CN104070083A - Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method - Google Patents
Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method Download PDFInfo
- Publication number
- CN104070083A CN104070083A CN201410293726.9A CN201410293726A CN104070083A CN 104070083 A CN104070083 A CN 104070083A CN 201410293726 A CN201410293726 A CN 201410293726A CN 104070083 A CN104070083 A CN 104070083A
- Authority
- CN
- China
- Prior art keywords
- elm
- model
- pca
- integrated
- modeling
- 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
Abstract
The invention discloses a model for measuring the rotating speed of the guiding disc of the perforating machine based on an integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method, aiming at the problems that the rotating speed of a guiding disc of an inclined-rolling perforating machine which is additionally provided with a diesel guiding disc can not be directly measured in the process of steel-blank perforation; an original method for measuring the rotating speed of the guiding disc by linear and nonlinear soft-measurement modeling is not ideal. According to the method, a PCA method, an ELM method and an integrated technology and the like applied in the integrated PCA-ELM method are introduced, and the established model is used in predication of the rotating speed of the guiding disc of the perforating machine. Proved by experiment and simulation, the model based on an integrated PCA-ELM algorithm has the advantages that the error in measurement of the rotating speed of the guiding disc of the perforating machine is effectively reduced, the accuracy in predicting the rotating speed of the guiding disc is improved, the model basis is provided for effectively controlling the rotating speed of the guiding disc and improving the production efficiency of a seamless steel pipe, the maintenance cost is low, the instantaneity is good and the reliability and the accuracy are high.
Description
Technical field
The present invention relates to a kind of punch guide disc rev flexible measurement method, its training speed is fast, generalization ability is strong, certainty of measurement is high.
Background technology
Steel pipe has very consequence in national economy, is called industrial blood vessel by people.Along with economic fast development, the use field of seamless steel pipe is in continuous expansion, also more and more higher to the requirement of product quality.The production process of seamless steel pipe mainly contains perforation, tube rolling and tube reducing.China has many iron and steel enterprises to adopt the Mannesmann piercing mill that installs diesel godet additional to carry out perforated steel billit, improves the quality of production of seamless steel pipe.But in Mannesmann piercing mill perforation procedure, due to the restriction of production objective condition, people cannot directly measure guide disc rev fast accurately.Directly measure fast in order to realize it, and then effectively control guide disc rev, enhance productivity, this patent proposes the flexible measurement method based on integrated PCA-ELM, indirectly measures dish rotating speed by other correlated variables.
Summary of the invention
Object of the present invention, be to provide a kind of method of measuring punch guide disc rev, can measure fast and accurately punch guide disc rev, solve due to objective factor, cannot directly fast measure accurately actual guide disc rev, and then cannot realize the problem of the control of guide disc rev.The method model maintenance expense is low, and real-time is good, and precision is high.
For the problem of in the soft measurement of punch guide disc rev, modeling speed being had relatively high expectations, the present invention proposes based on integrated PCA-ELM algorithm measurement punch guide disc rev model.ELM method than traditional neural network model have that model is simple, computational speed is fast, precision high, is relatively applicable to the modeling of online soft-sensing model.But in the soft measurement of guide disc rev, predetermined speed of ELM method and precision need further raising.The problem existing for guide disc rev soft sensor modeling, the present invention has proposed improvement to ELM method.First, extreme learning machine method is combined with PCA method, remove on-the-spot noise and reduce mode input variable number, improve modeling speed and forecast precision; Secondly the data that hand-held scene direct measurement device obtained are introduced, and obtain PCA-ELM soft-sensing model; Again, it is integrated that application integration technology is carried out multi-model, with the repeatedly modeling of data of collection in worksite, to the cumulative summation of the result of each modeling, try to achieve again mean value, complete based on the modeling of integrated PCA-ELM algorithm measurement punch guide disc rev, set up the non-linear relation model between each input quantity and the guide disc rev that affects guide disc rev; Finally utilize presence to answer screw measured value to revise result, obtain exact value.By its accuracy of simulation results show, result shows, the model based on integrated PCA-ELM algorithm has higher recurrence and precision of prediction than traditional ELM model, can meet the requirement of on-the-spot application.
