CN109710661A - Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis - Google Patents
Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis Download PDFInfo
- Publication number
- CN109710661A CN109710661A CN201811571162.5A CN201811571162A CN109710661A CN 109710661 A CN109710661 A CN 109710661A CN 201811571162 A CN201811571162 A CN 201811571162A CN 109710661 A CN109710661 A CN 109710661A
- Authority
- CN
- China
- Prior art keywords
- pressure heater
- data
- wavelet
- indicate
- simulated annealing
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 57
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 53
- 230000002068 genetic effect Effects 0.000 title claims abstract description 43
- 238000002922 simulated annealing Methods 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims description 68
- 238000000137 annealing Methods 0.000 claims abstract description 10
- 230000008859 change Effects 0.000 claims description 29
- 239000011159 matrix material Substances 0.000 claims description 29
- 230000008569 process Effects 0.000 claims description 17
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 17
- 230000009466 transformation Effects 0.000 claims description 15
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 5
- 238000009825 accumulation Methods 0.000 claims description 4
- 238000001816 cooling Methods 0.000 claims description 4
- 230000010429 evolutionary process Effects 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 230000005236 sound signal Effects 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 10
- 230000007547 defect Effects 0.000 abstract description 6
- 230000036632 reaction speed Effects 0.000 abstract description 4
- 230000000875 corresponding effect Effects 0.000 description 9
- 238000012423 maintenance Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 6
- 238000005259 measurement Methods 0.000 description 5
- 230000005611 electricity Effects 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002209 hydrophobic effect Effects 0.000 description 2
- 238000011068 loading method Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- FNMKZDDKPDBYJM-UHFFFAOYSA-N 3-(1,3-benzodioxol-5-yl)-7-(3-methylbut-2-enoxy)chromen-4-one Chemical compound C1=C2OCOC2=CC(C2=COC=3C(C2=O)=CC=C(C=3)OCC=C(C)C)=C1 FNMKZDDKPDBYJM-UHFFFAOYSA-N 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000009172 bursting Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000000571 coke Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005474 detonation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000001771 impaired effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000000491 multivariate analysis Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011426 transformation method Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This application provides one kind based on Global Genetic Simulated Annealing Algorithm to high-pressure heater state analysis method, the data of high-pressure heater operating status are obtained first, and the processing of data human error and the progress accidental error processing of the data problem as caused by unstable environment or unstable instrument and signal are carried out to the data problem as caused by equipment heat itself；The primary variables for influencing generator operation is found using the analysis of main elements, reduces the data of calculating, improves operation reaction speed, convenient for selecting representative measuring point；It is excavated based on running state data of the genetic annealing algorithms to high-pressure heater, finally calculate the degree of membership of high-pressure heater and obtains the optimal operational condition of high-pressure heater.The present invention is based on Global Genetic Simulated Annealing Algorithms to the FCM cluster result of high-pressure heater running state data, which had not only overcome FCM algorithm to fall into the defect of locally optimal solution, but also strengthened global search function, to have better convergence and ability of searching optimum.
Description
Technical field
This application involves high-pressure heater running state analysis technical fields, more particularly to one kind to be based on genetic simulated annealing
Method of the algorithm to high-pressure heater state analysis.
Background technique
Power industry is built upon the high concentration on the basis of modern electric energy conversion, transmission, distribution science and technology
The big industry of socialization, is supplied with the basic activity of the national economy energy, and is related to the public utilities of the living standards of urban and rural residents.
Power generation, transmission of electricity, distribution and user form a unified power system operation, any one link breaks down, can all influence
To the safe and stable operation of entire power grid.Currently, the power supply architecture of diversification and the method for operation of complexity are to electricity net safety stable
The new demand that the technology supervision of operation proposes.Wherein, unit, equipment safe and stable operation directly affect the stabilization of power grid
Operation level.
High-pressure heater is one of power generation main auxiliary equipment of fired power generating unit, for a long time the high pressure under high-temperature high-pressure state
Heater and system operation, bypass changeover, the factors such as water supply failure of pump, unit load mutation can all run larger shadow to it
It rings, high-pressure heater and the system failure is caused frequently to occur.High-pressure heater, which breaks down, may cause steam turbine water to rush
The safety of boiler operatiopn is hit, reduced, the efficiency of cycle is reduced, reduces economy and cause casualties.High-pressure heater hair
Raw failure, which has become, is only second to boiler tube bursting, influences one of the main reason for unit expires detonation hair.In the prior art, power plant is general
The operating status of high-pressure heater is monitored all over using the type of alarm of maximin.Based on repair time interval
Equipment repair and maintenance operation simultaneously formulates maintenance plan, maintenance mode, after equipment stops working operation or cancel closedown accident
Subsequent maintenance mode.
But electricity power enterprise is when in view of improving unit reliability and power generation hourage, reduction testing cost and maintenance
Between under conditions of, common maintenance mode can not in real time be monitored high-pressure heater, can not equipment early warning be timely existing
Field high-pressure heater and operation maintenance personnel propose the foundation of high-pressure heater operating status, can not remind operation maintenance personnel
Whether should monitoring dynamics to high-pressure heater reinforce, can not find the hidden fault of high-pressure heater in time, so as to
It is impaired that high-pressure heater can be will appear, entire network system is unable to operate stably, and electricity power enterprise's safety in production is obstructed, economy effect
Benefit can not ensure or even the significant consequences of casualties.Therefore, it is necessary to the operating status progress in real time to high-pressure heater
Analysis and early warning, but existing analysis method can not overcome the defect for falling into locally optimal solution again, not have global search function,
Good convergence and ability of searching optimum.
Summary of the invention
This application provides a kind of based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, to solve
The defect for falling into locally optimal solution can not be overcome, do not have global search function, good convergence and ability of searching optimum
Technical problem.
A method of based on Global Genetic Simulated Annealing Algorithm to high-pressure heater state analysis, the method includes following steps
It is rapid:
Step S100: high-pressure heater running state data is obtained；
Step S200: human error in running state data acquisition is handled using Pauta criterion；
Step S300: using the wavelet threshold denoising method of the smooth unbiased possibility predication of history to random in running state data acquisition
Error is handled；
Step S400: it is selected using operating status of the principle component analysis to high-pressure heater；
Step S500: FCM cluster result is carried out using operating status of the Global Genetic Simulated Annealing Algorithm to high-pressure heater；
Step S600: calculating the degree of membership of high-pressure heater and obtains the optimal operational condition of high-pressure heater.
