CN104515591B - A kind of high-precision coal pulverizer vibration detection process - Google Patents
A kind of high-precision coal pulverizer vibration detection process Download PDFInfo
- Publication number
- CN104515591B CN104515591B CN201410851654.5A CN201410851654A CN104515591B CN 104515591 B CN104515591 B CN 104515591B CN 201410851654 A CN201410851654 A CN 201410851654A CN 104515591 B CN104515591 B CN 104515591B
- Authority
- CN
- China
- Prior art keywords
- coal pulverizer
- data
- matrix
- parameter
- detection model
- 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.)
- Active
Links
Landscapes
- Disintegrating Or Milling (AREA)
- Crushing And Grinding (AREA)
Abstract
The invention discloses a kind of high-precision coal pulverizer vibration detection process, including：(a) gathered successively according to sequential n groups reflection coal pulverizer running situation history data, and by actual test obtain this n groups history data each corresponding to actual vibration intensity data；(b) coal pulverizer vibration detection model is built using phase space algorithm and extreme learning machine algorithm based on above-mentioned data, to build coal pulverizer vibration detection model, optimization then is performed to detection model；(c) the work at present data to coal pulverizer carry out on-line measurement, corresponding output result are then obtained according to optimal detection model, the output result is required coal pulverizer current vibration state outcome.By means of the invention it is possible in order to manipulate, high efficiency and accurately mode realizes the real-time detection to coal pulverizer oscillation intensity, and the application scenario for the large thermal power plant etc that is particularly suitable for use in.
Description
Technical field
The invention belongs to thermoelectricity electrification technique association area, is examined more particularly, to a kind of high-precision coal-grinding machine vibration
Survey process.
Background technology
Coal pulverizer is by coal crusher and wears into the machinery of coal dust, its important auxiliary equipment as coal-powder boiler, in fire coal
It is applied widely in thermal power plant.In actual motion, coal pulverizer generally easily vibrates, and this will necessarily cause to grind power
The problems such as reduction, abrasion, sealing life shortening, joint breaking and weld seam drawing crack, mutually reply power plant production guarantee are played unfavorable
Influence.Effectively its running status can be judged according to the oscillation intensity of coal pulverizer.Therefore, the coal pulverizer of power plant needs at present
Vibration measurement device is equipped with, to monitor the vibration state of coal pulverizer in real time.
Coal pulverizer vibration detection mode of the prior art is most to be examined for artificial point, is wasted time and energy and can not be realized real-time prison
Control, therefore, the ARIMA methods based on time series, the SVMs based on recurrence have been proposed, and ANN
The methods of network and gray theory, is modeled for the vibration to coal pulverizer.However, further investigations have shown that, it is above-mentioned existing
There is more deficiency or defect in measurement accuracy, model training and on-line prediction etc. in method, for example, ARIMA models
For linear autoregression smoothing prediction, the simulation less effective for non-linear values such as coal pulverizer vibratory outputs；Using artificial neuron
Network or SVMs, which are predicted, can perform vibratory output recurrence, but because the parameter of model needs iteration optimizing, model
Training can take a substantial amount of time, it is difficult to applied to on-line prediction, etc..Therefore, it is more perfect to need searching badly for association area
Solution, to realize high-acruracy survey and on-line prediction to power plant coal vibrational state.
The content of the invention
For the disadvantages described above or Improvement requirement of prior art, the invention provides a kind of high-precision coal-grinding machine vibration to examine
Process is surveyed, wherein by selecting coal pulverizer critical operational parameters and its historical vibration intensity data, and come based on this
Specific vibration measurement model and algorithm are built to perform on-line measurement, mutually should be able to high efficiency, accurately perform coal pulverizer
Real-time oscillation intensity measurement, is therefore particularly suitable for the application scenarios such as thermal power plant's coal pulverizer oscillation intensity high-acruracy survey.
