CN109145988A - Determination method, apparatus, equipment and the storage medium of the target operating condition of denitrating system - Google Patents
Determination method, apparatus, equipment and the storage medium of the target operating condition of denitrating system Download PDFInfo
- Publication number
- CN109145988A CN109145988A CN201810960110.0A CN201810960110A CN109145988A CN 109145988 A CN109145988 A CN 109145988A CN 201810960110 A CN201810960110 A CN 201810960110A CN 109145988 A CN109145988 A CN 109145988A
- Authority
- CN
- China
- Prior art keywords
- cluster
- distance
- sample data
- operating condition
- standard
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/231—Hierarchical techniques, i.e. dividing or merging pattern sets so as to obtain a dendrogram
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of determination methods of the target operating condition of denitrating system, before being clustered based on FCM clustering algorithm to sample data set, the standard cluster centre and standard cluster numbers of sample data can be determined based on fuzzy F- statistic, method when using single FCM clustering algorithm to determine target operating condition in compared with the prior art, this programme can obtain accurate standard cluster centre and standard cluster numbers, to guarantee that FCM clustering algorithm obtains accurate target operating condition using accurate standard cluster centre and standard cluster numbers, to guarantee the normal operation of out of stock system.In addition, the invention also discloses a kind of determining device of the target operating condition of denitrating system, equipment and storage medium, effect is as above.
Description
Technical field
The present invention relates to technical field of electric power, in particular to a kind of determination method of the target operating condition of denitrating system.Device,
Equipment and storage medium.
Background technique
With the fast development of science and technology, the application of various information systems has become the important of power plant's production and operation
Working foundation condition, the especially gradual perfection in SIS system data warehouse, the operation data of various production control systems are highly dense
The acquisition and long-term accumulation preservation of degree.Power plant and entrance " big data " epoch, although being rich in procedural information in these data,
It is a lack of effective information excavating tool.
It is fortune currently, being the key link in denitration running optimizatin to the determination of the denitrating system target operating condition in power plant
The basis of row Operating Guideline and fault diagnosis, based on SIS system actual history operating condition data, data is excavated poly-
Application of the class method in denitrating system operational objective operating condition and status monitoring, using actual motion optimum condition as denitrating system
The target operating condition of actual condition, i.e. unit (for the unit of denitration) stable operation history work optimal in selection same operating cluster
Target operating condition of the condition as the operating condition cluster represents the unit best stabilized operation level actually occurred in the operating condition cluster.
Currently, being to utilize FCM clustering algorithm, fuzzy F- statistics quantity algorithm, mould to the acquisition of target operating condition in denitrating system
Paste C means clustering algorithm etc. obtains the target operating condition of out of stock system.Common algorithm is FCM algorithm, and FCM algorithm gathers
Class result is highly susceptible to the influence of cluster numbers and initial cluster center, but use single FCM algorithm to cluster numbers and
When cluster centre is chosen, cluster centre and cluster numbers are often inaccurate, in this way, since initial cluster center and cluster numbers select
It takes inaccurately, to will cause influence to the precision of the cluster result of out of stock system, also you can't get the targets of denitrating system
Operating condition.Therefore, when determining target operating condition using single FCM clustering algorithm, if initial cluster center and cluster numbers selection be not smart
It is true then will lead to that finally obtained target operating condition is not accurate, further influence the operation of out of stock system.
Therefore, how accurately to determine that the target operating condition in denitrating system with the normal operation for ensuring out of stock system is this field
Technical staff's problem to be solved.
Summary of the invention
It is an object of the invention to disclose a kind of determination method of the target operating condition of denitrating system.Device, equipment and storage
Medium can accurately determine the target operating condition in denitrating system, thereby further ensure that the normal operation of out of stock system.
To achieve the above object, the embodiment of the invention discloses following technical solutions:
First, the embodiment of the invention discloses a kind of choosing methods of target operating condition, comprising:
Obtain sample data set corresponding with denitrating system；
The standard cluster centre and standard cluster numbers of the sample data set are determined based on fuzzy F- statistic；
According to the standard cluster centre and the standard cluster numbers, based on FCM clustering algorithm to the sample data set
It is clustered to obtain cluster result；
Target operating condition corresponding with the out of stock system is determined from the cluster result.
Preferably, described to determine that the standard cluster centre of the sample data set and standard cluster based on fuzzy F- statistic
Number includes:
The first distance between all samples that the sample data is concentrated is calculated, fuzzy similarity distance matrix is obtained；
The initial clustering number of the sample data set is determined according to the similarity distance matrix；
The second distance between the sample data in all kinds of in the initial clustering number is calculated, and generates corresponding layering
Clustering tree；
The hierarchical cluster tree is split according to the second distance, obtains multiple segmentation results；
Cluster numbers corresponding with each segmentation result are calculated, are chosen from each cluster numbers using fuzzy F- statistic
Optimal cluster numbers are clustered as the standard cluster numbers using the corresponding cluster centre of the standard cluster numbers as the standard
Center.