1. forecast model is set up
Obtain the required field data of modeling by PCA, the combination of ELM modeling method, and it is integrated to adopt integrated technology to carry out multi-model, finally obtains integrated PCA-ELM model.
(1) linear PCA method
PCA method is the Multivariable Statistical Methods of a kind of data compression and feature extraction, can effectively remove the correlation between data, reduces the complexity of calculating.Original index is reassembled into one group of new several overall target that have nothing to do mutually by PCA, therefrom extracts the original indication information of several less overall targets reflection as much as possible simultaneously.
Suppose initial data concentrate have
nindividual sample, each sample has
pindividual variable, has so just formed one
n×
pthe sample data matrix on rank, is designated as
X i=[
x i1,
x i2,…,
x ip]
T (1)
If
rfor
xcovariance matrix, ask for
rcharacteristic value
λi (i=1,2 ... and corresponding p)
pindividual Orthogonal Units characteristic vector
e i,
λ 1>=
λ 2>=...>=
λp>=0,
x iindividual principal component can be designated as
Y i=
e i T X=
e i1 X 1+
e i2 X 2+…+
e ip X p (2)
If
y ifor
x 1,
x 2...,
x pin any combination in variance maximum, claim
y 1for first principal component,
y 2be with
y 1incoherent
x 1,
x 2...,
x pall combinations in variance maximum, become Second principal component,, define by that analogy information that each principal component contributor rate refers to certain Principle component extraction and account for the share of total information, contribution rate more illustrates that more greatly the ability of corresponding principal component reflection integrated information is stronger.The
kthe contribution rate of individual principal component is
, generally, if front
mthe accumulation contribution rate of individual principal component
be greater than 85%, use front
mthe combination of individual principal component
y 1,
y 2,
y mreplace initial data
xthereby, the dimension of data is reduced, with in the case of losing the least possible original information, original variable is become to mutual uncorrelated variables.
(2) ELM method
For
nindividual different sample (
x i,
t i),
nindividual hidden node, excitation function is that the unified model of the SLFN of g (x) is
（
j=1,2,…N） (3)
Wherein:
x i=[
x i1,
x i2...
x in]
t∈
r n ,
ti=[
t i1 ,
t i2 ,
t im ]
t∈
r m ,
w ifor connecting the input weights of i hidden node and output node;
β ifor connecting
ithe output weights of individual hidden node and output node;
b ibe
ithe deviation of individual hidden node;
w i.
x irepresent
w iwith
x iinner product; Excitation function g (x) is that sigmoid. formula (3) can be expressed in matrix as
Hβ=T (4)
In formula:
(5).
hbe called the hidden layer output matrix of neutral net.
In the time that excitation function g (x) infinitely can be micro-, network parameter does not need all to adjust, and input connects weights
w isetover with hidden node
b iin the time that starting, training can select at random, and constant in training process.And output connection weights can fetch acquisition by the least square that solves following system of linear equations:
(6)
The least square solution of this linear equation is
(7)
In formula:
hbe called the generalized inverse of hidden layer output matrix H.
ELM algorithm can be summarized as three steps below:
Given training sample set (
x i,
t i) ∈
r n ×
r n ,
i=
1...
n, activation primitive g (x), and hidden node N,
(a) generate at random hidden node parameter (
w i,
b i),
i=
1...,
n;
(b) calculate hidden layer output matrix
h;
(c) computing network optimum is weighed outward
β,
(3) guide disc rev based on integrated PCA-ELM method is measured modeling
Modeling method based on integrated PCA-ELM combines PCA method and extreme learning machine ELM method to carry out modeling, and it is integrated to adopt integrated technology to carry out multi-model, finally obtains integrated PCA-ELM model.Concrete modeling procedure is as follows:
(a) utilize the variable data that affects guide disc rev that PCA method records scene to screen:
1) the variable sampled data matrix under procurement process normal operating condition, and it is carried out to standardization obtain matrix
x;
2) according to the data matrix after standardization
xcalculate covariance matrix
r;
3) according to covariance matrix
robtain characteristic value, principal component contributor rate and accumulative total variance contribution ratio, determine principal component number, institute's pivoting number should make contribution rate of accumulative total reach 80% ~ 95%.