To achieve the above object, it is proposed, according to the invention, a kind of high-precision coal pulverizer vibration detection process is provided, its
It is characterised by, this method comprises the following steps：
(a) gather the history data of n groups reflection coal pulverizer running situation successively according to sequential, and pass through actual test
Obtain this n groups history data each corresponding actual vibration intensity data；Wherein each group history data includes
Sampling moment numbering, wind powder mixture temperature, carbon monoxide content, frequency converter rotating speed and the lubricating oil temperature of discharge knockout drum,
The operating current and import primary air flow of coal pulverizer, the aperture of cooling water valve, the lubricating oil temperature of grinding roller bearing and motor driving
The bearing temperature of side, the tank temperature of hydraulic station, loading fuel pump outlet pressure, maximum ventilation resistance, First air controllable register
The operational data of this 17 aspects of coal calorific value, moisture and interior water used by aperture, and coal pulverizer；
(b) all data obtained to step (a) perform normalized, and phase space algorithm is used based on these data
With extreme learning machine algorithm to build coal pulverizer vibration detection model, and optimization, its specific mistake are performed to detection model
Journey is as follows：
(b1) parameter corresponding to being preset respectively to phase space algorithm and extreme learning machine algorithm first, i.e. Embedded dimensions m, prolong
Then these parameters are performed initialization by slow operator τ, penalty factor, nuclear parameter γ in a random basis；
(b2) according to the Embedded dimensions m and delay operator τ after initialization, the actual vibration intensity data of coal pulverizer is write as
Phase space matrix X as follows form：
Wherein, x1,x2,…,xnThe n group actual vibration intensity datas gathered in step (a) are represented respectively, and they are common
The phase point of phase space is constituted, and total line number of the phase space matrix is n- (m-1) τ, and total columns is m；
(b3) history data is written as to shown matrix Y form：
Wherein, i=1,2 ..., 17, j=1,2 ..., n；yi,jThen represent that i-th of parameter in above-mentioned 17 parameters exists
Numbering is the data corresponding to the j sampling moment；
(b4) the sampling moment corresponding to last row actual vibration intensity data in the phase space matrix X,
A series of history datas inscribed when taking out the sampling from the matrix Y are placed in the phase space rail as auxiliary parameter
Before last row in mark X, matrix Z as follows is thus obtained：
Wherein, the total line numbers of matrix Z are n- (m-1) τ, and total columns is m+17；
(b5) by each line number in the matrix Z obtained by step (b4) according to this and preceding m+16 row data make jointly
For the input data of extreme learning machine, while using last row in the matrix Z as the correction data exported；
(b6) according to the penalty factor after initialization and nuclear parameter γ, with reference to the input data in step (b5) and output
Correction data, the modeling error of extreme learning machine is calculated；
(b7) whether the modeling error that judgment step (b6) is calculated falls within default threshold value：If so, then will step
Suddenly the Embedded dimensions m employed in (b6), delay operator τ, penalty factor and nuclear parameter γ are as final detection model parameter
Optimum results, and the optimized results by the current detection model established using this as coal pulverizer vibration detection, otherwise, then turn
Enter step (b8)；
(b8) intersection is performed to above-mentioned Embedded dimensions m, delay operator τ, penalty factor and nuclear parameter γ using genetic algorithm
And variation, then perform circulation in the way of step (b2)~(b7), until the modeling error that is drawn drop into it is default
Untill within threshold value, while export the detection model of optimization and its corresponding model parameter optimized results；
(c) the work at present data for being related to above-mentioned 17 aspects to coal pulverizer carry out on-line measurement, then according to step (b)
The optimal detection model obtained obtains corresponding output result, and the output result is required coal pulverizer current vibration shape
State testing result.
As it is further preferred that in step (b6), it is preferred to use the method for 5 retransposings checking is described to be calculated
The modeling error of extreme learning machine.
As it is further preferred that by step (b) obtain optimal models after, preferably can also be periodically to its model
Parameter is updated, and in the process, newest sample data is added in sample set, while retains in original sample set
Classical sample data.
As it is further preferred that the coal pulverizer is preferably the coal pulverizer of large thermal power plant.
In general, by the contemplated above technical scheme of the present invention compared with prior art, due to taking into full account
Current coal pulverizer operational factor and historical vibration situation select to use phase to the multifactor impact of its current vibration intensity
Space arithmetic and extreme learning machine algorithm perform on-line checking to build appropriate detection model, mutually should be able in order to manipulate,
High efficiency and accurately mode realize the real-time detection to coal pulverizer oscillation intensity, and the large thermal power plant etc that is particularly suitable for use in
Application scenario.