Preferably, the first distance calculated between all samples that the sample data is concentrated includes:
Determine each sample point that the sample data is concentrated；
Target sample point is chosen from each sample；
The first distance of each sample point to the target sample point is calculated using euclidean distance method.
Preferably, the second distance between the sample data calculated in all kinds of in the initial clustering number, and it is raw
Include: at corresponding hierarchical cluster tree
Determine the number of plies of clustering tree corresponding with initial clustering number, the number of plies of the clustering tree and the initial clustering number phase
Together；
Determine the distance value of the second distance between the sample data in all kinds of in initial clustering number；
Hierarchical cluster number is determined according to the second distance between the sample data in all kinds of；
Each hierarchical cluster number is divided to the level of corresponding clustering tree；
The hierarchical cluster tree is generated according to the initial clustering number and the hierarchical cluster number.
Preferably, described that the hierarchical cluster tree is split according to the second distance, obtain multiple segmentation results
Include:
Determine the distance value of each second distance；
The hierarchical cluster tree is divided into quantity sub- clustering tree identical with each distance value；The segmentation knot
The quantity of fruit is identical as the quantity of each second distance, and each second distance is corresponding with quantity and the second distance
The identical sub- clustering tree of distance value.
Preferably, described according to the standard cluster centre and the standard cluster numbers, it is clustered based on FCM
Algorithm is clustered to obtain cluster result to the sample data set
Iteration ends number is determined according to the sample data set；
According to described in the determination of the standard cluster centre, the iteration ends number and the standard cluster numbers
Sample data concentrates the degree of membership of each sample；
The determining subordinated-degree matrix corresponding with each sample；
Chosen from the subordinated-degree matrix the corresponding class of maximum membership degree as with the subordinated-degree matrix pair
The cluster result for the sample answered.
Second, the embodiment of the invention discloses a kind of selecting devices of target operating condition, comprising:
Module is obtained, for obtaining sample data set corresponding with out of stock system；
Determining module, for determining the standard cluster centre and standard cluster of sample data set based on fuzzy F- statistic
Number；
Cluster module, for being based on FCM clustering algorithm to institute according to the standard cluster centre and the standard cluster numbers
Sample data set is stated to be clustered to obtain cluster result；
Module is chosen, for choosing target operating condition corresponding with the out of stock system from the cluster result.
Preferably, the determining module includes:
First computing unit obtains distance for calculating the first distance between all samples that the sample data is concentrated
Matrix；
Determination unit, for the determining initial clustering number with the sample data set；
Second computing unit for calculating the second distance between all kinds of in the initial clustering number, and generates correspondence
Hierarchical cluster tree；
Cutting unit obtains multiple segmentation knots for being split according to the second distance to the hierarchical cluster tree
Fruit；
Computing unit determines the mark using K mean algorithm for calculating cluster numbers corresponding with each segmentation result
Quasi- cluster centre and the standard cluster numbers.
Second, the embodiment of the invention discloses a kind of selected equipments of target operating condition, comprising:
Memory, for storing computer program；
Processor, for executing the computer program stored in the memory to realize as described in any of the above one kind
The step of determination method of the target operating condition of denitrating system.
4th, the embodiment of the invention discloses a kind of computer readable storage medium, deposited on computer readable storage medium
Computer program is contained, the target operating condition of as above any denitrating system is realized when computer program is executed by processor
Determination method the step of.
As it can be seen that a kind of determination method of the target operating condition of denitrating system disclosed by the embodiments of the present invention, obtains and de- first
The corresponding sample data set of pin system, be then based on fuzzy F- statistic determine sample data set primary standard cluster centre and
Standard cluster numbers are clustered to obtain most based on FCM algorithm according to standard cluster centre and standard cluster numbers to sample data set
Whole cluster result finally determines target operating condition from cluster result again.Therefore, using this programme, it is being based on FCM clustering algorithm
Before clustering to sample data set, the standard cluster centre and standard cluster of sample data can be determined based on fuzzy F- statistic
Number, compared with the prior art in method when determining target operating condition using single FCM clustering algorithm, this programme can obtain accurately
Standard cluster centre and standard cluster numbers, to guarantee that FCM clustering algorithm is poly- using accurate standard cluster centre and standard
Class number obtains accurate target operating condition, to guarantee the normal operation of out of stock system.In addition, the embodiment of the invention also discloses one
Determining device, equipment and the storage medium of the target operating condition of kind denitrating system, effect are as above.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of determination method flow schematic diagram of the target operating condition of out of stock system disclosed by the embodiments of the present invention；
Fig. 2 is a kind of determination apparatus structure schematic diagram of the target operating condition of out of stock system disclosed by the embodiments of the present invention；
Fig. 3 is a kind of target operating condition of denitrating system disclosed by the embodiments of the present invention locking equipment structural schematic diagram really.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of determination methods of the target operating condition of denitrating system.Device, equipment and storage are situated between
Matter can accurately determine the target operating condition in denitrating system, thereby further ensure that the normal operation of out of stock system.