Further, the high-pressure heater running state data, including main feedwater flow, extraction temperature, feed pressure,
Extraction pressure, the exit water temperature of high-pressure heater, inlet water temperature, water level, drain temperature, upper end difference and lower end are poor.
Further, described that human error in running state data acquisition is handled using Pauta criterion, pass through
Observed quantity running state data change rate size judges whether there is human error,
The running state data inscribed when current time, x1, x2 ..., xn and each expression of i >=2 is indicated with i, with Δ xi=
xi-xi-1Indicate two neighboring moment status data change rate；WithIndicate the variation of observed quantity running state data
The mean value of rate；WithIndicate the standard deviation of observed quantity running state data change rate；
If the absolute value when adjacent two moment observed quantity status data change rate is greater than 3 σi, | Δ xi| 3 σ of >i, current time
There are human errors for the status data of observed quantity, and the human error of the current time status data is deleted and corrected.
Further, the human error of the current time status data is deleted and is corrected, and is included the following steps:
Calculate the standard deviation sigma of current time observed quantity status data change ratei；
Judgement | Δ xi| with 3 σiSize, as | Δ xi| 3 σ of <i, then it is the status data of normal observation, when | Δ xi| >
3σi, observing the status data that refers to, there are human errors；
Current time observation state data are human errors, to Δ xiPositive negative judgement；If Δ xi> 0, then use xi=
xi-1+σiIt corrects and replaces the human error；If Δ xi< 0, then use xi=xi-1-σiIt corrects and replaces the human error.
Subsequent time observation state data adopt into repeating the above process.
Further, the wavelet threshold denoising method using the smooth unbiased possibility predication of history is in running state data acquisition
Random error is handled,
The step of Wavelet-denoising Method includes:
Wavelet transformation is carried out to f containing noise signal (x)=s (x)+n (x), wherein s (x) indicates original signal, n (x) table
Show noise signal；
Wavelet coefficient on each decomposition scale of threshold process obtains and estimates new wavelet coefficient wj,x, so thatTo the greatest extent
It is possible small；
To estimation wavelet coefficient wj,xCarry out wavelet reconstruction, the estimation signal after being denoised
The step of wavelet threshold denoising method includes:
Collected original signal is subjected to wavelet transformation, obtains the wavelet coefficient of each decomposition level；
If f (t) ∈ L2(R), wavelet transformation is indicated with following formula,
Wherein a indicates scale factor, and b indicates translation
The factor,Indicate analysis wavelet；
Threshold process is carried out to wavelet coefficient, obtains new wavelet coefficient；
You is carried out to new wavelet coefficient to convert, and obtains reconstruction signal；
The step of history smooth unbiased possibility predication includes:
The estimated value of actual signal is calculated according to original signal, and make actual signal estimated value and actual signal it is square
It is poor minimum；
Wherein, the original signal containing noise signal is indicated with X=[x0, x1 ..., xN-1] T, formula is
xi=si+ei, (i=0,1 ..., N-1), siIndicate the true signal value at i moment, eiIndicate the noise figure at i moment,
N indicates signal length；
It can be obtained using mean square deviation replacement for mathematic expectaion:
Wherein, X indicates original signal, and s indicates actual signal,Indicate the estimated value of actual signal s.
Further, the wavelet basis that wavelet transformation and inversion are used instead be all it is orthogonal, mean square deviation can be indicated with following formula:
Wherein, wj, k indicate actual signal s k-th of wavelet coefficient on j-th of decomposition level；Indicate withIt is corresponding
Wavelet coefficient.
Further, the analysis of main elements simplifies the data of high-pressure heater operating status；
Comparison obtains the primary process variable influenced in former variable on operating status；
The analysis of main elements includes:
Observation state data are standardized, to remove shadow of the difference to original variable of the dimension order of magnitude
It rings, variation is normalized to observation state data, indicate each variable after normalization with X=[x1, x2 ..., xp], meet
E (xi)=0, Var (xi)=1, i=1,2 ..., p；
Covariance matrix after calculating each variable indicates covariance matrix with S；
Calculate characteristic value and feature vector；The characteristic value of solution S and in magnitude order 1 >=λ of λ 2 >=... p >=0 λ, feature
Vector normal orthogonal turns to e1, e2 ..., ep,(j=1,2 ..., p) indicates j-th of pivot of x, and former variable is logical
Cross the transformed main former score of characteristic variable；
Accumulation contribution rate is calculated, the variance of p is indicated with λ j,Indicate contribution rate,It indicates
Preceding k pivot accumulates contribution rate.
Further, the Global Genetic Simulated Annealing Algorithm to the high-pressure heater running state data of acquisition carry out emulation and
Analysis, is calculated the degree of membership of high-pressure heater and the optimal operational condition of high-pressure heater.
Further, the Global Genetic Simulated Annealing Algorithm includes:
Control parameter initialization.N is population scale；Pm is mutation probability；T is annealing initial temperature；α is cooling for temperature
Parameter；
The initial system of solutions generates at random；
Step is executed as follows to the system of solutions of generation, until next-generation heredity comes out；Each of group individual is assessed
Its fitness value f (xi) (i=1,2,, N)；It randomly chooses two individual xi and xj and carrys out crisscross inheritance, and generate two newly
Then individual xi' and xj' evaluates the fitness value f (xi') and f (xj') of two new individuals, then connects according to certain probability
By new individual；Hereditary variation is executed after individual intersection heredity, judges whether to receive the new explanation after variation；
If Tk+1=α TK, step is gone back to, or if meeting the condition of convergence, evolutionary process terminates.
The beneficial effect of the application is:
From the above technical scheme, this application provides one kind to be based on Global Genetic Simulated Annealing Algorithm to high-pressure heater shape
The method of state analysis obtains the data of high-pressure heater operating status first, and asks the data as caused by equipment heat itself
Topic carries out the processing of data human error and the data problem as caused by unstable environment or unstable instrument and signal carries out
Accidental error processing；The primary variables for influencing generator operation is found using the analysis of main elements, is reduced the data of calculating, is mentioned
Height operation reaction speed, convenient for selecting representative measuring point；Based on genetic annealing algorithms to the operation shape of high-pressure heater
State data are excavated, and are finally calculated the degree of membership of high-pressure heater and are obtained the optimal operational condition of high-pressure heater.This
Based on Global Genetic Simulated Annealing Algorithm to the FCM cluster result of high-pressure heater running state data, which was both overcome for invention
FCM algorithm falls into the defect of locally optimal solution, and strengthens global search function, to be provided with better convergence and complete
Office's search capability.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is the method flow diagram that the application analyzes high-pressure heater state based on genetic annealing algorithms.