Brief description of the drawings
Fig. 1 is the concrete technology flow process figure according to the coal pulverizer vibration detection process constructed by the present invention；
Fig. 2 is the principle operation diagram according to the coal pulverizer method for detecting vibration of the present invention.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as in addition, technical characteristic involved in each embodiment of invention described below
Conflict can is not formed each other to be mutually combined.
Fig. 1 be according to the concrete technology flow process figure of the coal pulverizer vibration detection process constructed by the present invention, Fig. 2 be by
According to the principle operation diagram of the coal pulverizer method for detecting vibration of the present invention.As shown in Figures 1 and 2, the process mainly includes
Following technological operation step：
First, it is screening and the acquisition step of coal pulverizer running parameter.
Real-time parameter up to 207 (such as primary air flow, a wind-warm syndrome etc.) in coal pulverizer work system, coal pair in mill
The ature of coal information answered has 6 (such as low heat valve, moisture as received coal etc.), if all parameters are involved in modeling, it will so that
Model complex, generalization ability weakens, and is likely to result in dimension disaster.Therefore, first passed through in the present invention between dimension
Coefficient correlation is screened, and Pearson correlation coefficient is more than in 0.99 two dimensions and removes one of them, then based on genetic algorithm
Feature selecting, using wrapper structure, have chosen 17 parameters therein, wherein by SIS real-time data bases obtain 14, bag
Include：Sampling moment value, discharge knockout drum wind powder mixture temperature, outlet carbon monoxide content, discharge knockout drum frequency converter rotating speed,
Motor inboard bearing temperature, coal pulverizer electric current, separator lubricating oil temperature, import primary air flow, cooling water valve aperture, grinding roller
Bearing lubrication oil temperature, heat primary air controllable register aperture, hydraulic station tank temperature, maximum ventilation resistance, hydraulic oil unit loading
Fuel pump outlet pressure；Ature of coal information 3, including：Coal calorific value, moisture, interior water.
Specifically, in this step, the history run number of n groups reflection coal pulverizer running situation is gathered successively according to sequential
According to, and this n groups history data each corresponding actual vibration intensity data is obtained by actual test；It is wherein described each
Group history data includes sampling moment numbering, the wind powder mixture temperature of discharge knockout drum, carbon monoxide content, frequency conversion
Device rotating speed and lubricating oil temperature, the operating current and import primary air flow of coal pulverizer, aperture, the profit of grinding roller bearing of cooling water valve
The bearing temperature of oil temperature and motor driving side, the tank temperature of hydraulic station, loading fuel pump outlet pressure, maximum ventilation resistance
The aperture of power, First air controllable register, and the work used by coal pulverizer in terms of 17, coal calorific value, moisture and interior water etc.
Data.
Then, it is ready to after training set and test set, normalizing is performed to above-mentioned obtained all data in the present invention
Change is handled, and is then based on these data, and selection builds the inspection of coal-grinding machine vibration using phase space algorithm and extreme learning machine algorithm
Model is surveyed, and optimization is performed to detection model.For its principle, substantially step includes setting phase space weight first for it
Structure parameter (Embedded dimensions m, delay operator τ), extreme learning machine parameter (penalty factor, nuclear parameter γ) optimizing space, in optimizing
This four parameters are initialized at random in space, then establish measurement mould using training set and the parameter being randomly derived
Type, the precision of test set detection model is recycled, passes through the structure again that intersects and make a variation if model accuracy can not reach requirement
Shape parameter is modeled, untill obtaining the model of precision satisfaction, finally the structure of the model is stored and measured for actual；
The further specific explanations of above-mentioned operation principle can be found in the related introduction of phase space algorithm and extreme learning machine, therefore herein not
Repeat again.