Referring to Figure 1, Fig. 1 is a kind of determination method flow of the target operating condition of out of stock system disclosed by the embodiments of the present invention
Schematic diagram, comprising:
S101, sample data set corresponding with out of stock system is obtained.
Specifically, sample data set corresponding with out of stock system includes each parameter in denitrating system, such as in the present embodiment
The concentration of inlet, the exhaust gas volumn in out of stock system and the unit load data in out of stock system absorb NOXSlurry
PH value.Target operating condition corresponding with out of stock system can be with are as follows: out of stock efficiency, out-of-stock cost etc., it is each in out of stock system by choosing
The specific size of parameter determines the target operating condition of denitrating system.Sample data set in the embodiment of the present invention can be denitration
History denitration data in system, due to the operating condition of denitrating system and the flue gas flow of inlet and NOXConcentration it is direct
Correlation, the optimal value of the parameter setting of denitrating system also with unit operating condition and change, therefore, in the embodiment of the present invention
As preferred embodiment, by the NO of the historical data of the unit operation in denitrating system and inletXConcentration as this hair
Sample data in bright embodiment.So as to choose and unit load and NOXThe corresponding target operating condition of concentration, such as target work
Condition is to NOXDenitration efficiency is minimum and the operational efficiency highest of unit.Certainly, as target operating condition specifically why type,
The implementation of the embodiment of the present invention can't be impacted, according to the technical solution of the present invention, realize the choosing to target operating condition
It takes, i.e., how to determine optimal target operating condition, so that it is determined that going out to determine the parameter of the optimal objective operating condition of denitrating system most
Excellent value.
S102, the standard cluster centre and standard cluster numbers that sample data set is determined based on fuzzy F- statistic.
Specifically, the concept of fuzzy F- statistic may refer to the prior art in the present embodiment, it is applied to the present invention and implements
It is specific as follows as preferred embodiment in example: to acquire the distance that sample data concentrates each sample point first, generate corresponding
Fuzzy similarity distance matrix, i.e., by the Distance conformability degree of each sample point higher sample distance as in fuzzy similarity matrix
Element.It is determining initial poly- with sample data set using fuzzy similarity distance matrix after obtaining fuzzy similarity distance matrix
Then class number calculates the second distance of the class between different initial clustering numbers, it is poly- to generate corresponding layering using second distance
Hierarchical cluster tree is split to obtain multiple segmentation results (about this portion based on second distance for hierarchical cluster tree by class tree
Point, hierarchical cluster tree can be divided into the quantity for the sub- clustering tree that segmentation obtains according to the size of the value of each second distance
Equal with the value of second distance, i.e. the quantity of segmentation result and the quantity of second distance is identical, and number is included in each segmentation result
Measure the equal sub- clustering tree of the value of corresponding with each segmentation result second distance), corresponding in each segmentation result, having pair
The type of the number of clusters and cluster answered, using K mean algorithm (may refer to the prior art) determine standard cluster centre and
Standard cluster numbers.About this partial content, will introduce in greater detail below.
S103, according to standard cluster centre and standard cluster numbers, sample data set is clustered based on FCM clustering algorithm
Obtain cluster result.
Specifically, in the present embodiment, after determining standard cluster centre and standard cluster numbers, by the standard cluster centre
With standard cluster numbers as the cluster centre and cluster numbers in FCM clustering algorithm, then carry out the cluster operation of next step.Wherein,
It FCM clustering algorithm and is consistent in the prior art, detailed process are as follows: the size first according to the data volume of sample data set is true
Iteration ends number (number for allowing iteration) is determined, then according to the standard cluster centre and mark determined by F- fuzzy statistics amount
Each sample data that quasi- cluster numbers determine that sample data is concentrated belongs to person in servitude all kinds of in standard cluster centre and standard cluster numbers
Category degree, each sample standard deviation are corresponding with quantity degree of membership identical with standard cluster numbers, generate the subordinated-degree matrix of each sample, often
It is corresponding with maximum membership degree in a sample subordinated-degree matrix, at this point, using the corresponding class of maximum membership degree as the cluster of sample
As a result.About this partial content, it will describe in detail in embodiment, wouldn't elaborate herein later.