4) ask square prediction error, accurately choose pivot number.
5) according to (4), (5) definite pivot number
p, determine load matrix
p.Make score matrix
t=XP,
tbe the learning sample of setting up ELM model.
(b) set up ELM network model.The influence factor historical data that guide disc rev is exerted an influence, after step (a) analyzing and processing, is obtained to reconstruct data sample, set up ELM network model.Then through ELM algorithm, these data are trained.Given training sample set (
x i,
t i) ∈
r n ×
r n ,
i=
1...
n, activation primitive g (x), and hidden node N.
1) be input weights
w iand threshold value
bi random assignment,
i=
1...,
n;
2) calculate matrix according to formula (3)
h;
3) calculate output weights
β,
.
(c) data of collection in worksite are done to five modelings according to step (1), step (2), adopt integrated technology by build for five times model integrated, to result averaged, finally obtain integrated PCA-ELM model, integrated PCA-ELM model modeling procedure chart is as shown in Fig. 1.
(d) utilize presence to answer the measured value of screw to proofread and correct institute's established model.Answering as presence the guide disc rev that screw records is n, the guide disc rev that the integrated PCA-ELM model of building obtains is m, both differences are n-m, be t the update time of model, within the t time, the guide disc rev value that built integrated PCA-ELM model is obtained all adds/subtracts difference n-m, completes the correction to institute's established model.
2. punch guide disc rev modeling experiment
Certain steel mill's piercing mill of seamless steel tube group creation data, comprises 16 variablees that affect guide disc rev, and each variable is totally 10000 groups of data.Data are divided into two groups: first 5000 groups are used for setting up punch godet and turn and read model, latter 5000 groups are used for inspection institute's established model, check its precision of prediction to guide disc rev.Application MATLAB software, by programming, sets up respectively ELM method model, PCA-ELM method model, integrated PCA-ELM method model, and predicting the outcome of three kinds of methods compared.Fig. 2 is true output and the network output comparison diagram of ELM Method Modeling data, Fig. 3 is the true output of ELM method check data and network output comparison diagram, Fig. 4 is true output and the network output comparison diagram of PCA-ELM Method Modeling data, Fig. 5 is the true output of PCA-ELM method check data and network output comparison diagram, Fig. 6 is true output and the network output comparison diagram of integrated PCA-ELM Method Modeling data, and Fig. 7 is the true output of integrated PCA-ELM method check data and network output comparison diagram.Table 1 is the error comparison sheet of three kinds of method institute established model modeling datas and check data.By more known, integrated PCA-ELM Method Modeling is for the precision of prediction of prediction punch guide disc rev, and error is less.
The error comparison sheet of three kinds of method institute established model modeling datas of table 1 and check data
Extreme learning machine method is combined with PCA method, remove on-the-spot noise and reduce mode input variable number, improve modeling speed and forecast precision; The data of secondly presence being answered screw directly measure are introduced, and obtain PCA-ELM soft-sensing model; Finally, for improving stability, it is integrated that application integration technology is carried out multi-model, with the repeatedly modeling of data of collection in worksite, to the cumulative summation of the result of each modeling, try to achieve again mean value, complete based on the modeling of integrated PCA-ELM algorithm measurement punch guide disc rev, set up the non-linear relation model between each input quantity and the guide disc rev that affects guide disc rev.By its accuracy of simulation results show, result shows, the model based on integrated PCA-ELM algorithm has higher recurrence and precision of prediction than traditional ELM model, can meet the requirement of on-the-spot application.