More specifically, the structure coal pulverizer vibration detection model and to detection model perform optimization process such as
Under：
(1) parameter corresponding to being preset respectively to phase space algorithm and extreme learning machine algorithm first, i.e. Embedded dimensions m, prolong
Then these parameters are performed initialization by slow operator τ, penalty factor, nuclear parameter γ in a random basis；
(2) according to the Embedded dimensions m and delay operator τ after initialization, the actual vibration intensity data of coal pulverizer is write as
Phase space matrix X as follows form：
Wherein, x1,x2,…,xnThe n group actual vibration intensity datas gathered in step (a) are represented respectively, and they are common
With the phase point for constituting phase space, and total line number of the phase space matrix is n- (m-1) τ, and total columns is m；
(3) history data is written as to shown matrix Y form：
Wherein, i=1,2 ..., 17, j=1,2 ..., n；yi,jThen represent that i-th of parameter in above-mentioned 17 parameters exists
Numbering is the data corresponding to the j sampling moment；
(4) the sampling moment corresponding to last row actual vibration intensity data in the phase space matrix X, from
A series of history datas inscribed when taking out the sampling in the matrix Y are placed in the trajectory of phase space as auxiliary parameter
Before last row in X, matrix Z as follows is thus obtained：
Wherein, the total line numbers of matrix Z are n- (m-1) τ, and total columns is m+17；
(5) by each line number in the matrix Z obtained by previous step according to this and preceding m+16 row data make jointly
For the input data of extreme learning machine, while using last row in the matrix Z as the correction data exported；
(6) according to the penalty factor after initialization and nuclear parameter γ, with reference to the input data obtained by previous step and defeated
Go out correction data, the modeling error of extreme learning machine is calculated；
(7) whether the modeling error that judgment step is calculated falls within default threshold value：If so, then it is used
Embedded dimensions m, delay operator τ, penalty factor and nuclear parameter γ as final detection model parameter optimization result, and will
Optimized results using the current detection model that this is established as coal pulverizer vibration detection, otherwise, are then transferred to next step；
(b8) intersection is performed to above-mentioned Embedded dimensions m, delay operator τ, penalty factor and nuclear parameter γ using genetic algorithm
And variation, circulation is then performed in the way of above step, until the modeling error that is drawn drop into default threshold value it
Untill interior, while export the detection model of optimization and its corresponding model parameter optimized results.
Finally, it is actual coal-grinding machine vibration on-line checking operation.
For example by interface degree, the work at present data for being related to above-mentioned 17 aspects to coal pulverizer again are surveyed online
Amount, while the history coal pulverizer oscillation intensity data needed for calculating are transferred, these data are organized into the lattice of mode input requirement
Formula, and normalizing, then utilize according to the optimal detection model above obtained to obtain corresponding output result, the output
As a result it is required coal pulverizer current vibration state-detection result.
In addition, over time, equipment can also be gradually degraded, it is therefore necessary to periodically model is modified；Together
When the obtained sample data of check staff's measurement be continuously increased, can also further correction model parameter, improve model accuracy.
During Modifying model, newest sample data is added in sample set as far as possible, while retained in original sample set
Classical sample data, remove repetition, very similar sample data.Using the sample set re -training model newly obtained, replace
Change original model.The training detailed process and preceding step flow of model are similar.
To sum up, the present invention is based on the measurement model of phase space reconfiguration and extreme learning machine can basis be real-time, performs exactly
The data estimation of coal pulverizer oscillation intensity, and substantial amounts of actual test result shows, its relative error can guarantee that 10% with
It is interior, thus there is very high use value, and the applied fields such as thermal power plant's coal pulverizer oscillation intensity high-acruracy survey that are particularly suitable for use in
Close.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, all any modification, equivalent and improvement made within the spirit and principles of the invention etc., all should be included
Within protection scope of the present invention.
