forecasting |
CN104699890A (en) * | 2015-01-07 | 2015-06-10 | 国家电网公司 | Modeling method of short-term forewarning model for heavy overload of distribution transformer of urban power network |
CN107588906A (en) * | 2017-09-11 | 2018-01-16 | 北京金风慧能技术有限公司 | Liquid leakage method for early warning and device for liquid-cooling circulating system |
CN107909230A (en) * | 2018-01-12 | 2018-04-13 | 国网福建省电力有限公司 | A kind of modeling method of the short-term Early-warning Model of rural power grids distribution transforming heavy-overload |
-
2018
- 2018-08-07 CN CN201810890568.3A patent/CN109087008A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103793601A (en) * | 2014-01-20 | 2014-05-14 | 广东电网公司电力科学研究院 | Turbine set online fault early warning method based on abnormality searching and combination forecasting |
CN104699890A (en) * | 2015-01-07 | 2015-06-10 | 国家电网公司 | Modeling method of short-term forewarning model for heavy overload of distribution transformer of urban power network |
CN107588906A (en) * | 2017-09-11 | 2018-01-16 | 北京金风慧能技术有限公司 | Liquid leakage method for early warning and device for liquid-cooling circulating system |
CN107909230A (en) * | 2018-01-12 | 2018-04-13 | 国网福建省电力有限公司 | A kind of modeling method of the short-term Early-warning Model of rural power grids distribution transforming heavy-overload |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11740619B2 (en) | 2019-05-31 | 2023-08-29 | Nanjing University Of Aeronautics And Astronautics | Malfunction early-warning method for production logistics delivery equipment |
CN110766054A (en) * | 2019-09-23 | 2020-02-07 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
CN110766054B (en) * | 2019-09-23 | 2022-05-10 | 东软集团股份有限公司 | Early warning method and device for fracturing sand plugging, storage medium and equipment |
CN110930108A (en) * | 2019-10-21 | 2020-03-27 | 深圳技术大学 | Government affair data processing method, system, device and storage medium |
CN112649153A (en) * | 2020-11-27 | 2021-04-13 | 国网冀北电力有限公司电力科学研究院 | Method, device and equipment for detecting air tightness of closed equipment |
CN112579391A (en) * | 2020-12-14 | 2021-03-30 | 浪潮云信息技术股份公司 | Distributed database automatic operation and maintenance method and system based on artificial intelligence |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109087008A (en) | A kind of equipment fault early-warning method and device | |
CN109766334B (en) | Processing method and system for online monitoring abnormal data of power equipment | |
EP2813978A1 (en) | Industrial asset health model update | |
CN108287317B (en) | Battery power prediction model generation method and system and power prediction method and system | |
Zhao et al. | Which is better for replacement policies with continuous or discrete scheduled times? | |
JP5684813B2 (en) | Abnormality prediction diagnosis method for oil-filled electrical equipment | |
US7908300B2 (en) | Guided entry system for individuals for annotating process deviations | |
Kwon et al. | Fatigue life assessment and lifetime management of aluminum ships using life-cycle optimization | |
CN111722058A (en) | Power information system fault detection method, device and medium based on knowledge graph | |
JP7053152B2 (en) | Systems and methods for optimizing recommended inspection intervals | |
CN110046152A (en) | A method of processing electricity consumption data missing values | |
CN116224137A (en) | UPS equipment fault early warning method and system | |
CN108663501A (en) | A kind of predicting model for dissolved gas in transformer oil method and system | |
CN107657121B (en) | Aircraft structure performance prediction processing method and system based on corrosion level evaluation | |
CN117029968A (en) | Traffic data diagnosis method, system, storage medium and electronic equipment | |
CN113987027A (en) | Equipment state determination method, device and medium based on fault prediction | |
CN117193240A (en) | Electric control cabinet fault early warning system for electrochemical aluminum production | |
CN116720983A (en) | Power supply equipment abnormality detection method and system based on big data analysis | |
CN116702463A (en) | Method, device, equipment and storage medium for predicting residual life of heating surface pipe | |
CN116316635A (en) | Electric power cooperative control method and system based on measurement information | |
CN112418534B (en) | Method and device for predicting quantity of collected parts, electronic equipment and computer readable storage medium | |
CN117150032B (en) | Intelligent maintenance system and method for hydropower station generator set | |
Chen et al. | A permutation entropy-based importance measure for condition monitoring data fusion in fault diagnosis | |
JP2013214256A (en) | Integrated maintenance system for power plant | |
Peifeng | Structural analysis of the remote performance monitoring system used in modern agricultural machinery |
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: 20181225 |