3. the invention has the advantages that
(1) adopt PCA method to carry out pretreatment to field data, remove on-the-spot noise and reduce mode input variable number, guide disc rev soft sensor modeling speed is faster and precision of prediction is higher.
(2) adopt extreme learning machine ELM method to carry out modeling, because whole learning process once completes, pace of learning is fast, and mass data shows that ELM method Generalization Capability is good, extreme learning machine method is combined with PCA method, improve modeling speed and forecast precision, therefore, in guide disc rev soft sensor modeling process, PCA-ELM method is better than traditional E LM method.
(3) it is integrated that employing integrated technology carries out multi-model, with the repeatedly modeling of data of collection in worksite, to the cumulative summation of the result of each modeling, then tries to achieve mean value, and less error, reaches better prediction effect.
Simulation results show measure feasibility and the validity of punch guide disc rev based on integrated PCA-ELM method.
brief description of the drawings:
Fig. 1 is integrated PCA-ELM model modeling procedure chart
Fig. 2 is true output and the network output comparison diagram of ELM Method Modeling data
Fig. 3 is true output and the network output comparison diagram of ELM method check data
Fig. 4 is true output and the network output comparison diagram of PCA-ELM Method Modeling data
Fig. 5 is true output and the network output comparison diagram of PCA-ELM method check data
Fig. 6 is true output and the network output comparison diagram of integrated PCA-ELM Method Modeling data
Fig. 7 is true output and the network output comparison diagram of integrated PCA-ELM method check data.
Claims (2)
1. for the problem of in the soft measurement of punch guide disc rev, modeling speed being had relatively high expectations, the present invention proposes that to have model based on integrated PCA-ELM algorithm measurement punch guide disc rev model .ELM method than traditional neural network model simple, computational speed is fast, precision high, relatively be applicable to the modeling of online soft-sensing model. but in the soft measurement of guide disc rev, predetermined speed of ELM method and precision need further raising. the problem existing for guide disc rev soft sensor modeling, the present invention has proposed improvement to ELM method. first, extreme learning machine method is combined with PCA method, remove on-the-spot noise and reduce mode input variable number, improve modeling speed and forecast precision, secondly the data that hand-held scene direct measurement device obtained are introduced, and obtain PCA-ELM soft-sensing model, again, it is integrated that application integration technology is carried out multi-model, with the repeatedly modeling of data of collection in worksite, to the cumulative summation of the result of each modeling, try to achieve again mean value, complete based on the modeling of integrated PCA-ELM algorithm measurement punch guide disc rev, set up the non-linear relation model between each input quantity and the guide disc rev that affects guide disc rev, finally utilize presence to answer screw measured value to revise result, obtain exact value. by its accuracy of simulation results show, result shows, the model based on integrated PCA-ELM algorithm has higher recurrence and precision of prediction than traditional ELM model, can meet the requirement of on-the-spot application.
2. the modeling method based on integrated PCA-ELM combines PCA method and extreme learning machine ELM method to carry out modeling, and it is integrated to adopt integrated technology to carry out multi-model, finally obtains integrated PCA-ELM model. and concrete modeling procedure is as follows:
(1) utilize the variable data that affects guide disc rev that PCA method records scene to screen.
(2) set up ELM network model. by the influence factor historical data that guide disc rev is exerted an influence after step (a) analyzing and processing, obtain reconstruct data sample, set up ELM network model. then through ELM algorithm, these data are trained.
(3) data of collection in worksite are done to five modelings according to step (1), step (2), adopt integrated technology by build for five times model integrated, to result averaged, finally obtain integrated PCA-ELM model.