S104, target operating condition corresponding with out of stock system is determined from cluster result.
Specifically, the sample data that degree of membership is not much different is as a class, then, from all kinds of in the present embodiment
The corresponding sample data of optimal degree of membership is chosen as target operating condition corresponding with out of stock system (for example, objective function is de-
Out of stock efficiency in pin system then chooses the optimized parameter in denitrating system when being optimal out of stock efficiency, for example, optimal
The gas concentration of inlet, the optimal unit load in the exhaust gas volumn and out of stock system in optimal out of stock system).
As it can be seen that a kind of determination method of the target operating condition of denitrating system disclosed by the embodiments of the present invention, obtains and de- first
The corresponding sample data set of pin system, be then based on fuzzy F- statistic determine sample data set primary standard cluster centre and
Standard cluster numbers are clustered to obtain most based on FCM algorithm according to standard cluster centre and standard cluster numbers to sample data set
Whole cluster result finally determines target operating condition from cluster result again.Therefore, using this programme, it is being based on FCM clustering algorithm
Before clustering to sample data set, the standard cluster centre and standard cluster of sample data can be determined based on fuzzy F- statistic
Number, compared with the prior art in method when determining target operating condition using single FCM clustering algorithm, this programme can obtain accurately
Standard cluster centre and standard cluster numbers, to guarantee that FCM clustering algorithm is poly- using accurate standard cluster centre and standard
Class number obtains accurate target operating condition, to guarantee the normal operation of out of stock system.
Based on the above embodiment, as preferred embodiment, step S102 includes:
The first distance between all samples that sample data is concentrated is calculated, fuzzy similarity distance matrix is obtained；
The initial clustering number of sample data set is determined according to similarity distance matrix；
The second distance between all kinds of in initial clustering number is calculated, and generates corresponding hierarchical cluster tree；
Hierarchical cluster tree is split according to second distance, obtains multiple segmentation results；
Cluster numbers corresponding with each segmentation result are calculated, determine that standard cluster centre and standard cluster using K mean algorithm
Number.
Specifically, in the present embodiment, sample data concentrates this first distance of various kinds that can pass through euclidean distance method meter
It obtains, i.e., the Euclidean distance between each sample point, in the present embodiment, arbitrarily chooses sample centered on a sample point
Point calculates each sample to the distance of the central sample point, the distance that distance value is not much different then is determined from each distance
Value Types, and the corresponding first distance of the distance value of each type is formed into similarity distance matrix.
Wherein, it as preferred embodiment, calculates sample data and concentrates the first distances of all samples to include:
Determine each sample point that sample data is concentrated；
Target sample point is chosen from each sample；
Using euclidean distance method calculate the center of each sample to target sample center of a sample first distance.
Specifically, each sample point that can be concentrated to sample data is summed it up and seeks sample data in the present embodiment
Then sample data is concentrated the sample point closest to the average sample point as target sample point, Europe by the average sample point of collection
Formula Furthest Neighbor may refer to the prior art, after determining target sample point, can calculate each sample using euclidean distance method
This point arrives the first distance of target sample point.
Wherein, as preferred embodiment, calculate between the sample data in all kinds of in initial clustering number second away from
From, and generate corresponding hierarchical cluster tree and include:
Determine the number of plies of clustering tree corresponding with initial clustering number, the number of plies of the clustering tree and the initial clustering number phase
Together；
Determine the distance value of the second distance between the sample data in all kinds of in initial clustering number；
Hierarchical cluster number is determined according to the second distance between the sample data in all kinds of；
Each hierarchical cluster number is divided to the level of corresponding clustering tree；
The hierarchical cluster tree is generated according to the initial clustering number and the hierarchical cluster number.
After obtaining the initial clustering number of sample data set, then calculate initial clustering number in it is all kinds of in sample number
Second distance between corresponds to the sample data corresponding with fuzzy similarity distance matrix in each class, calculate each
The second distance (can also be calculated using euclidean distance method) of each sample data in a fuzzy similarity distance matrix.So
The second distance for obscuring each sample in similarity distance matrix according to each afterwards determines in the fuzzy distance matrix again
Cluster numbers, after the cluster numbers for obtaining each sample data in each fuzzy similarity distance matrix, generate hierarchical cluster
Tree, wherein each of hierarchical cluster tree greatly branch into each class of initial clustering number, each of hierarchical cluster tree
Class in big branch comprising the corresponding cluster numbers of second distance in fuzzy similarity distance matrix corresponding with the branch.