Claims (4)
1. a kind of high-precision coal pulverizer vibration detection process, it is characterised in that this method comprises the following steps：
(a) gather the history data of n groups reflection coal pulverizer running situation successively according to sequential, and obtained by actual test
The respective corresponding actual vibration intensity data of this n groups history data；Wherein each group history data includes sampling
Moment numbers, wind powder mixture temperature, carbon monoxide content, frequency converter rotating speed and the lubricating oil temperature of discharge knockout drum, coal-grinding
The operating current and import primary air flow of machine, the aperture of cooling water valve, the lubricating oil temperature of grinding roller bearing and motor driving side
Bearing temperature, the tank temperature of hydraulic station, loading fuel pump outlet pressure, maximum ventilation resistance, the aperture of First air controllable register,
And the operational data used by coal pulverizer in terms of this 17, coal calorific value, moisture and interior water；
(b) all data obtained to step (a) perform normalized, and phase space algorithm and pole are used based on these data
Limit learning machine algorithm performs optimization to build coal pulverizer vibration detection model to detection model, and its detailed process is such as
Under：
(b1) parameter corresponding to being preset respectively to phase space algorithm and extreme learning machine algorithm first, i.e. Embedded dimensions m, delay are calculated
Then these parameters are performed initialization by sub- τ, penalty factor, nuclear parameter γ in a random basis；
(b2) according to the Embedded dimensions m and delay operator τ after initialization, the actual vibration intensity data of coal pulverizer is written as
Shown phase space matrix X form：
Wherein, x1,x2,…,xnThe n group actual vibration intensity datas gathered in step (a) are represented respectively, and they are collectively formed
The phase point of phase space, and total line number of the phase space matrix is n- (m-1) τ, and total columns is m；
(b3) history data is written as to shown matrix Y form：
Wherein, i=1,2 ..., 17, j=1,2 ..., n；yi,jThen represent that i-th of parameter in above-mentioned 17 parameters is being numbered
For the data corresponding to the j sampling moment；
(b4) the sampling moment corresponding to last row actual vibration intensity data in the phase space matrix X, from institute
State and a series of history datas for being inscribed during the sampling are taken out in matrix Y be placed in the phase space matrix X as auxiliary parameter
In last row before, thus obtain matrix Z as follows：
Wherein, total line number of the matrix Z is n- (m-1) τ, and total columns is m+17；
(b5) by each line number in the matrix Z obtained by step (b4) according to this and preceding m+16 row data collectively as pole
Limit the input data of learning machine, while the correction data using last row in the matrix Z as output；
(b6) according to the penalty factor after initialization and nuclear parameter γ, contrasted with reference to the input data in step (b5) and output
Data, the modeling error of extreme learning machine is calculated；
(b7) whether the modeling error that judgment step (b6) is calculated falls within default threshold value：If so, then by step
(b6) the Embedded dimensions m, delay operator τ, penalty factor and nuclear parameter γ employed in are excellent as final detection model parameter
Change result, and the optimized results by the current detection model established using this as coal pulverizer vibration detection, otherwise, be then transferred to
Step (b8)；
(b8) above-mentioned Embedded dimensions m, delay operator τ, penalty factor and nuclear parameter γ are performed using genetic algorithm and intersects and become
It is different, circulation then is performed in the way of step (b2)~(b7), until the modeling error drawn drops into default threshold value
Within untill, while export the detection model of optimization and its corresponding model parameter optimized results；
(c) the work at present data for being related to above-mentioned 17 aspects to coal pulverizer carry out on-line measurement, are then obtained according to step (b)
Optimal detection model obtain corresponding output result, the output result is required coal pulverizer current vibration state inspection
Survey result.
2. coal pulverizer vibration detection process as claimed in claim 1, it is characterised in that in step (b6), using 5 weights
The method of cross validation is calculated the modeling error of the extreme learning machine.
3. coal pulverizer vibration detection process as claimed in claim 1 or 2, it is characterised in that obtained by step (b)
After optimal models, also periodically its model parameter is updated, in the process, newest sample data is added to sample
Concentrate, while retain the classical sample data in original sample set.
4. coal pulverizer vibration detection process as claimed in claim 1, it is characterised in that the coal pulverizer is large electric power plant
The coal pulverizer of factory.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410851654.5A CN104515591B (en) | 2014-12-31 | 2014-12-31 | A kind of high-precision coal pulverizer vibration detection process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410851654.5A CN104515591B (en) | 2014-12-31 | 2014-12-31 | A kind of high-precision coal pulverizer vibration detection process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104515591A CN104515591A (en) | 2015-04-15 |