(4) utilize presence to answer the measured value of screw to proofread and correct institute's established model. answering as presence the guide disc rev that screw records is n, the guide disc rev that the integrated PCA-ELM model of building obtains is m, both differences are n-m, be t the update time of model, within the t time, the guide disc rev value that built integrated PCA-ELM model is obtained all adds/subtracts difference n-m, completes the correction to institute's established model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410293726.9A CN104070083A (en) | 2014-06-27 | 2014-06-27 | Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410293726.9A CN104070083A (en) | 2014-06-27 | 2014-06-27 | Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104070083A true CN104070083A (en) | 2014-10-01 |
Family
ID=51591916
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410293726.9A Pending CN104070083A (en) | 2014-06-27 | 2014-06-27 | Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104070083A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354347A (en) * | 2015-07-15 | 2016-02-24 | 东北大学 | Magnesite ore grade classification modeling method |
| CN107367936A (en) * | 2017-07-31 | 2017-11-21 | 广东工业大学 | Piezoelectric ceramic actuator modeling, control method and system based on OS ELM |
| CN107390546A (en) * | 2017-07-31 | 2017-11-24 | 广东工业大学 | Piezoelectric Driving locating platform modeling method, control method and system based on EOS ELM |
| CN107505840A (en) * | 2017-07-31 | 2017-12-22 | 广东工业大学 | Piezoelectric Driving FTS modeling methods, control method and system based on FReOS ELM |
| CN107944182A (en) * | 2017-12-08 | 2018-04-20 | 东北大学 | Determination methods, electronic equipment and the computer-readable storage medium of corner track failure |
| CN110018322A (en) * | 2019-04-18 | 2019-07-16 | 北京先见智控科技有限公司 | A kind of Rotating speed measring method and system based on deep learning |
| CN111031558A (en) * | 2019-12-23 | 2020-04-17 | 安徽理工大学 | Mobile traffic prediction method and device, computer equipment and storage medium |
| CN113486587A (en) * | 2021-07-06 | 2021-10-08 | 太原科技大学 | Seamless steel pipe perforation process model parameter prediction system based on Matlab |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110106734A1 (en) * | 2009-04-24 | 2011-05-05 | Terrance Boult | System and appartus for failure prediction and fusion in classification and recognition |
| CN102708381A (en) * | 2012-05-09 | 2012-10-03 | 江南大学 | Improved extreme learning machine combining learning thought of least square vector machine |
| CN103413050A (en) * | 2013-08-20 | 2013-11-27 | 北京工业大学 | Motor imagery electroencephalogram voting strategy sorting method based on extreme learning machines |
| CN103745093A (en) * | 2013-12-25 | 2014-04-23 | 中国矿业大学 | Principal component analysis-extreme learning machine (PCA-ELM) based coal mine water inrush prediction method |
-
2014
- 2014-06-27 CN CN201410293726.9A patent/CN104070083A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110106734A1 (en) * | 2009-04-24 | 2011-05-05 | Terrance Boult | System and appartus for failure prediction and fusion in classification and recognition |
| CN102708381A (en) * | 2012-05-09 | 2012-10-03 | 江南大学 | Improved extreme learning machine combining learning thought of least square vector machine |
| CN103413050A (en) * | 2013-08-20 | 2013-11-27 | 北京工业大学 | Motor imagery electroencephalogram voting strategy sorting method based on extreme learning machines |
| CN103745093A (en) * | 2013-12-25 | 2014-04-23 | 中国矿业大学 | Principal component analysis-extreme learning machine (PCA-ELM) based coal mine water inrush prediction method |
Non-Patent Citations (3)
| Title |
|---|
| 严东等: "基于特征提取和极限学习机的软测量方法", 《控制工程》, vol. 20, no. 1, 20 January 2013 (2013-01-20), pages 55 - 58 * |
| 何群等: "基于PCA和多变量极限学习机的轴承剩余寿命预测", 《中国机械工程》, vol. 25, no. 7, 2 April 2014 (2014-04-02), pages 984 - 989 * |
| 肖冬等: "基于改进PCA-ELM方法的穿孔机导盘转速测量", 《控制理论与应用》, vol. 27, no. 01, 15 January 2010 (2010-01-15), pages 19 - 24 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354347A (en) * | 2015-07-15 | 2016-02-24 | 东北大学 | Magnesite ore grade classification modeling method |
| CN105354347B (en) * | 2015-07-15 | 2018-06-29 | 东北大学 | A kind of magnesite ore grade classification model construction method |