Wherein, as preferred embodiment, hierarchical cluster tree is split according to second distance, obtains multiple segmentation knots
Fruit includes:
Determine the distance value of each second distance；
Each hierarchical cluster tree is divided into quantity sub- clustering tree identical with each distance value, the quantity of segmentation result and each the
The quantity of two distances is identical, and it is identical with the distance value of second distance from clustering tree that each second distance is corresponding with quantity.
Specifically, in the present embodiment, after obtaining hierarchical cluster tree, according to the second distance in all kinds of by hierarchical cluster
Tree is split, i.e., hierarchical cluster tree is divided by hierarchical cluster tree using the second distance in all kinds of as segmentation foundation
The quantity of sub- clustering tree and the equal sub- clustering tree of the value of second distance (i.e. each second distance corresponding a segmentation result),
How many second distance just how many segmentation result.Then for the sub- clustering tree for including in each segmentation result is calculated
Number, how many sub- clustering tree just how many cluster numbers, is then acquired in all segmentation results using fuzzy F- statistic
The maximum value of cluster numbers, it is using maximum cluster numbers as optimal cluster numbers (standard cluster numbers), the standard cluster numbers are corresponding
Cluster centre as standard cluster centre.
Based on the above embodiment, as preferred embodiment, step S103 includes:
Iteration ends number is determined according to sample data set.
Establishing criteria cluster centre, iteration ends number and standard cluster numbers determine that sample data concentrates being subordinate to for various kinds sheet
Degree.
Determine the corresponding subordinated-degree matrix of each sample.
Cluster of the corresponding class of maximum membership degree as sample corresponding with subordinated-degree matrix is chosen from subordinated-degree matrix
As a result.
Specifically, obtaining sample data set, (history that can be the unit operation in out of stock system is negative in the present embodiment
Lotus data) after, iteration ends number is then determined, if the sample data volume that sample data is concentrated is larger, by iteration ends
Number is arranged larger, on the contrary.Calculating about degree of membership can use following formula calculating, specific as follows:
The basic principle of FCM algorithm may refer to the prior art, for parameter therein, also may refer to the prior art.
In above formula, step refers to the number of iterations,iRefer to the group number that the sample data for concentrating sample data is grouped, what k referred to
It is the sample vector data in Vector Groups i, what the m in formula was referred to is parameter (the i.e. input change of the sample data concentration of input
The number of amount).C refers to cluster numbers.dikAnd djkWhat is indicated is the distance function of i-th of class and j-th of class.
After obtaining above-mentioned degree of membership, new cluster master mould is obtained using FCM algorithm, specific as follows:
That ε is indicated is the number of iterations, xkWhat is indicated is in sample data vector.uikWhat is indicated is k-th of sample in i-th group
The degree of membership of notebook data vector.Vstep+1What is indicated is thestep+1The cluster numbers of secondary iteration.It is calculated when by above-mentioned calculation formula
After new cluster master mould, judge whether to meet condition
||V(step+1)-V(step+i)| | < ε illustrates to have reached greatest iteration time at this time if meeting the condition at this time
Number, therefore, will finally obtain result as cluster result.
A kind of determining device of the target operating condition of denitrating system disclosed by the embodiments of the present invention is introduced below, please be join
See that Fig. 2, Fig. 2 are a kind of determination apparatus structure schematic diagram of the target operating condition of out of stock system disclosed by the embodiments of the present invention, the dress
It sets and includes:
Module 201 is obtained, for obtaining sample data set corresponding with out of stock system；
Determining module 202, the standard cluster centre and standard for determining sample data set based on fuzzy F- statistic are poly-
Class number；
Cluster module 203, for being based on FCM clustering algorithm to sample number according to standard cluster centre and standard cluster numbers
It is clustered to obtain cluster result according to collection；
Module 204 is chosen, for choosing target operating condition corresponding with out of stock system from cluster result.
As it can be seen that a kind of determining device of the target operating condition of denitrating system disclosed by the embodiments of the present invention, obtains and de- first
The corresponding sample data set of pin system, be then based on fuzzy F- statistic determine sample data set primary standard cluster centre and
Standard cluster numbers are clustered to obtain most based on FCM algorithm according to standard cluster centre and standard cluster numbers to sample data set
Whole cluster result finally determines target operating condition from cluster result again.Therefore, using this programme, it is being based on FCM clustering algorithm
Before clustering to sample data set, the standard cluster centre and standard cluster of sample data can be determined based on fuzzy F- statistic
Number, compared with the prior art in method when determining target operating condition using single FCM clustering algorithm, this programme can obtain accurately
Standard cluster centre and standard cluster numbers, to guarantee that FCM clustering algorithm is poly- using accurate standard cluster centre and standard
Class number obtains accurate target operating condition, to guarantee the normal operation of out of stock system.
