CN106778816B - Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition - Google Patents

Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition Download PDF

Info

Publication number
CN106778816B
CN106778816B CN201611047414.5A CN201611047414A CN106778816B CN 106778816 B CN106778816 B CN 106778816B CN 201611047414 A CN201611047414 A CN 201611047414A CN 106778816 B CN106778816 B CN 106778816B
Authority
CN
China
Prior art keywords
combustion
length
gray level
mixing coefficient
black dragon
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.)
Expired - Fee Related
Application number
CN201611047414.5A
Other languages
Chinese (zh)
Other versions
CN106778816A (en
Inventor
刘禾
李新利
杨国田
于磊
胡叙畅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
North China Electric Power University
Original Assignee
North China Electric Power University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by North China Electric Power University filed Critical North China Electric Power University
Priority to CN201611047414.5A priority Critical patent/CN106778816B/en
Publication of CN106778816A publication Critical patent/CN106778816A/en
Application granted granted Critical
Publication of CN106778816B publication Critical patent/CN106778816B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20021—Dividing image into blocks, subimages or windows
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20212—Image combination
    • G06T2207/20224—Image subtraction

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Geometry (AREA)
  • Control Of Combustion (AREA)

Abstract

The invention belongs to the technical field of combustion flame image processing, and particularly relates to a combustion stability judging method based on combustion mixing coefficients and fuzzy recognition. The method fully considers the influence of the combustion mixing coefficient on the combustion stability, has good capability of identifying the combustion stability of the flame, can correctly distinguish the conditions of the combustion stability and the instability, and has higher accuracy.