Specific embodiment
Here embodiment will be illustrated in detail, the example is illustrated in the accompanying drawings.In the following description when referring to the accompanying drawings,
Unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Implement described in following embodiment
Mode does not represent all embodiments consistent with the application.
Referring to Fig. 1, the method flow diagram that high-pressure heater state is analyzed based on genetic annealing algorithms for the application.
A method of based on Global Genetic Simulated Annealing Algorithm to high-pressure heater state analysis, the method includes following steps
It is rapid:
Step S100: high-pressure heater running state data is obtained；
Step S200: human error in running state data acquisition is handled using Pauta criterion；
Step S300: using the wavelet threshold denoising method of the smooth unbiased possibility predication of history to random in running state data acquisition
Error is handled；
Step S400: it is selected using operating status of the principle component analysis to high-pressure heater；
Step S500: FCM cluster result is carried out using operating status of the Global Genetic Simulated Annealing Algorithm to high-pressure heater；
Step S600: calculating the degree of membership of high-pressure heater and obtains the optimal operational condition of high-pressure heater.
Step S100 specifically:
High-pressure heater status data is the various letters for representing state that high temperature heater (HTH) generates during operating status
Number, these status representative signals contain the useful status information of high-pressure heater, also contain interference information.Above-mentioned interference
Information has an impact the reliability and precision of the operating status of high-pressure heater.Therefore, first to high-pressure heater status number
According to being pre-processed, consider to obtain the accuracy of measurement result obtained in high-pressure heater status data, consistency and complete
Property.
Further, the high-pressure heater running state data, including main feedwater flow, extraction temperature, feed pressure,
Extraction pressure, the exit water temperature of high-pressure heater, inlet water temperature, water level, drain temperature, upper end difference and lower end are poor.
Step S200 specifically:
There are measurement error in obtaining the measurement result that high-pressure heater status data obtains, measurement error includes fault
Error and random error.Out-of-date error includes process leakage, equipment fault, measuring instrumentss zero drift phenomenon, it is serious when further include
Measuring instrumentss failure etc..When generating human error, can quality to measurement result and state analysis generation seriously affect.
The detection method of human error is commonly used specifically, judging human error according to the change rate size of observed quantity.Root
The method that human error is judged according to change rate is performed an analysis to past time series behavioral characteristics, and threshold value is manually to set
, if the change rate in current time observed quantity at moment with before is not less than threshold value, typical time interval is set as 1s,
Then the observed quantity at current time is exactly a human error.Threshold value is set it should be understood that the amount of being observed historical data feature.
When observed quantity quantity is greater than 185, using Pauta criterion, it is assumed that one group of detection data contains only random error,
Standard deviation is obtained to its calculation processing, according to one section of certain determine the probability, it is believed that as long as being more than the mistake in this section
Difference is not just random error but human error, should reject the data containing the error.Using Pauta criterion and commonly use
Detection human error in conjunction with coming to observed quantity change rate method of inspection.
It is described that human error in running state data acquisition is handled using Pauta criterion, it is run by observed quantity
Status data change rate size judges whether there is human error,
The running state data inscribed when current time, x1, x2 ..., xn and each expression of i >=2 is indicated with i, with Δ xi=
xi-xi-1Indicate two neighboring moment status data change rate；
WithIndicate the mean value of observed quantity running state data change rate；
WithIndicate the standard deviation of observed quantity running state data change rate；
If the absolute value when adjacent two moment observed quantity status data change rate is greater than 3 σi, | Δ xi| 3 σ of >i, current time
There are human errors for the status data of observed quantity, and the human error of the current time status data is deleted and corrected.
To be combined with Pauta criterion by the existing method for detecting human error using change rate.Pass through judgement
The standard deviation of the absolute value of adjacent moment observed quantity status data change rate and 3 times of observed quantity running state data change rate
It compares, if the former is greater than the latter, the status data for obtaining current time observed quantity is human error.Using the above method
It avoids, the existing method that human error is detected using change rate, it is also necessary to by analyze really to historical data feature
Threshold value is determined, thus the method for simplifying human error processing.
Step S300 specifically:
Random error be influenced by unstable enchancement factor, such as unstable environmental condition, unstable instrument and
The noise etc. that signal is mingled with, almost without from avoiding, but there are certain statistical laws.Wavelet noise-eliminating method is a kind of based on small
The coefficients model denoising method of wave conversion, the corresponding wavelet coefficient of noise is Uniformly distributed, and quantity is more, but absolute value compared with
It is small.And the negligible amounts of the corresponding wavelet coefficient of echo signal, but absolute value is larger.Using the small of the smooth unbiased possibility predication of history
Wave Threshold denoising handles random error in running state data acquisition.
The wavelet threshold denoising method using the smooth unbiased possibility predication of history is to random error in running state data acquisition
It is handled,
The step of Wavelet-denoising Method includes:
Wavelet transformation is carried out to f containing noise signal (x)=s (x)+n (x), wherein s (x) indicates original signal, n (x) table
Show noise signal；
Wavelet coefficient on each decomposition scale of threshold process obtains and estimates new wavelet coefficient wj,x, so thatTo the greatest extent
It is possible small；
To estimation wavelet coefficient wj,xCarry out wavelet reconstruction, the estimation signal after being denoised
The step of wavelet threshold denoising method includes:
Collected original signal is subjected to wavelet transformation, obtains the wavelet coefficient of each decomposition level；
If f (t) ∈ L2(R), wavelet transformation is indicated with following formula,
Wherein a indicates scale factor, and b indicates translation
The factor,Indicate analysis wavelet；
When the Fourier transformation of ψ (t)Meet:When, ψ (t) is wavelet mother function condition
It sets up.
Wavelet transformation is original function f (t) and analysis waveletInner product, obtain original function after wavelet transformation and analyzing
Projection on small echo.If the characteristic information of original function is all more prominent on all scales, it is necessary to the maximum extent will be former
On Function Projective to analysis wavelet, it is desirable that the two has the waveform of higher similarity.Selection wavelet basis function is needing to consider it just
The property handed over, compact sup-port, symmetry, flatness etc..
Threshold process is carried out to wavelet coefficient, obtains new wavelet coefficient；
You is carried out to new wavelet coefficient to convert, and obtains reconstruction signal；
The step of history smooth unbiased possibility predication includes:
The estimated value of actual signal is calculated according to original signal, and make actual signal estimated value and actual signal it is square
It is poor minimum；
Wherein, the original signal containing noise signal is indicated with X=[x0, x1 ..., xN-1] T, formula is
xi=si+ei, (i=0,1 ..., N-1), siIndicate the true signal value at i moment, eiIndicate the noise figure at i moment,
N indicates signal length；
It can be obtained using mean square deviation replacement for mathematic expectaion:
Wherein, X indicates original signal, and s indicates actual signal,Indicate the estimated value of actual signal s.
Biggish absolute value coefficient retained or shunk, and the lesser absolute value coefficient of zero setting obtains estimation wavelet systems
Number, signal are reconstructed by estimation wavelet coefficient, realize signal denoising.
Further, the wavelet basis that wavelet transformation and inversion are used instead be all it is orthogonal, mean square deviation can be indicated with following formula:
Wherein, wj, k indicate actual signal s k-th of wavelet coefficient on j-th of decomposition level；Indicate withIt is corresponding
Wavelet coefficient.
Step S400 specifically:
Principle component analysis is to be established one group of new hidden comprehensive variable based on initial data variable and become to reduce initial data
The dimension of amount simplifies complexity, and Main change information is extracted from new projector space again, and obtains statistical nature, from
And reinforce understanding the variable characteristics of initial data.The application is using principle component analysis to the big of high-pressure heater operating status
Data reduction is measured, comparison obtains the primary process variable influenced in former variable on operating status.
Principle component analysis (PCA), also known as Principal Component Analysis are a kind of multivariate statistics disobeyed and be against accurate mathematical model
Analysis method obtains the mutually indepedent variable space of low-dimensional, mentions by carrying out dimensionality reduction projection process higher-dimension correlated variables space
The data characteristics of complex process is taken, and the principal component model of respective process is established.Principal component model becomes to retain characterize data
Different Main way and abandon part residual error, achieved the purpose that eliminate system interference and extraction system information.
Sample space is that have data array constructed by M sample value of N number of variable of correlativity, passes through pivot
Analytic approach establishes less comprehensive variable, can closely reflect the change information contained in original N number of variable.Base
This method is the primary and secondary position to determine change direction, to compare the variance size of data variation, is obtained by primary and secondary arrangement
Each pivot, these be all each other independent pivot.With the help of principle component analysis, change information can be extracted to drop
The complexity of low data analysis.
If the matrix X of m*n is raw data matrix, m sampled value (or sample) and n variable are indicated, these numbers
According to correlation may be it is very high, then result in covariance matrix and approach singular matrix, this is that general numerical method can not be gone point
Analysis.The purpose of principle component analysis seeks to building K (K < n) a new variable, and as far as possible can ground make original letter of n variable
Breath is kept, and pivot is exactly this K new variables, they are mutually independent two-by-two.Pivot analysis is by mentioning to sample information
It takes and determines obtained some main shafts (main shaft on a multiple normal distribution isodensity ellipsoid).This is the classical theory of statistics
Viewpoint.When data matrix dimension is more, the reduction of this variable quantity can undoubtedly mitigate the difficulty of problem analysis, reduce
Calculation step improves efficiency.
Principle component analysis (PCA) is a kind of common feature extracting method, has and eliminates data dependence, reduces data dimension
The excellent performances such as number.Assuming that X ∈ Rm*nFor historical data sample obtained, wherein m indicates the number of sample, and n indicates variable
Number.
X covariance matrix is calculated firstThen S is the correlation matrix of X, carries out singular value decomposition: S to S
=UDU ',
Wherein, U is the eigenvectors matrix of S, D=diag (λ1,λ2,…λn), λ is n characteristic value.To which matrix X can
It is decomposed into
Pivot score matrix and residual error score matrix are respectively T ∈ Rm*k,And pivot matrix of loadings and residual error
Matrix of loadings is respectively P=[P1,P2,…Pk]∈Rn*k,K is pivot number, selection
The result that whether rationally can directly select the precision of Principal Component Analysis Model and measuring point impacts.
The standard of pivot contribution rate method selection pivot:
Wherein, c is thresholding contribution rate, is typically chosen 80% or more.
It is selected using operating status of the principle component analysis to high-pressure heater；
The analysis of main elements simplifies the data of high-pressure heater operating status；
Comparison obtains the primary process variable influenced in former variable on operating status；
The analysis of main elements includes:
Observation state data are standardized, to remove shadow of the difference to original variable of the dimension order of magnitude
It rings, variation is normalized to observation state data, indicate each variable after normalization with X=[x1, x2 ..., xp], meet
E (xi)=0, Var (xi)=1, i=1,2 ..., p；
Covariance matrix after calculating each variable indicates covariance matrix with S；
Calculate characteristic value and feature vector；The characteristic value of solution S and in magnitude order 1 >=λ of λ 2 >=... p >=0 λ, feature
Vector normal orthogonal turns to e1, e2 ..., ep,(j=1,2 ..., p) indicates j-th of pivot of x, and former variable passes through
The transformed main former score of characteristic variable；
Accumulation contribution rate is calculated, the variance of p is indicated with λ j,Indicate contribution rate,It indicates
Preceding k pivot accumulates contribution rate.
Step S500 specifically:
FCM cluster result is carried out using operating status of the Global Genetic Simulated Annealing Algorithm to high-pressure heater；
Clustering is one of algorithm of data mining, carries out data mining by statistical clustering method.It is poly-
Alanysis goes the distribution situation for understanding data by the way that a large amount of data are divided into the identical subclass of property.Cluster is exactly basis
One group of individual is divided into several classifications by similitude.Its target be as much as possible reduce the same category on individual between away from
From, and increase the distance of the individual to belong to a different category as far as possible.To obtain the set i.e. knot of cluster of one group of data object
Fruit, the cluster also referred to as clustered (class).Data in class are mutually similar, and the data in other classes are mutually different.
If the data X={ x1, x2 ..., xn } of sample, c class (2≤c≤n) are the class numbers for being divided into data sample
Number, { A1, A2 ..., Ac } indicate corresponding c classification, and U is its subordinated-degree matrix, and uij is person in servitude of the sample xj for class Ai
Category degree, V={ v1, v2 ..., vc } are cluster centre matrix.Then FCM objective function are as follows:
In formula: w is weighting coefficient, w ∈ [1 ,+∞]；Dij is each sample at a distance from cluster centre.J (U, V) is every
The feature of class sample to cluster centre Weighted distance quadratic sum, when J (U, V) numerical value reaches minimum, it was demonstrated that Clustering Effect is most
It is excellent.It requires a sample to be subordinate to angle value and for each cluster for 1, it may be assumed that
FCM algorithm needs to give cluster number c in advance, and it is not generally known that number is clustered in actual conditions.Cause
This, with the difference of cluster number, obtained cluster result is also different.We usually use Validity Index to scan for
The number for determining optimum cluster judges optimal cluster numbers from a large amount of cluster numbers, to obtain optimal cluster knot
Fruit.There is the Validity Index of many FCM clusters, but some Validity Indexes are only related to the geometry of data, but and fuzzy
It divides degree of membership not to be associated with, certainly exists limitation.Therefore, effective using Xie-Beni herein in order to improve clustering precision
Property index (while considering the geometry of data and the degree of membership of data) search for preferable clustering number.Document points out a kind of tool
The fuzzy clustering Validity Index for having better effects is Xie-Beni index, is calculated as following formula:
And when premium class number c* is cluster numbers c, VXB value is minimum, while it is meant that VXB value is smaller, it is better to obtain
Fuzzy clustering effect.
SearchInteger between section calculates the corresponding VXB value of each integer, obtains the smallest c value of VXB value
As optimum clustering number c*, here it is the methods for determining preferable clustering number c*.
Its algorithm steps are as follows:
Normalized sample data, nondimensionalization.Initial data is normalized using range transformation method,
Eliminate dimension.
Given cluster numbersSubordinated-degree matrix U and cluster centre matrix V are calculated with following formula.
Find optimum cluster result.Given cluster numbersIt calculates each c and corresponds to Validity Index VXB
Value, search for the corresponding preferable clustering number c* value of the smallest VXB value.It calculates preferable clustering number c* and corresponds to cluster centre matrix, i.e.,
Optimal classification mode, each sample data are assigned to specific classification, and optimum cluster result is acquired.
Steps are as follows for Global Genetic Simulated Annealing Algorithm:
Step 1: control parameter initializes.N is population scale；Pm is mutation probability；T is annealing initial temperature；α is temperature
Spend cooling parameter.
Step 2: the initial system of solutions generates at random.
Step 3: the system of solutions to generation executes step as follows, until next-generation heredity comes out.
(1) to each of group individual assessment its fitness value f (xi) (i=1,2,, N).
(2) two individual x are randomly choosediWith xjCarry out crisscross inheritance, and generates two new individual xi' and xj', then assess
Fitness value f (the x of two new individuals outi') and f (xj'), then receive new individual according to certain probability.
(3) hereditary variation is executed after individual intersection heredity, is judged whether according to the method in (2) new after receiving to make a variation
Solution.
Step 4: if Tk+1=α TK, turns third step.Or if meeting the condition of convergence, evolutionary process terminates.
Step S600 specifically:
It calculates the degree of membership of high-pressure heater and obtains the optimal operational condition of high-pressure heater.
From the above technical scheme, this application provides divided based on Global Genetic Simulated Annealing Algorithm high-pressure heater state
The method of analysis, first obtain high-pressure heater operating status data, and to as equipment itself heat caused by data problem into
The processing of row data human error and the data problem as caused by unstable environment or unstable instrument and signal carry out accidental
Error processing；The primary variables for influencing generator operation is found using the analysis of main elements, reduces the data of calculating, improves fortune
Row reaction speed, convenient for selecting representative measuring point；Based on genetic annealing algorithms to the operating status number of high-pressure heater
According to being excavated, finally calculates the degree of membership of high-pressure heater and obtain the optimal operational condition of high-pressure heater.The present invention
Based on Global Genetic Simulated Annealing Algorithm to the FCM cluster result of high-pressure heater running state data, which both overcomes FCM
Algorithm falls into the defect of locally optimal solution, and strengthens global search function, to be provided with better convergence and the overall situation is searched
Suo Nengli.
The c state that high-pressure heater equipment may occur are as follows: Ω={ ω1,ω2,…,ωc, then equipment state early warning
The forms of characterization of evidence is as follows:
In formula, xi is the measuring point vector of selection, and mi is the evidence being defined on the Ω of equipment state space, the coke member that A is mi
For reflecting the output variable for corresponding to xiA possibility that value.
The building of high-pressure heater status early warning evidence is described as follows: first in status data pretreatment and the choosing of state feature
On the basis of fixed, status data is analyzed, selection can characterize the running state data of equipment full working scope；Secondly, base
Clustering is carried out to status data in Data Clustering Algorithm and excavates the high-pressure heater status information lain in data,
It can be obtained the subjection degree uij that arbitrary data xi is under the jurisdiction of a certain classification wj；Third is excavated state wj to these and is divided
Analysis, determines " status categories " in its physical significance；Finally, combined data cluster result and expertise are to excavation classification
It is qualitative, high-pressure heater status early warning evidence set TR (i.e. evidence library) can be constructed, wherein mi (wj)=uij.
After pre-processing to the historical data that thermal power plant provides, 2 states (w1, w2) are clustered into, there are also used
The exit water temperature to 3754 groups of high-pressure heaters, extraction pressure, inlet water temperature, hydrophobic is obtained based on Global Genetic Simulated Annealing Algorithm
The subordinated-degree matrix of temperature and active power data clusters such as following table 5.1.Xi is i-th group of data, and wj is jth classification.Such as m5
(w2)=u52=0.2517, i.e., the 2nd classification of the 5th group data are 0.2517.
For the reasonability in experimental evidence library, 10 groups of number of faults are added in the status data of 3754 groups of high-pressure heaters
According to carrying out the FCM cluster based on Global Genetic Simulated Annealing Algorithm to it, and be polymerized to 3 classes, cluster is respectively such as table 5.2.3rd class packet
It containing 10 groups of data, and is exactly 10 groups of fault datas being added.
1 subordinated-degree matrix U of table
2 clustering distribution of table
The principal states parameter of high-pressure heater includes: exit water temperature, feedwater flow, inlet water temperature, extraction pressure, dredges
Coolant-temperature gage etc..Operating status measuring point exit water temperature, the inlet water temperature, hydrophobic temperature of high-pressure heater are determined using principle component analysis
Degree and active power, therefore, the operating status of heater can be determined by above-mentioned 4 state parameters.It is high in feed heating system
This 4 state parameters of pressure heater have apparent linear dependence, this can be also as a kind of rejecting of data prediction
Standard, that is, the data for being unsatisfactory for corresponding linear rule can be determined that be rejected for abnormal data, in order to avoid influence equipment state
The correctness of model.In addition, the very big data of numerical fluctuations are considered abnormal in a short time, directly reject.
Status data is obtained as training set sample from power plant, after being standardized to training set sample, carries out base
It is clustered in the FCM of Global Genetic Simulated Annealing Algorithm.Effect is best when training sample point is gathered for 2 class, i.e., by all normal fortune
Row state point is divided into 2 kinds of different operating statuses, and table 4.3 is all kinds of class center situation.Its subordinated-degree matrix U is 3754*2
Matrix.I.e. constituting a height adds operating status model, i.e. table 5.1 for these sample points and its subordinated-degree matrix.
After establishing the state model of high-pressure heater, each sample point corresponds to a degree of membership in model, indicates
The sample point is under the jurisdiction of a possibility that each known operating status, these degrees of membership are that the state recognition of new operating status point mentions
For evidence, the operating condition of state point to be identified may finally be determined by evidence fusion.Utilize above-mentioned model and evidence k-NN
Operating status locating for new operation data of the algorithm to high-pressure heater, which is identified, (judges that the high-pressure heater is in
A possibility that knowing 2 kinds of states).When the operating parameter of high-pressure heater changes, algorithm may determine that its operating status
Variation.
When the high-pressure heater is in known 2 kinds of states, evidence library is added in new operation data.When high-pressure heater is unknown
The degree of membership of operating status is greater than after early warning threshold values, carries out fault pre-alarming；Into after early warning, judgement leads to the correlation of early warning
Variable, and into the early warning reason library based on expert with operation instruction, search the early warning reason of possible energy；If artificial judgment,
This early warning is new operating status, then evidence library can be added in new state.
Further, the human error of the current time status data is deleted and is corrected, and is included the following steps:
Calculate the standard deviation sigma of current time observed quantity status data change ratei；
Judgement | Δ xi| with 3 σiSize, as | Δ xi| 3 σ of <i, then it is the status data of normal observation, when | Δ xi| >
3σi, observing the status data that refers to, there are human errors；
Current time observation state data are human errors, to Δ xiPositive negative judgement；If Δ xi> 0, then use xi=
xi-1+σiIt corrects and replaces the human error；If Δ xi< 0, then use xi=xi-1-σiIt corrects and replaces the human error.
Subsequent time observation state data adopt into repeating the process of step 1 to step 3.
Further, the Global Genetic Simulated Annealing Algorithm to the high-pressure heater running state data of acquisition carry out emulation and
Analysis, is calculated the degree of membership of high-pressure heater and the optimal operational condition of high-pressure heater.
Further, the Global Genetic Simulated Annealing Algorithm includes:
Control parameter initialization.N is population scale；Pm is mutation probability；T is annealing initial temperature；α is cooling for temperature
Parameter；
The initial system of solutions generates at random；
Step is executed as follows to the system of solutions of generation, until next-generation heredity comes out；Each of group individual is assessed
Its fitness value f (xi) (i=1,2,, N)；It randomly chooses two individual xi and xj and carrys out crisscross inheritance, and generate two newly
Then individual xi' and xj' evaluates the fitness value f (xi') and f (xj') of two new individuals, then connects according to certain probability
By new individual；Hereditary variation is executed after individual intersection heredity, judges whether to receive the new explanation after variation；
If Tk+1=α TK, step is gone back to, or if meeting the condition of convergence, evolutionary process terminates.
From the above technical scheme, this application provides one kind to be based on Global Genetic Simulated Annealing Algorithm to high-pressure heater shape
The method of state analysis obtains the data of high-pressure heater operating status first, and asks the data as caused by equipment heat itself
Topic carries out the processing of data human error and the data problem as caused by unstable environment or unstable instrument and signal carries out
Accidental error processing；The primary variables for influencing generator operation is found using the analysis of main elements, is reduced the data of calculating, is mentioned
Height operation reaction speed, convenient for selecting representative measuring point；Based on genetic annealing algorithms to the operation shape of high-pressure heater
State data are excavated, and are finally calculated the degree of membership of high-pressure heater and are obtained the optimal operational condition of high-pressure heater.This
Based on Global Genetic Simulated Annealing Algorithm to the FCM cluster result of high-pressure heater running state data, which was both overcome for invention
FCM algorithm falls into the defect of locally optimal solution, and strengthens global search function, to be provided with better convergence and complete
Office's search capability.
Similar portion cross-reference between embodiment provided by the present application, specific embodiment provided above is only
It is several examples under the total design of the application, does not constitute the restriction of the application protection scope.For those skilled in the art
For member, any other embodiment that foundation application scheme is expanded without creative efforts is all
Belong to the protection scope of the application.
Claims (9)
1. it is a kind of based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, which is characterized in that the method
The following steps are included:
Step S100: high-pressure heater running state data is obtained；
Step S200: human error in running state data acquisition is handled using Pauta criterion；
Step S300: using the wavelet threshold denoising method of the smooth unbiased possibility predication of history to random error in running state data acquisition
It is handled；
Step S400: it is selected using operating status of the principle component analysis to high-pressure heater；
Step S500: FCM cluster result is carried out using operating status of the Global Genetic Simulated Annealing Algorithm to high-pressure heater；
Step S600: calculating the degree of membership of high-pressure heater and obtains the optimal operational condition of high-pressure heater.
2. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In, the high-pressure heater running state data, including main feedwater flow, extraction temperature, feed pressure, extraction pressure, high pressure
The exit water temperature of heater, inlet water temperature, water level, drain temperature, upper end is poor and lower end is poor.
3. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In, it is described that human error in running state data acquisition is handled using Pauta criterion, pass through observed quantity operating status
Data variation rate size judges whether there is human error,
The running state data inscribed when current time, x1, x2 ..., xn and each expression of i >=2 is indicated with i, with Δ xi=xi-xi-1
Indicate two neighboring moment status data change rate；WithIndicate the equal of observed quantity running state data change rate
Value；WithIndicate the standard deviation of observed quantity running state data change rate；
If the absolute value when adjacent two moment observed quantity status data change rate is greater than 3 σi, | Δ xi| 3 σ of >i, current time observation
There are human errors for the status data of amount, and the human error of the current time status data is deleted and corrected.
4. as claimed in claim 3 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In the human error of the current time status data is deleted and correction, includes the following steps:
Calculate the standard deviation sigma of current time observed quantity status data change ratei；
Judgement | Δ xi| with 3 σiSize, as | Δ xi| 3 σ of <i, then it is the status data of normal observation, when | Δ xi| 3 σ of >i,
There are human errors for the status data that observation refers to；
Current time observation state data are human errors, to Δ xiPositive negative judgement；If Δ xi> 0, then use xi=xi-1+σi
It corrects and replaces the human error；If Δ xi< 0, then use xi=xi-1-σiIt corrects and replaces the human error；
Subsequent time observation state data adopt into repeating the above process.
5. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In, the wavelet threshold denoising method using the smooth unbiased possibility predication of history obtains running state data at random error
Reason,
The step of Wavelet-denoising Method includes:
Wavelet transformation is carried out to f containing noise signal (x)=s (x)+n (x), wherein s (x) indicates original signal, and n (x) expression is made an uproar
Sound signal；
Wavelet coefficient on each decomposition scale of threshold process obtains and estimates new wavelet coefficient wj,x, so thatAs far as possible
It is small；
To estimation wavelet coefficient wj,xCarry out wavelet reconstruction, the estimation signal after being denoised
The step of wavelet threshold denoising method includes:
Collected original signal is subjected to wavelet transformation, obtains the wavelet coefficient of each decomposition level；
If f (t) ∈ L2(R), wavelet transformation is indicated with following formula,
Wherein a indicate scale factor, b indicate translation because
Son,Indicate analysis wavelet；
Threshold process is carried out to wavelet coefficient, obtains new wavelet coefficient；
You is carried out to new wavelet coefficient to convert, and obtains reconstruction signal；
The step of history smooth unbiased possibility predication includes:
According to original signal calculate actual signal estimated value, and make actual signal estimated value and actual signal mean square deviation most
It is small；
Wherein, the original signal containing noise signal is indicated with X=[x0, x1 ..., xN-1] T, formula is
xi=si+ei, (i=0,1 ..., N-1), siIndicate the true signal value at i moment, eiIndicate the noise figure at i moment, N table
Show signal length；
It can be obtained using mean square deviation replacement for mathematic expectaion:
Wherein, X indicates original signal, and s indicates actual signal,Indicate the estimated value of actual signal s.
6. as claimed in claim 5 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
All be in the wavelet basis that, wavelet transformation and inversion are used instead it is orthogonal, mean square deviation can be indicated with following formula:
Wherein, wj, k indicate actual signal s k-th of wavelet coefficient on j-th of decomposition level；Indicate withIt is corresponding small
Wave system number.
7. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In the analysis of main elements simplifies the data of high-pressure heater operating status；
Comparison obtains the primary process variable influenced in former variable on operating status；
The analysis of main elements includes:
Observation state data are standardized, it is right to remove influence of the difference to original variable of the dimension order of magnitude
Variation is normalized in observation state data, with X=[x1, x2 ..., xp] indicate normalization after each variable, meet E (xi)=
0, Var (xi)=1, i=1,2 ..., p；
Covariance matrix after calculating each variable indicates covariance matrix with S；
Calculate characteristic value and feature vector；The characteristic value of solution S and in magnitude order 1 >=λ of λ 2 >=... p >=0 λ, feature vector
Normal orthogonal turns to e1, e2 ..., ep,Indicate j-th of pivot of x, former variable passes through spy
Main former score after levying change of variable；
Accumulation contribution rate is calculated, the variance of p is indicated with λ j,Indicate contribution rate,K master before indicating
Member accumulation contribution rate.
8. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In the Global Genetic Simulated Annealing Algorithm is emulated and analyzed to the high-pressure heater running state data of acquisition, is calculated
The degree of membership of high-pressure heater and the optimal operational condition of high-pressure heater.