Based on above embodiments, as preferred embodiment, determining module 202 includes:
First computing unit, the first distance between all samples for calculating sample data concentration, obtains distance matrix；
Determination unit, for the determining initial clustering number with sample data set；
Second computing unit for calculating the second distance between all kinds of in initial clustering number, and generates corresponding point
Layer clustering tree；
Cutting unit obtains multiple segmentation results for being split according to second distance to hierarchical cluster tree；
Computing unit is determined in standard cluster for calculating cluster numbers corresponding with each segmentation result using K mean algorithm
The heart and standard cluster numbers.
Fig. 3 is referred to, Fig. 3 is a kind of target operating condition of denitrating system disclosed by the embodiments of the present invention locking equipment structure really
Schematic diagram, comprising:
Memory 301, for storing computer program；
Processor 302, for executing the computer program stored in the memory to realize what any of the above item was mentioned
The step of determination method of the target operating condition of denitrating system.
The target operating condition of denitrating system provided in this embodiment locking equipment really calls storage due to that can pass through processor
The computer program of device storage, realizes the step of the determination method of the target operating condition of the denitrating system provided such as above-mentioned any embodiment
Suddenly, so this detection device has the same actual effect of determination method of the target operating condition with above-mentioned denitrating system.
This programme in order to better understand, a kind of computer readable storage medium disclosed by the embodiments of the present invention, computer
It is stored with computer program on readable storage medium storing program for executing, realizes that any embodiment as above is mentioned when computer program is executed by processor
Denitrating system target operating condition determination method the step of.
Computer readable storage medium provided in this embodiment, since computer-readable storage can be called by processor
The computer program of media storage, the determination method that the target operating condition of the denitrating system provided such as above-mentioned any embodiment is provided
Step, so this detection device has the same actual effect of determination method of the target operating condition with above-mentioned denitrating system.
Above to a kind of determination method, apparatus, equipment and the storage of the target operating condition of denitrating system disclosed in the present application
Medium is described in detail.Specific examples are used herein to illustrate the principle and implementation manner of the present application, with
The explanation of upper embodiment is merely used to help understand the present processes and its core concept.It should be pointed out that being led for this technology
For the those of ordinary skill in domain, under the premise of not departing from the application principle, can also to the application carry out it is several improvement and
Modification, these improvement and modification are also fallen into the protection scope of the claim of this application.
Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with other realities
The difference of example is applied, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part illustration
?.
It should also be noted that, in the present specification, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Claims (10)
1. a kind of determination method of the target operating condition of denitrating system characterized by comprising
Obtain sample data set corresponding with denitrating system；
The standard cluster centre and standard cluster numbers of the sample data set are determined based on fuzzy F- statistic；
According to the standard cluster centre and the standard cluster numbers, the sample data set is carried out based on FCM clustering algorithm
Cluster obtains cluster result；
Target operating condition corresponding with the out of stock system is determined from the cluster result.
2. the determination method of the target operating condition of denitrating system according to claim 1, which is characterized in that described based on fuzzy
F- statistic determines the standard cluster centre of the sample data set and standard cluster numbers include:
The first distance between all samples that the sample data is concentrated is calculated, fuzzy similarity distance matrix is obtained；
The initial clustering number of the sample data set is determined according to the similarity distance matrix；
The second distance between the sample data in all kinds of in the initial clustering number is calculated, and generates corresponding hierarchical cluster
Tree；
The hierarchical cluster tree is split according to the second distance, obtains multiple segmentation results；
Cluster numbers corresponding with each segmentation result are calculated, are chosen from each cluster numbers using fuzzy F- statistic optimal
Cluster numbers as the standard cluster numbers, using the corresponding cluster centre of the standard cluster numbers as in standard cluster
The heart.
3. the determination method of the target operating condition of denitrating system according to claim 2, which is characterized in that described in the calculating
Sample data concentrate all samples between first distance include:
Determine each sample point that the sample data is concentrated；
Target sample point is chosen from each sample；
The first distance of each sample point to the target sample point is calculated using euclidean distance method.
4. the determination method of the target operating condition of denitrating system according to claim 2, which is characterized in that described in the calculating
The second distance between sample data in all kinds of in initial clustering number, and generate corresponding hierarchical cluster tree and include:
Determine that the number of plies of clustering tree corresponding with initial clustering number, the number of plies of the clustering tree are identical as the initial clustering number；
Determine the distance value of the second distance between the sample data in all kinds of in initial clustering number；
Hierarchical cluster number is determined according to the second distance between the sample data in all kinds of；
Each hierarchical cluster number is divided to the level of corresponding clustering tree；
The hierarchical cluster tree is generated according to the initial clustering number and the hierarchical cluster number.
5. the determination method of the target operating condition of denitrating system according to claim 2, which is characterized in that described according to
Second distance is split the hierarchical cluster tree, obtains multiple segmentation results and includes:
Determine the distance value of each second distance；
The hierarchical cluster tree is divided into quantity sub- clustering tree identical with each distance value；The segmentation result
Quantity is identical as the quantity of each second distance, each second distance be corresponding with quantity and the second distance away from
Sub- clustering tree identical from value.
6. the determination method of the target operating condition of denitrating system according to claim 1, which is characterized in that described according to
Standard cluster centre and the standard cluster numbers cluster the sample data set based on FCM clustering algorithm
Result includes:
Iteration ends number is determined according to the sample data set；
Determine that the sample data is concentrated according to the standard cluster centre, the iteration ends number and the standard cluster numbers
The degree of membership of each sample；
The determining subordinated-degree matrix corresponding with each sample；
The corresponding class of maximum membership degree is chosen from the subordinated-degree matrix as sample corresponding with the subordinated-degree matrix
Cluster result.
7. a kind of determining device of the target operating condition of denitrating system characterized by comprising
Module is obtained, for obtaining sample data set corresponding with out of stock system；
Determining module, for determining the standard cluster centre and standard cluster numbers of sample data set based on fuzzy F- statistic；
Cluster module, for being based on FCM clustering algorithm to the sample according to the standard cluster centre and the standard cluster numbers
Notebook data collection is clustered to obtain cluster result；
Module is chosen, for choosing target operating condition corresponding with the out of stock system from the cluster result.
8. the determining device of the target operating condition of denitrating system according to claim 7, which is characterized in that the determining module
Include:
First computing unit obtains distance matrix for calculating the first distance between all samples that the sample data is concentrated；
Determination unit, for the determining initial clustering number with the sample data set；
Second computing unit for calculating the second distance between all kinds of in the initial clustering number, and generates corresponding point
Layer clustering tree；
Cutting unit obtains multiple segmentation results for being split according to the second distance to the hierarchical cluster tree；
Computing unit determines that the standard is gathered using K mean algorithm for calculating cluster numbers corresponding with each segmentation result
Class center and the standard cluster numbers.
9. a kind of target operating condition of denitrating system locking equipment really characterized by comprising
Memory, for storing computer program；
Processor, for executing the computer program stored in the memory to realize as described in any one of claim 1 to 6
Denitrating system target operating condition determination method the step of.
10. a kind of computer readable storage medium, computer program, feature are stored on the computer readable storage medium
It is, the computer program is executed by processor to realize the target such as denitrating system as claimed in any one of claims 1 to 6
The step of determination method of operating condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810960110.0A CN109145988A (en) | 2018-08-22 | 2018-08-22 | Determination method, apparatus, equipment and the storage medium of the target operating condition of denitrating system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810960110.0A CN109145988A (en) | 2018-08-22 | 2018-08-22 | Determination method, apparatus, equipment and the storage medium of the target operating condition of denitrating system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109145988A true CN109145988A (en) | 2019-01-04 |
Family
ID=64790622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810960110.0A Pending CN109145988A (en) | 2018-08-22 | 2018-08-22 | Determination method, apparatus, equipment and the storage medium of the target operating condition of denitrating system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109145988A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111581581A (en) * | 2020-04-23 | 2020-08-25 | 大唐环境产业集团股份有限公司 | Method and system for detecting NOx concentration at SCR inlet under multi-boundary condition |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020095260A1 (en) * | 2000-11-28 | 2002-07-18 | Surromed, Inc. | Methods for efficiently mining broad data sets for biological markers |
| US20030175720A1 (en) * | 2002-03-18 | 2003-09-18 | Daniel Bozinov | Cluster analysis of genetic microarray images |
| US20150286703A1 (en) * | 2014-04-08 | 2015-10-08 | International Business Machines Corporation | Adaptive variable selection for data clustering |
| CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
| US20180073446A1 (en) * | 2015-02-20 | 2018-03-15 | Bingo Interests, Llc | System to control and condition the supply of natural gas to bi-fuel engines |
| CN108038561A (en) * | 2017-09-21 | 2018-05-15 | 南京航空航天大学 | A kind of Multipurpose Optimal Method of SCR denitration preformed catalyst |
-
2018
- 2018-08-22 CN CN201810960110.0A patent/CN109145988A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020095260A1 (en) * | 2000-11-28 | 2002-07-18 | Surromed, Inc. | Methods for efficiently mining broad data sets for biological markers |
| US20030175720A1 (en) * | 2002-03-18 | 2003-09-18 | Daniel Bozinov | Cluster analysis of genetic microarray images |
| US20150286703A1 (en) * | 2014-04-08 | 2015-10-08 | International Business Machines Corporation | Adaptive variable selection for data clustering |
| US20180073446A1 (en) * | 2015-02-20 | 2018-03-15 | Bingo Interests, Llc | System to control and condition the supply of natural gas to bi-fuel engines |
| CN106709662A (en) * | 2016-12-30 | 2017-05-24 | 山东鲁能软件技术有限公司 | Electrical equipment operation condition classification method |
| CN108038561A (en) * | 2017-09-21 | 2018-05-15 | 南京航空航天大学 | A kind of Multipurpose Optimal Method of SCR denitration preformed catalyst |
Non-Patent Citations (4)
| Title |
|---|
| 周金海等: "基于遗传算法的模糊聚类在临床决策分析中的研究", 《医学信息》 * |
| 姜伦 等: "关于模糊C-均值(FCM)聚类算法的改进", 《计算机与数字工程》 * |
| 闵敏: "基于F统计量层次聚类的组合式模糊聚类算法研究", 《电脑与信息技术》 * |
| 顾慧 等: "一种基于EKFCM算法的电站脱硫系统目标工况库的建立方法", 《中国电机工程学报》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111581581A (en) * | 2020-04-23 | 2020-08-25 | 大唐环境产业集团股份有限公司 | Method and system for detecting NOx concentration at SCR inlet under multi-boundary condition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104462184B (en) | A kind of large-scale data abnormality recognition method based on two-way sampling combination | |
| CN101630376B (en) | Soft-sensing modeling method and soft meter of multi-model neural network in biological fermentation process | |
| CN109858572A (en) | A kind of modified hierarchy clustering method for sewage abnormality detection | |
| CN103135009B (en) | Electric appliance detection method and system based on user feedback information | |
| CN106295899B (en) | Wind power probability density Forecasting Methodology based on genetic algorithm Yu supporting vector quantile estimate | |
| CN110264116A (en) | A kind of Electrical Power System Dynamic safety evaluation method explored based on relationship with regression tree | |
| CN110889535B (en) | Multi-point wind speed prediction method in wind power plant based on convolution cyclic neural network | |
| CN113011481B (en) | Electric energy meter function abnormality assessment method and system based on decision tree algorithm | |
| CN110263998B (en) | Double-layer correction method for multisource numerical weather forecast set | |
| CN105701554A (en) | Electrical equipment energy efficiency chaos analysis method based on massive measurement data | |
| CN104809658A (en) | Method for rapidly analyzing low-voltage distributing network area line loss | |
| CN107133632A (en) | A kind of wind power equipment fault diagnosis method and system | |
| CN107909208A (en) | Damage method drops in a kind of taiwan area distribution | |
| CN108921322A (en) | A kind of ultra-short term node load prediction technique and system based on Spot Price | |
| CN106779219A (en) | A kind of electricity demand forecasting method and system | |
| Li et al. | A spatiotemporal directed graph convolution network for ultra-short-term wind power prediction | |
| CN110321934A (en) | A kind of method and system detecting user power utilization abnormal data | |
| Yu et al. | An efficient substation placement and sizing strategy based on GIS using semi-supervised learning | |
| CN109697566A (en) | Electronic product processing technology evaluation system and its evaluation method | |
| CN105005711A (en) | Method and device for acquiring statistical line loss | |
| CN108132423B (en) | A kind of method for rapidly positioning based on state transition probability power system monitoring data distortion | |
| CN108536130A (en) | A kind of Fault Diagnosis in Chemical Process system of colony intelligence optimizing | |
| CN105631465A (en) | Density peak-based high-efficiency hierarchical clustering method | |
| CN109307852A (en) | A kind of method and system of the measurement error of determining electric automobile charging pile electric energy metering device | |
| CN108230183A (en) | A kind of processing method of the grid equipment various dimensions comprehensive warning based on time scale measurement |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200713 Address after: Guangzhou Yuexiu District City, Guangdong province 510060 Dongfeng Road No. 757 Applicant after: GUANGDONG POWER GRID Co.,Ltd. Applicant after: ELECTRIC POWER RESEARCH INSTITUTE OF GUANGDONG POWER GRID Co.,Ltd. Applicant after: GUANGDONG YUELONG POWER GENERATION CO.,LTD. Address before: 510600 No. 757 Dongfeng East Road, Guangzhou, Guangdong, Yuexiu District Applicant before: GUANGDONG POWER GRID Co.,Ltd. Applicant before: GUANGDONG DIANKEYUAN ENERGY TECHNOLOGY Co.,Ltd. Applicant before: GUANGDONG YUELONG POWER GENERATION CO.,LTD. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190104 |