Description

Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition
Technical Field
The invention belongs to the technical field of combustion flame image processing, and particularly relates to a combustion stability judging method based on combustion mixing coefficients and fuzzy recognition.
Background
Coal is one of the main energy sources in China, and the consumption of coal-fired boilers accounts for a considerable proportion. The combustion stability of the utility boiler is directly related to the reliability and safety of boiler operation, unstable combustion or poor combustion adjustment can not only cause steam parameters to fluctuate and reduce the thermal efficiency of the boiler, but also can cause fire extinguishing of a hearth and even explosion of the hearth, thereby causing huge economic loss. Therefore, the method for detecting the combustion flame of the boiler in real time, establishing a combustion stability judging method, improving the combustion efficiency and reducing the combustion pollution has important engineering application significance.
In the current stage, a research method for boiler combustion stability mainly measures flame characteristic signals such as flame brightness, flame intensity, flame spectrum and the like, and then performs mathematical processing such as spectrum analysis on the flame characteristic signals to obtain characteristic information reflecting combustion conditions. In recent years, most power station boilers in China widely use flame image monitoring systems to provide visual information of combustion states for hearth flame detection, and researchers complete judgment of the combustion stability states of the boilers by processing flame images, extracting corresponding characteristic quantities and combining methods such as a neural network, a support vector machine, a rough set theory and fuzzy recognition.
It is worth noting that the mixing effect of the pulverized coal and the heat flow in the furnace also has an important influence on the combustion condition in the furnace, the combustion mixing parameter is a characteristic parameter for representing the mixing effect of the pulverized coal and the heat flow in the furnace, if the mixing effect of the pulverized coal and the heat flow in the furnace is good, the combustion condition in the furnace is good, and the combustion is stable; if the mixing effect is poor, the combustion time of the pulverized coal can be prolonged, the combustion efficiency of the pulverized coal is reduced, and the combustion is unstable, so that the combustion mixing coefficient has important significance for the research on the combustion stability of the boiler.
Therefore, the combustion stability judging method based on the flame image at the present stage is not mature enough, and the influence of the combustion mixing coefficient on the combustion stability of the boiler is not considered, so that the accuracy of the combustion stability judging result is low.
Disclosure of Invention
In order to solve the problems, the invention provides a combustion stability judging method based on combustion mixing coefficients and fuzzy recognition, which comprises the following steps:
step one, taking a flame combustion area as a characteristic area, extracting a corresponding gray level parameter G and a gray level position distribution parameter D, and calculating a combustion mixing coefficient H and combustion mixing coefficient fluctuation △ H;
step two, performing edge detection on the combustion image, and calculating a black dragon length DI and a black dragon length fluctuation parameter △ DI;
and step three, establishing a combustion stability judging method based on fuzzy recognition by taking the combustion mixing coefficient H, the combustion mixing coefficient fluctuation △ H, the black dragon length DI and the black dragon length fluctuation parameter △ DI as characteristic indexes.
Preferably, in the first step, the combustion mixture ratio H is calculated as follows: taking the flame combustion area as a characteristic area, carrying out block processing on the characteristic area, and extracting gray level distribution information in each block areaThe gray scale parameter G is obtained by constructing Euclidean distances between feature vectors of pixel gray scale statistical information in each block region and the optimal feature vector and weighting, and the gray scale position distribution parameter D is constructed by constructing gray scale information of each sub-region and taking the distance from the sub-region to the edge of the black dragon on the image as position information of the sub-region
Figure BDA0001160299730000021
Wherein N is the number of the sub-regions,
Figure BDA0001160299730000022
for the sub-region gray scale features,
Figure BDA0001160299730000023
for the ith gray level of the jth sub-region, the number of pixel points, piIs the ith component in the weight vector P, n is the number of gray levels, siThe distance from the center point of the ith sub-area to the segmentation straight line;
determining a combustion mixing coefficient according to the gray level parameter G and the gray level position distribution parameter D
Figure BDA0001160299730000024
Wherein α is a weight coefficient with a value range of (0,1),
Figure BDA0001160299730000025
in order to normalize the gray-scale parameters,
Figure BDA0001160299730000026