| CN107367936A (en) * | 2017-07-31 | 2017-11-21 | 广东工业大学 | Piezoelectric ceramic actuator modeling, control method and system based on OS ELM |
| CN107390546A (en) * | 2017-07-31 | 2017-11-24 | 广东工业大学 | Piezoelectric Driving locating platform modeling method, control method and system based on EOS ELM |
| CN107505840A (en) * | 2017-07-31 | 2017-12-22 | 广东工业大学 | Piezoelectric Driving FTS modeling methods, control method and system based on FReOS ELM |
| CN107944182A (en) * | 2017-12-08 | 2018-04-20 | 东北大学 | Determination methods, electronic equipment and the computer-readable storage medium of corner track failure |
| CN110018322A (en) * | 2019-04-18 | 2019-07-16 | 北京先见智控科技有限公司 | A kind of Rotating speed measring method and system based on deep learning |
| CN111031558A (en) * | 2019-12-23 | 2020-04-17 | 安徽理工大学 | Mobile traffic prediction method and device, computer equipment and storage medium |
| CN113486587A (en) * | 2021-07-06 | 2021-10-08 | 太原科技大学 | Seamless steel pipe perforation process model parameter prediction system based on Matlab |
| CN113486587B (en) * | 2021-07-06 | 2022-08-05 | 太原科技大学 | Seamless steel pipe perforation process model parameter prediction system based on Matlab |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104070083A (en) | Method for measuring rotating speed of guiding disc of perforating machine based on integrated PCA-ELM (Principal Component Analysis)-(Extrem Learning Machine) method | |
| CN108764568B (en) | Data prediction model tuning method and device based on LSTM network | |
| CN112365029B (en) | Missing value processing method for air conditioner load prediction and air conditioner load prediction system | |
| CN111241657A (en) | Cold-rolled strip steel plate shape prediction method based on deep learning | |
| Truong et al. | Design of an advanced time delay measurement and a smart adaptive unequal interval grey predictor for real-time nonlinear control systems | |
| CN104317269A (en) | Comprehensive forecast and iterative learning control method based on 2D theory | |
| CN110866640A (en) | Power load prediction method based on deep neural network | |
| CN110458601B (en) | Method and device for processing resource data, computer equipment and storage medium | |
| CN103106535A (en) | Method for solving collaborative filtering recommendation data sparsity based on neural network | |
| CN107578093A (en) | The Elman neural network dynamic Forecasting Methodologies of Landslide Deformation | |
| CN103268525A (en) | WD-RBF (wavelet denoising-radial basis function)-based analogue prediction method of hydrological time sequence | |
| CN108460462A (en) | A kind of Interval neural networks learning method based on interval parameter optimization | |
| CN103106331A (en) | Photo-etching line width intelligence forecasting method based on dimension-reduction and quantity-increment-type extreme learning machine | |
| CN104330972A (en) | Comprehensive prediction iterative learning control method based on model adaptation | |
| Wan Ahmad et al. | Arima model and exponential smoothing method: A comparison | |
| CN103279030B (en) | Dynamic soft measuring modeling method and device based on Bayesian frame | |
| CN105069214B (en) | A kind of reliability of technology appraisal procedure based on nonlinear correlation analysis | |
| CN104102131B (en) | The linear quadratic fault tolerant control method of the batch process that Infinite horizon optimizes | |
| CN106683001A (en) | Thermal power plant set identification data selection method based on historical operation data | |
| CN104165711A (en) | Verification table temperature measuring method based on Kalman filtering | |
| Zhu et al. | Prediction of air quality index based on wavelet transform combination model | |
| CN110991641B (en) | Oil reservoir type analysis method and device and electronic equipment | |
| CN109376957B (en) | Method for predicting load of thermal power plant | |
| Wu et al. | Prediction of energy consumption time series using Neural Networks combined with exogenous series | |
| CN102393628A (en) | Dynamic matrix control (DMC) engineering method based on model simplification and prediction error correction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141001 |