9. as described in claim 1 based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis, feature exists
In the Global Genetic Simulated Annealing Algorithm includes:
Control parameter initialization, N is population scale；Pm is mutation probability；T is annealing initial temperature；α is temperature cooling parameter；
The initial system of solutions generates at random；
Step is executed as follows to the system of solutions of generation, until next-generation heredity comes out；To the assessment of each of group individual, it is suitable
Answer angle value f (xi) (i=1,2,, N)；It randomly chooses two individual xi and xj and carrys out crisscross inheritance, and generate two new individual xi'
With xj', the fitness value f (xi') and f (xj') of two new individuals are then evaluated, then receives new according to certain probability
Body；Hereditary variation is executed after individual intersection heredity, judges whether to receive the new explanation after variation；
If Tk+1=α TK, step is gone back to, or if meeting the condition of convergence, evolutionary process terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811571162.5A CN109710661A (en) | 2018-12-21 | 2018-12-21 | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811571162.5A CN109710661A (en) | 2018-12-21 | 2018-12-21 | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109710661A true CN109710661A (en) | 2019-05-03 |
Family
ID=66256055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811571162.5A Pending CN109710661A (en) | 2018-12-21 | 2018-12-21 | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109710661A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056858A (en) * | 2019-05-15 | 2019-07-26 | 贵州电网有限责任公司 | A kind of fired power generating unit Heater Terminal Temperature Difference adaptive regulation method and device |
| CN110132342A (en) * | 2019-05-22 | 2019-08-16 | 江苏方天电力技术有限公司 | Power station air preheater dust stratification monitoring method based on Clustering |
| CN110579709A (en) * | 2019-08-30 | 2019-12-17 | 西南交通大学 | fault diagnosis method for proton exchange membrane fuel cell for tramcar |
| CN111090048A (en) * | 2019-12-19 | 2020-05-01 | 重庆大学 | Vehicle-mounted data self-adaptive time interval transmission method for new energy automobile |
| CN112309167A (en) * | 2020-11-13 | 2021-02-02 | 河南观潮智能科技有限公司 | Intelligent parking management system and parking method |
| CN112528227A (en) * | 2020-11-18 | 2021-03-19 | 浙江蓝卓工业互联网信息技术有限公司 | Sensor abnormal data identification method based on mathematical statistics |
| CN113794725A (en) * | 2021-09-14 | 2021-12-14 | 北京本地通科技有限公司 | Electric digital data transmission method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104636813A (en) * | 2013-11-12 | 2015-05-20 | 中国科学院沈阳计算技术研究所有限公司 | Hybrid genetic simulated annealing algorithm for solving job shop scheduling problem |
| CN105260815A (en) * | 2015-09-18 | 2016-01-20 | 国网辽宁省电力有限公司鞍山供电公司 | Electric power system payment channel evaluating method based on large data |
| CN105488266A (en) * | 2015-11-25 | 2016-04-13 | 江苏方天电力技术有限公司 | Actual power generation power prediction method of thermal power generating unit |
| CN106936627A (en) * | 2016-09-28 | 2017-07-07 | 清华大学 | A kind of thermal power generating equipment performance monitoring method based on big data analysis mining |
-
2018
- 2018-12-21 CN CN201811571162.5A patent/CN109710661A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104636813A (en) * | 2013-11-12 | 2015-05-20 | 中国科学院沈阳计算技术研究所有限公司 | Hybrid genetic simulated annealing algorithm for solving job shop scheduling problem |
| CN105260815A (en) * | 2015-09-18 | 2016-01-20 | 国网辽宁省电力有限公司鞍山供电公司 | Electric power system payment channel evaluating method based on large data |
| CN105488266A (en) * | 2015-11-25 | 2016-04-13 | 江苏方天电力技术有限公司 | Actual power generation power prediction method of thermal power generating unit |
| CN106936627A (en) * | 2016-09-28 | 2017-07-07 | 清华大学 | A kind of thermal power generating equipment performance monitoring method based on big data analysis mining |
Non-Patent Citations (3)
| Title |
|---|
| 刘秋菊: ""基于遗传模拟退火算法的模糊聚类方法"", 《微计算机信息》 * |
| 苏志刚: ""球磨机制粉出力的软测量研究"", 《优秀硕士论文》 * |
| 闫奕荣: ""基于主元分析的多元指标综合评价方法研究"", 《统计与决策》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056858A (en) * | 2019-05-15 | 2019-07-26 | 贵州电网有限责任公司 | A kind of fired power generating unit Heater Terminal Temperature Difference adaptive regulation method and device |
| CN110132342A (en) * | 2019-05-22 | 2019-08-16 | 江苏方天电力技术有限公司 | Power station air preheater dust stratification monitoring method based on Clustering |
| CN110579709A (en) * | 2019-08-30 | 2019-12-17 | 西南交通大学 | fault diagnosis method for proton exchange membrane fuel cell for tramcar |
| CN111090048A (en) * | 2019-12-19 | 2020-05-01 | 重庆大学 | Vehicle-mounted data self-adaptive time interval transmission method for new energy automobile |
| CN111090048B (en) * | 2019-12-19 | 2021-09-24 | 重庆大学 | Vehicle-mounted data self-adaptive time interval transmission method for new energy automobile |
| CN112309167A (en) * | 2020-11-13 | 2021-02-02 | 河南观潮智能科技有限公司 | Intelligent parking management system and parking method |
| CN112528227A (en) * | 2020-11-18 | 2021-03-19 | 浙江蓝卓工业互联网信息技术有限公司 | Sensor abnormal data identification method based on mathematical statistics |
| CN113794725A (en) * | 2021-09-14 | 2021-12-14 | 北京本地通科技有限公司 | Electric digital data transmission method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109710661A (en) | Based on Global Genetic Simulated Annealing Algorithm to the method for high-pressure heater state analysis | |
| CN110441065B (en) | Gas turbine on-line detection method and device based on LSTM | |
| US11840998B2 (en) | Hydraulic turbine cavitation acoustic signal identification method based on big data machine learning | |
| CN112016251B (en) | Nuclear power device fault diagnosis method and system | |
| CN106055918A (en) | Power system load data identification and recovery method | |
| CN108921230A (en) | Method for diagnosing faults based on class mean value core pivot element analysis and BP neural network | |
| CN112036089A (en) | Coal mill fault early warning method based on DPC-MND and multivariate state estimation | |
| CN109740859A (en) | Transformer condition evaluation and system based on Principal Component Analysis and support vector machines | |
| CN109034191A (en) | One-dimensional telemetry exception interpretation method based on ELM | |
| CN111401749A (en) | Dynamic safety assessment method based on random forest and extreme learning regression | |
| CN111767657B (en) | Nuclear power system fault diagnosis method and system | |
| CN111680875A (en) | Unmanned aerial vehicle state risk fuzzy comprehensive evaluation method based on probability baseline model | |
| CN111899905A (en) | Fault diagnosis method and system based on nuclear power device | |
| CN109711435A (en) | A kind of support vector machines on-Line Voltage stability monitoring method based on genetic algorithm | |
| CN110175682A (en) | A kind of optimization core pivot element analysis fault monitoring method based on Chaos-Particle Swarm Optimization | |
| CN109685136A (en) | A kind of high-pressure heater status data analysis method | |
| Mao et al. | Comparative study on prediction of fuel cell performance using machine learning approaches | |
| CN110705831A (en) | Power angle instability mode pre-judgment model construction method after power system fault and application thereof | |
| CN114116832A (en) | Power distribution network abnormity identification method based on data driving | |
| Ding et al. | Deep Forest‐Based Fault Diagnosis Method for Chemical Process | |
| CN109389313B (en) | Fault classification diagnosis method based on weighted neighbor decision | |
| CN115017978A (en) | Fault classification method based on weighted probability neural network | |
| CN111814394B (en) | Power system safety assessment method based on correlation and redundancy detection | |
| CN115130564A (en) | High-pressure heater online monitoring method based on parallel GMM-LCKSVD | |
| CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network |
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: 20190503 |