the gray level position distribution parameters are normalized, wherein t is a sub-region pixel point.
Preferably, the combustion mixing coefficient fluctuation parameter △ H is calculated by taking the mixing coefficient difference Δ H of two frames of continuous time flame video imagesi＝|Hi-Hi-1Describing the fluctuation condition of the combustion mixing coefficient at the moment, so as to reflect the stable combustion of the boiler, wherein deltaHiA combustion mixing coefficient fluctuation parameter at time i, HiCombustion mixing coefficient at time i, Hi-1Is the combustion mixture coefficient at time i-1.
Preferably, in the second step, the calculation method of the black dragon length DI is as follows: determining n straight lines to be detected according to the effective area of the length of the black dragon and the moving direction of the pulverized coal airflow, searching the maximum gray gradient point of each straight line through a first-order difference algorithm to obtain n groups of black dragon edge coordinates, wherein the distance from each edge point to the boundary of the flame image is
Figure BDA0001160299730000031
The average value is the length of the black dragon of the flame image
Figure BDA0001160299730000032
Wherein (x)i0,yi0) Is the ith straight line initial pixel point coordinate, (x)ij,yij) And the coordinate of the pixel point with the maximum gray gradient of the ith straight line is obtained.
Preferably, the black dragon length fluctuation parameter △ DI is calculated by taking the black dragon length difference delta DI of two continuous time flame video imagesi＝|DIi-DIi-1I describe the fluctuation of the black dragon length at that moment, where Δ DIiA black dragon length fluctuation parameter at time i, DIiLength of black dragon at time i, DIi-1Is the length of the black dragon at the time i-1.
Preferably, in the third step, the method for determining combustion stability based on the combustion mixture coefficient and fuzzy recognition comprises the following steps:
(1) selecting membership functions of the combustion mixing coefficient H and the black dragon length parameter DI as a large-scale ridge distribution function and a middle-scale quadratic parabolic distribution function respectively, and selecting membership functions of the combustion mixing coefficient fluctuation parameter △ H and the black dragon length fluctuation parameter △ DI as small-scale quadratic parabolic distributions;
(2) selecting the membership degree of characteristic indexes of the stable combustion flame video image as a stable standard fuzzy set U1Selecting the membership degree of the characteristic index of the unstable combustion flame video image as an unstable standard fuzzy set U2；
(3) Extract the waiting judgmentDetermining flame image characteristic parameters, calculating characteristic index membership, and constructing a factor fuzzy set A of a combustion stability judging methodiSeparately calculate AiAnd a stable standard fuzzy set U1And an unstable standard fuzzy set U2The euclidean closeness of (c) is determined by using a closeness selection principle.
According to the method, the combustion stability discrimination method based on the combustion mixing coefficient and fuzzy recognition is constructed according to the pulverized coal combustion image, the discrimination of the boiler combustion stability is realized, the influence of the combustion mixing coefficient on the combustion stability is fully considered, the method has good capability of recognizing the flame combustion stability, the combustion stability and the instability can be correctly discriminated, and the accuracy is high.
Drawings
FIG. 1 is a flow chart of a combustion stability determination method according to the present invention
FIG. 2 is a combustion image of an embodiment of the present invention
FIG. 3 is a schematic diagram of the effective area of the burning image black dragon length
Detailed Description
The embodiments are described in detail below with reference to the accompanying drawings.
In the combustion process of the boiler, the combustion mixing coefficient directly influences the combustion condition of the boiler, and the method disclosed by the invention judges the combustion stability by the method shown in the figure 1, and specifically comprises the following steps of:
taking a flame combustion area as a characteristic area, extracting corresponding gray level parameters G and gray level position distribution parameters D, and calculating a combustion mixing coefficient H and combustion mixing coefficient fluctuation △ H, wherein the details are as follows:
the combustion image is collected by a flame image sensor, and the embodiment of the present invention uses the combustion image with the size of 320 × 240, see fig. 2. The method comprises the steps of taking a flame combustion area as a characteristic area, carrying out blocking processing on the characteristic area, dividing the characteristic area into N adjacent non-gap subregions with the size of t multiplied by t, carrying out gray level grading processing on each subregion in sequence, and extracting gray level distribution information in each blocking area, wherein the gray level distribution information comprises a gray level parameter G and a gray level position distribution parameter D, so that the image gray level distribution condition of coal powder pixels in a furnace at a spatial position is introduced into the measurement of the mixing effect of coal powder and heat flow in the furnace, namely the combustion mixing coefficient.
The gray parameter G is obtained by constructing Euclidean distance between the feature vector of the pixel gray level statistical information in each block region and the optimal feature vector and weighting. Correspondingly, in the n-level gray level image, when the mixing effect is the worst, the pixel points in the characteristic region are distributed in a lower gray level set; when the mixing effect is optimal, the pixels in the region are concentrated in the higher gray level set. Meanwhile, the more the gray level set pixel points with smaller gray values are, the poorer the mixing effect is; the more gray level set pixels with larger gray values, the better the mixing effect, so the influence of different gray level sets on the evaluation of the mixing effect is measured by weighting.
And obtaining the number of pixel points belonging to each gray level set in the characteristic region according to the number of pixel points belonging to each gray level set in each subregion. By constructing an n-dimensional feature vector A ═ a1,…,ai,…,an) Wherein a isiThe ratio of the number of pixels in the ith gray level set to the total number of pixels in the area is adopted, under an ideal condition, when the mixing effect of pulverized coal and heat flow in the furnace is the best, the combustion effect is the best, the brightness of the characteristic area of a combustion image is the highest, and therefore, A is usedb(0, …,0, …,0,1) as the n-dimensional optimal feature vector. At the same time with
Figure BDA0001160299730000051
Calculating gray scale parameters as weight vectors
Figure BDA0001160299730000052
Wherein p isi,ai,abiRespectively representing vectors P, AbN is the number of sub-regions.
The gray level position distribution parameter D is constructed by taking the distance from the sub-region to the black dragon edge on the image as the position information of the gray level information of each sub-region
Figure BDA0001160299730000053
Wherein N is the number of the sub-regions,
Figure BDA0001160299730000054
for the sub-region gray scale features,
Figure BDA0001160299730000055
for the ith gray level of the jth sub-region, the number of pixel points, piIs the ith component in the weight component P, n is the number of gray levels, siThe distance from the center point of the ith sub-region to the dividing straight line.
Determining a mixing coefficient according to the gray level parameter G and the gray level position distribution parameter D
Figure BDA0001160299730000056
Wherein α is a weight coefficient with a value range of (0,1),
Figure BDA0001160299730000057
in order to normalize the gray-scale parameters,
Figure BDA0001160299730000058
the gray level position distribution parameters are normalized, wherein t is a sub-region pixel point.
When the conditions such as the airflow speed of the pulverized coal or the temperature in the furnace are changed, the mixing effect of the pulverized coal and the heat flow in the furnace is also changed, and the combustion mixing coefficient is greatly fluctuated. Taking the combustion mixing coefficient difference delta H of two frames of continuous time flame video imagesi＝|Hi-Hi-1Describing the fluctuation condition of the combustion mixing coefficient at the moment to reflect whether the combustion of the boiler is stable or not, wherein deltaHiThe combustion mixture ratio fluctuates at time i.
Step two, performing edge detection on the combustion image, and calculating the black dragon length DI and the black dragon length fluctuation parameter △ DI, wherein the details are as follows:
in the black dragon length measurement process, the middle area of the mixed gas flow plays a role, and the effective area of the characteristic extraction is shown in fig. 3. When the pulverized coal airflow enters the initial combustion zone from the unburned zone, the gray level of pixel points of the flame image along the airflow direction is greatly increased, and from the initial combustion zone to the unburned zone, the gray level is rapidly reduced, the gray level gradients in two directions are respectively calculated, and the pixel points at the edge of the length of the black dragon can be determined. And performing edge detection on the flame image by adopting a first-order difference algorithm. The black dragon edge is determined by finding the maximum forward difference, i.e. the maximum point of the gray scale gradient in the image in the specified direction.
Determining n straight lines to be detected according to the effective area of the length of the black dragon and the moving direction of the pulverized coal airflow, and searching the maximum gray gradient point of each straight line through a first-order difference algorithm to obtain n groups of black dragon edge coordinates. Each edge point is at a distance from the flame image boundary of
Figure BDA0001160299730000061
The average value is the length of the flame image black dragon
Figure BDA0001160299730000062
Wherein (x)i0,yi0) Is the ith straight line initial pixel point coordinate, (x)ij,yij) And the coordinate of the pixel point with the maximum gray gradient of the ith straight line is obtained.
When the airflow or flame expansion speed of the pulverized coal changes due to the change of external conditions, the black dragon moves back and forth, the length of the black dragon changes, the pulverized coal is not stable in ignition and even the flame is blown out, so that the stable combustion condition can be reflected by the change of the length of the black dragon along with the change of time, and the difference delta DI between the lengths of the black dragon of two continuous-time flame video images is obtainedi＝|DIi-DIi-1I describe the fluctuation of the black dragon length at that moment, where Δ DIiThe black dragon length at time i fluctuates.
Step three, establishing combustion stability judgment based on combustion mixing coefficients and fuzzy recognition:
the combustion mixing coefficient H should be close to 1 in value at optimum combustion conditions, with the combustion effect being worse the further away from 1. When the combustion mixing coefficient is near 0 or 1, the numerical value of the combustion mixing coefficient is increased to have little influence on the combustion effect, otherwise, the influence is larger, so that the partial large-scale ridge-shaped distribution is selected as the membership function mu of the optimal combustion mixing coefficient1Taking x as combustion mixing coefficientH，x∈[0,1]Then, then
Figure BDA0001160299730000063
When the fluctuation delta H of the combustion mixing coefficient is large, the combustion effect is large in change, the combustion instability is increased, and therefore the small quadratic parabolic distribution is adopted as the optimal combustion mixing coefficient fluctuation membership function u2. Taking x as the fluctuation 50 x delta H of the mixing coefficient, wherein x belongs to [0,50 ]]Then, then
Figure BDA0001160299730000071
The length of black dragon length DI is in moderate state during best combustion situation, and black dragon length is close to the image middle part more, and the combustion effect is better, and the combustion effect when drawing close to the centre promotes more slowly, otherwise, the combustion effect is worse, but the combustion effect when drawing close to the centre promotes more fast. Therefore, taking the intermediate quadratic parabolic distribution as the best membership function u of the black dragon length3Get it
Figure BDA0001160299730000072
DImaxIs the maximum length of black dragon, x belongs to [0,1 ]]Then, then
Figure BDA0001160299730000073
The combustion effect is optimal when the fluctuation delta DI of the length of the black dragon approaches to 0, the combustion effect becomes worse along with the increase of the fluctuation of the length of the black dragon, the speed of the deterioration of the combustion effect becomes fast, and therefore the small quadratic parabolic distribution is selected as the membership function u of the fluctuation of the length of the optimal black dragon4. Get
Figure BDA0001160299730000074
x∈[0,5]Then, then
Figure BDA0001160299730000075
When the combustion condition is optimal under the ideal condition, the mixing effect of the pulverized coal and the heat flow in the furnace is in the optimal state, namelyThe combustion mixture ratio H is 1, the black dragon length DI should be in the middle of the flame image, and the combustion mixture ratio fluctuation Δ H and the black dragon length fluctuation Δ DI are minimum, i.e., the fluctuation value is 0. At this time, the degree of membership u of each characteristic index1＝1,u2＝1,u3＝1,u41, selecting a stable standard fuzzy set U1(1,1,1, 1). On the contrary, when the combustion condition is worst under the ideal condition, the combustion mixing coefficient H is small, the soot length DI should be too long or too short, and the combustion mixing coefficient fluctuation Delta H and the soot length fluctuation Delta DI are large, so that the unstable standard fuzzy set U is selected2＝(0,0,0,0)。
Respectively extracting a combustion mixing coefficient H, combustion mixing coefficient fluctuation delta H, black dragon length DI and black dragon fluctuation delta DI based on a combustion flame image, calculating the membership degree of the combustion mixing coefficient H, the combustion mixing coefficient fluctuation delta H, the black dragon length DI and the black dragon fluctuation delta DI to the optimal combustion condition, and obtaining a fuzzy set X to be detectedi. Respectively calculating the fuzzy sets X to be detected according to a near selection principleiAnd stable standard fuzzy set U1Unstable standard fuzzy set U2Euclidean closeness of
Figure BDA0001160299730000081
Wherein n is the number of burners of the acquired flame video image, and in this embodiment, n is 4; i is 1,2, …, m is the number of the obtained fuzzy sets to be detected, and m is 60 in the embodiment; j is 1, 2.
And judging the combustion stability according to a selection principle. If N (X)i,U1)>N(Xi,U2) The fuzzy set X is explainediAnd a stable standard fuzzy set U1The flame image boiler is closer, and the combustion condition of the corresponding flame image boiler is stable; otherwise, the combustion condition of the boiler is unstable.
The present invention is not limited to the above embodiments, and any changes or substitutions that can be easily made by those skilled in the art within the technical scope of the present invention are also within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (6)

1. A combustion stability judging method based on combustion mixing coefficients and fuzzy recognition is characterized in that the method comprises the following steps of extracting flame image features collected by a flame image sensor, calculating combustion mixing coefficients, combustion mixing coefficient fluctuation parameters, black dragon length and black dragon length fluctuation parameters, and establishing a combustion stability judging method based on fuzzy recognition:
step one, taking a flame combustion area as a characteristic area, extracting a gray level parameter and a gray level position distribution parameter, and calculating a combustion mixing coefficient and a combustion mixing coefficient fluctuation parameter;
secondly, performing edge detection on the flame image, and calculating the length of the black dragon and the fluctuation parameter of the length of the black dragon;
and step three, establishing a combustion stability judging method based on the combustion mixing coefficient and fuzzy recognition by taking the combustion mixing coefficient, the combustion mixing coefficient fluctuation parameter, the blackdragon length and the blackdragon length fluctuation parameter as characteristic indexes.
2. The combustion stability determination method according to claim 1, wherein in the first step, the combustion mixture coefficient is calculated by: taking a flame combustion area as a characteristic area, carrying out block processing on the flame combustion area, extracting gray level distribution information in each block area, wherein the gray level distribution information comprises gray level parameters and gray level position distribution parameters, the gray level parameters are obtained by constructing Euclidean distances between characteristic vectors of pixel gray level statistical information in each block area and optimal characteristic vectors and weighting, the gray level position distribution parameters are obtained by constructing gray level information of each subarea and taking the distance from each subarea to the edge of a black dragon on an image as position information of each subarea, and the gray level position distribution parameters are constructed by constructing the gray level information of each subarea
Figure FDA0002284044570000011
Wherein D is a gray level position distribution parameter, N is the number of sub-regions,
Figure FDA0002284044570000012
for the sub-region gray scale features,
Figure FDA0002284044570000013
for the ith gray level of the jth sub-region, the number of pixel points, piIs the ith component in the weight vector P, n is the number of gray levels, siThe distance from the center point of the ith sub-area to the segmentation straight line;
determining a combustion mixing coefficient according to the gray level parameter and the gray level position distribution parameter
Figure FDA0002284044570000021
Wherein α is a weight coefficient with a value range of (0,1),
Figure FDA0002284044570000022
in order to normalize the gray-scale parameters,
Figure FDA0002284044570000023
the gray level position distribution parameters are normalized, wherein t is a sub-region pixel point.
3. The combustion stability determination method according to claim 2, wherein the combustion mixture coefficient fluctuation parameter calculation method is as follows: taking the combustion mixing coefficient difference delta H of two continuous time flame imagesi＝|Hi-Hi-1The fluctuation condition of the combustion mixing coefficient at the moment i is described, so that the combustion stability of the boiler is reflected, wherein deltaHiA combustion mixing coefficient fluctuation parameter at time i, HiCombustion mixing coefficient at time i, Hi-1Is the combustion mixture coefficient at time i-1.
4. The combustion stability determination method according to claim 1, wherein in the second step, the calculation method of the black dragon length is as follows: determining n straight lines to be detected according to the effective area of the length of the black dragon and the moving direction of the pulverized coal airflow, searching the maximum gray gradient point of each straight line through a first-order difference algorithm to obtain n groups of black dragon edge coordinates, wherein the distance from each edge point to the boundary of the flame image is
Figure FDA0002284044570000024
The average value is the length of the black dragon of the flame image
Figure FDA0002284044570000025
Wherein (x)i0,yi0) Is the ith straight line initial pixel point coordinate, (x)ij,yij) And the coordinate of the pixel point with the maximum gray gradient of the ith straight line is obtained.
5. The combustion stability determination method according to claim 4, wherein the soot length fluctuation parameter calculation method is as follows: black dragon length difference delta DI of two-frame continuous time flame imagei＝|DIi-DIi-1I describes the fluctuation of the length of the dragon at time i, where Δ DIiA black dragon length fluctuation parameter at time i, DIiLength of black dragon at time i, DIi-1Is the length of the black dragon at the time i-1.
6. The method of determining combustion stability according to claim 1, wherein in the third step, the method of establishing combustion stability based on the combustion mixture coefficient and the fuzzy recognition includes:
(1) setting membership functions of the combustion mixing coefficient and the black dragon length parameter as a partial large ridge-shaped distribution function and a middle secondary parabolic distribution function respectively, and setting membership functions of the combustion mixing coefficient fluctuation parameter and the black dragon length fluctuation parameter as partial small secondary parabolic distributions;
(2) selecting the characteristic index membership degree of the stable combustion flame image as a stable standard fuzzy set, and selecting the characteristic index membership degree of the unstable combustion flame image as an unstable standard fuzzy set;
(3) extracting the characteristic parameters of the flame image to be judged, calculating the membership degree of the characteristic indexes, constructing a factor fuzzy set of the combustion stability judging method, respectively calculating the Euclidean closeness of the factor fuzzy set and a stable standard fuzzy set and an unstable standard fuzzy set, and judging the combustion stability by using a proximity principle.
CN201611047414.5A 2016-11-23 2016-11-23 Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition Expired - Fee Related CN106778816B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611047414.5A CN106778816B (en) 2016-11-23 2016-11-23 Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611047414.5A CN106778816B (en) 2016-11-23 2016-11-23 Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition

Publications (2)

Publication Number Publication Date
CN106778816A CN106778816A (en) 2017-05-31
CN106778816B true CN106778816B (en) 2020-06-05

Family

ID=58974514

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611047414.5A Expired - Fee Related CN106778816B (en) 2016-11-23 2016-11-23 Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition

Country Status (1)

Country Link
CN (1) CN106778816B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108445845B (en) * 2018-02-12 2020-02-14 国网山东省电力公司电力科学研究院 Intelligent boiler combustion stability judgment system and method suitable for large power station
CN109214332A (en) * 2018-08-31 2019-01-15 华北电力大学 A kind of combustion stability method of discrimination based on furnace flame image fractal characteristic
CN113674280A (en) * 2021-10-25 2021-11-19 启东万惠机械制造有限公司 Method for measuring temperature of hearth of power station boiler
CN114754353B (en) * 2022-04-13 2023-05-30 山西大学 Circulating fluidized bed boiler combustion optimization method integrating neighborhood rough set machine learning

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2390675A (en) * 2002-07-10 2004-01-14 Univ Greenwich Flame characteristic monitor using digitising image camera
CN103077394B (en) * 2012-12-31 2017-02-08 天津大学 Method for automatically monitoring flame combustion stability
CN105183935B (en) * 2015-07-20 2018-11-27 昆明理工大学 A kind of evaluation method of flame combustion situation and stability
CN105912872B (en) * 2016-04-27 2018-06-22 华北电力大学 Hot-fluid mixed effect measure in a kind of coal dust and stove based on burning image

Also Published As

Publication number Publication date
CN106778816A (en) 2017-05-31

Similar Documents

Publication Publication Date Title
CN106778816B (en) Combustion stability judging method based on combustion mixing coefficient and fuzzy recognition
CN107314819B (en) A kind of detection of photovoltaic plant hot spot and localization method based on infrared image
CN106845443B (en) Video flame detection method based on multi-feature fusion
Qiu et al. An autoadaptive edge-detection algorithm for flame and fire image processing
CN107609470B (en) Method for detecting early smoke of field fire by video
CN106846305B (en) A kind of boiler combustion stability monitoring method based on the more characteristics of image of flame
Wang et al. Pattern recognition for measuring the flame stability of gas-fired combustion based on the image processing technology
CN107481237B (en) A kind of infrared array image hot spot detection method based on multiframe temperature characterisitic
CN105868165B (en) A kind of station boiler operation data cleaning method
CN110245850A (en) A kind of sintering process operating mode's switch method and system considering timing
CN107730548A (en) It is a kind of based on average gray and area towards heating furnace flame real-time detection method
Salamanca et al. On the detection of solar panels by image processing techniques
CN111461003B (en) Coal-fired working condition identification method based on video image sequence feature extraction
CN108830311A (en) A kind of infrared image photovoltaic array recognition methods based on Adaboost
CN112733900B (en) Boiler combustion state stability judging method based on deep migration learning
CN105912872B (en) Hot-fluid mixed effect measure in a kind of coal dust and stove based on burning image
Garami et al. Reaction zone monitoring in biomass combustion
CN113963013B (en) Markless power transmission tower displacement vibration identification method based on computer vision
CN108460353B (en) Infrared image human body detection method based on time sequence tracking
Yang et al. A Multi-Modal Data-Driven Decision Fusion Method for Process Monitoring in Metal Powder Bed Fusion Additive Manufacturing
Tao et al. A new flame image edge detection algorithm for reconstruction of the internal temperature field in the furnace
CN112396101A (en) Intelligent image analysis and decision-making method based on firepower plant hearth flame television
CN111860923A (en) Boiler flue gas NO based on multi-model clustering integrationxEmission prediction algorithm
Zhao Application research of image processing technology for fire detection and fire alarm based on blockchain
CN105976359A (en) Method for measuring diffusion angle of pulverized coal in furnace of boiler burner based on image technology

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20200605

Termination date: 20201123