| CN104515591B true CN104515591B (en) | 2018-01-05 |
Family
ID=52791240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410851654.5A Active CN104515591B (en) | 2014-12-31 | 2014-12-31 | A kind of high-precision coal pulverizer vibration detection process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104515591B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106597149B (en) * | 2016-11-22 | 2019-01-25 | 电子科技大学 | A kind of oscillator remaining life estimation method based on acceleration sensitivity |
| CN112270419B (en) * | 2020-11-02 | 2024-02-23 | 北京京能能源技术研究有限责任公司 | Grinding roller wear prediction method, grinding roller wear prediction device, computer equipment and readable storage medium |
| CN117610890B (en) * | 2024-01-19 | 2024-04-30 | 天津美腾科技股份有限公司 | Dynamic calculation method, device, equipment and medium for parameters of coal preparation plant |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103048041A (en) * | 2012-12-20 | 2013-04-17 | 北京信息科技大学 | Fault diagnosis method of electromechanical system based on local tangent space and support vector machine |
| CN103400210A (en) * | 2013-08-13 | 2013-11-20 | 广西电网公司电力科学研究院 | Short-term wind-speed combination forecasting method |
| US8725676B1 (en) * | 2011-03-31 | 2014-05-13 | Rockwell Collins, Inc. | State change detection |
| CN103902776A (en) * | 2014-04-02 | 2014-07-02 | 沈阳化工大学 | Wet type ball grinder load parameter integrated modeling method based on EEMD (ensemble empirical mode decomposition) |
-
2014
- 2014-12-31 CN CN201410851654.5A patent/CN104515591B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8725676B1 (en) * | 2011-03-31 | 2014-05-13 | Rockwell Collins, Inc. | State change detection |
| CN103048041A (en) * | 2012-12-20 | 2013-04-17 | 北京信息科技大学 | Fault diagnosis method of electromechanical system based on local tangent space and support vector machine |
| CN103400210A (en) * | 2013-08-13 | 2013-11-20 | 广西电网公司电力科学研究院 | Short-term wind-speed combination forecasting method |
| CN103902776A (en) * | 2014-04-02 | 2014-07-02 | 沈阳化工大学 | Wet type ball grinder load parameter integrated modeling method based on EEMD (ensemble empirical mode decomposition) |
Non-Patent Citations (2)
| Title |
|---|
| Extreme learning machine: Theory and applications;Guang-Bin Huang等;《Neurocomputing》;20060516(第70期);第489-501页 * |
| 基于集合经验模态分解和改进极限学习机的短期风速组合预测研究;张翌晖等;《电力系统保护与控制》;20140516;第42卷(第10期);第29-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104515591A (en) | 2015-04-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Thelen et al. | A comprehensive review of digital twin—part 2: roles of uncertainty quantification and optimization, a battery digital twin, and perspectives | |
| Wang | Towards zero-defect manufacturing (ZDM)—a data mining approach | |
| Rodríguez et al. | The use of artificial neural network (ANN) for modeling the useful life of the failure assessment in blades of steam turbines | |
| CN100416438C (en) | Method and device for monitoring a technical installation comprising several systems, in particular an electric power station | |
| US10318664B2 (en) | Determining life consumption of a mechanical part | |
| CN103970965B (en) | Test run method for accelerated life test of gas turbine engine | |
| Nielsen et al. | A case study on risk-based maintenance of wind turbine blades with structural health monitoring | |
| Tinga et al. | Aligning PHM, SHM and CBM by understanding the physical system failure behaviour | |
| Huynh | A hybrid condition-based maintenance model for deteriorating systems subject to nonmemoryless imperfect repairs and perfect replacements | |
| CN104515591B (en) | A kind of high-precision coal pulverizer vibration detection process | |
| Aydin et al. | Applicability of a fuzzy genetic system for crack diagnosis in Timoshenko beams | |
| CN108204944A (en) | The Buried Pipeline rate prediction method of LSSVM based on APSO optimizations | |
| CN107545110B (en) | Dynamic stress accelerated life test profile compiling method | |
| CN102486833A (en) | Method for predicting efficiency and detecting fault of device | |
| CN110046409A (en) | A kind of components of steam turbine health state evaluation method based on ResNet | |
| Majumder et al. | Real time reliability monitoring of hydro‐power plant by combined cognitive decision‐making technique | |
| CN117252051A (en) | Cable tunnel monitoring and early warning method and system based on digital twinning | |
| Barbieri et al. | Sensor-based degradation prediction and prognostics for remaining useful life estimation: Validation on experimental data of electric motors | |
| Gao et al. | Reliability and availability models of belt drive systems considering failure dependence | |
| Zhang et al. | Condition based maintenance and operation of wind turbines | |
| Sambrekar et al. | Maintenance strategies for realizing Industry 4.0: An overview | |
| CN107956731A (en) | The Forecasting Methodology that the change of Air Cooling Fan for Power Station frequency influences unit back pressuce parameter | |
| CN117391464A (en) | Project progress prediction method and system based on RBF neural network | |
| JP4885008B2 (en) | Maintenance planning method | |
| Ziegler | Assessment of monopiles for lifetime extension of offshore wind turbines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |