CN112561230A - Environment-friendly equipment state monitoring method based on electrical characteristics - Google Patents
Environment-friendly equipment state monitoring method based on electrical characteristics Download PDFInfo
- Publication number
- CN112561230A CN112561230A CN202011230180.4A CN202011230180A CN112561230A CN 112561230 A CN112561230 A CN 112561230A CN 202011230180 A CN202011230180 A CN 202011230180A CN 112561230 A CN112561230 A CN 112561230A
- Authority
- CN
- China
- Prior art keywords
- environmental protection
- protection equipment
- cluster
- electrical
- equipment
- 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.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 27
- 230000007613 environmental effect Effects 0.000 claims abstract description 93
- 238000009826 distribution Methods 0.000 claims abstract description 26
- 238000007621 cluster analysis Methods 0.000 claims abstract description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 33
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 230000014509 gene expression Effects 0.000 claims description 11
- 238000007476 Maximum Likelihood Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 abstract description 5
- 238000012360 testing method Methods 0.000 description 7
- 230000005611 electricity Effects 0.000 description 5
- 239000003245 coal Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000006477 desulfuration reaction Methods 0.000 description 2
- 230000023556 desulfurization Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 238000009776 industrial production Methods 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 102000003780 Clusterin Human genes 0.000 description 1
- 108090000197 Clusterin Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/219—Managing data history or versioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Water Supply & Treatment (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Public Health (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The embodiment of the application provides an environmental protection equipment state monitoring method based on electrical characteristics, which comprises the steps of obtaining historical electrical data of normally operating environmental protection equipment, carrying out cluster analysis on the historical electrical data, fitting each obtained cluster to obtain a binary Gaussian distribution model, and calculating parameters of the binary Gaussian distribution model; calculating the average probability density of each cluster based on the obtained binary Gaussian model distribution parameters; the method comprises the steps of periodically collecting an electrical data time sequence of the environment-friendly equipment, and calculating the membership degree of the electrical data time sequence and the electrical characteristics of the environment-friendly equipment; and comparing the electrical characteristic membership degree with a preset threshold value, and outputting a monitoring result of the running state of the environmental protection equipment. The service condition of the environmental protection equipment of an industrial enterprise is monitored on line by analyzing the electrical characteristics of the environmental protection equipment such as current, power and the like, and the negative environmental protection behaviors of the enterprise are found in time, so that the environmental protection department of the government is assisted to carry out effective supervision.
Description
Technical Field
The application belongs to the field of database management, and particularly relates to an environment-friendly equipment state monitoring method based on electrical characteristics.
Background
For a long time, the problem of environmental pollution caused by industrial production is always a hot topic of attention of all circles of society, and therefore, a series of measures are carried out by the country to encourage industrial enterprises to install environment-friendly equipment so as to reduce the pollutant discharge amount in the industrial production as much as possible and realize green production. In 2014, the national development reform committee and the environmental protection department have introduced an environment-friendly electricity price policy for coal and electricity enterprises, and as long as the coal and electricity enterprises install environment-friendly equipment meeting the requirements, the coal and electricity enterprises can obtain a certain maximum added price subsidy of 4.5 minutes per kilowatt-hour.
However, after some enterprises report and install the environmental protection equipment to the government, the environmental protection equipment is not used according to the regulations in order to save the operation and maintenance cost. In the first year when the policy of 'environment-friendly electricity price' is implemented, a plurality of thermal power plants subordinate to five power generation groups are heavily penalized due to unauthorized shutdown of the desulfurization facility of the power generation group; according to the data published by the environmental protection department in 2015, only enterprises with outstanding problems in desulfurization and denitrification facilities in 2014 have 17 families. The negative environmental protection behavior not only volatilizes subsidy funds issued by the country, but also seriously hinders the smooth implementation of the environmental protection policy.
At present, environmental protection departments master the implementation situation of enterprises on environmental protection policies mainly by monitoring pollutant discharge amount, and lack effective supervision on the use situation of environmental protection equipment, and the vulnerability is a main reason for the frequent occurrence of the bad behaviors. ZL201420121769.4 discloses a remote automatic monitoring data acquisition technology of environmental protection equipment, aims at monitoring the power consumption state of the environmental protection equipment in real time, and assists environmental protection law enforcement personnel to carry out online supervision on the service condition of the enterprise environmental protection equipment. However, the technology focuses on data acquisition, and simply judges whether the running state of the environmental protection equipment is abnormal or not according to the ratio of the actual power to the rated power of the environmental protection equipment.
Disclosure of Invention
In order to overcome the defects and shortcomings in the prior art, the method for monitoring the state of the environment-friendly equipment based on the electrical characteristics is provided, and the running state of the environment-friendly equipment is periodically evaluated through the power utilization data of the environment-friendly equipment.
Specifically, the method for monitoring the state of the environmental protection equipment based on the electrical characteristics provided by the embodiment of the application comprises the following steps:
step 1, acquiring historical electrical data of normally-operated environment-friendly equipment, performing cluster analysis on the historical electrical data, fitting each obtained cluster to obtain a binary Gaussian distribution model, and calculating parameters of the binary Gaussian distribution model;
step 2, calculating the average probability density of each cluster based on the obtained binary Gaussian model distribution parameters;
step 3, regularly collecting the electrical data time sequence of the environmental protection equipment, and calculating the membership degree of the electrical data time sequence and the electrical characteristics of the environmental protection equipment;
and 4, comparing the electrical characteristic membership degree with a preset threshold value, and outputting a monitoring result of the running state of the environmental protection equipment.
Optionally, step 1 includes:
selecting active power P and reactive power Q as electrical data of the environmental protection equipment to be analyzed, and clustering historical power data of the environmental protection equipment by using a mean shift algorithm to obtain clusters with different numerical value distribution characteristics;
fitting a binary Gaussian distribution model for each cluster from which the cluster is generated, wherein the probability density function is in the form of:
wherein x is [ P, Q ═ Q]TMu and Σ are the mean vector parameter and covariance matrix parameter of the random variable x, respectively, and can be obtained by maximum likelihood estimation, as shown in the following formula:
wherein XiRepresents the observed sample of the random variable x, and N is the total number of observed samples.
Optionally, step 2 includes:
let K be the total number of clusters of the mean shift algorithm, and c be the kth clusterk;
The mean probability density level is defined as: c is tokEach sample point in (a) is substituted into the corresponding gaussian score of the clusterIn the probability density function expression of the cloth model, obtaining different probability densities and calculating the average value of the probability densities, wherein the average value is the average probability density level P of the clusterkThe expression is as follows:
optionally, step 3 includes:
let the input power sequence of the algorithm be S ═ S1,s2,…,sTIn which s isi=[Pi,Qi]TT is the input monitoring period length;
defining each point s in the input sequence based on the mean probability density level P defined in step 3iAt ckMembership b in a clusterk(si):
Definition of B(s)i) Is s isiRelative to the membership degree of the electrical characteristics of the environment-friendly equipment, the calculation method is as follows:
B(si)=max{b1(si),b2(si),...,bk(si)},
then, calculating the electrical characteristic membership B (S) of the whole input power sequence relative to the environmental protection equipment by using the following formula:
optionally, the step 4 includes:
and marking the threshold value of the membership degree B (S) of the electrical characteristics as sigma, when the membership degree B (S) is more than or equal to sigma, determining that the sequence accords with the electrical characteristics when the environmental protection equipment normally operates, and judging that the state of the environmental protection equipment is normal, otherwise, judging that the environmental protection equipment is abnormal.
Optionally, the method for obtaining the threshold σ includes:
dividing historical power data into a plurality of time sequences meeting algorithm input requirements;
and calculating the membership degree of the electrical characteristics of each sequence relative to the environmental protection equipment, and taking the minimum value of the membership degrees as a threshold value sigma.
The beneficial effect that technical scheme that this application provided brought is: the service condition of the environmental protection equipment of an industrial enterprise is monitored on line by analyzing the electrical characteristics of the environmental protection equipment such as current, power and the like, and the negative environmental protection behaviors of the enterprise are found in time, so that the environmental protection department of the government is assisted to carry out effective supervision.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for monitoring a state of an environmental protection device based on electrical characteristics according to an embodiment of the present disclosure;
fig. 2 is a clustering result of history numbers based on a mean shift algorithm according to an embodiment of the present application.
Detailed Description
To make the structure and advantages of the present application clearer, the structure of the present application will be further described with reference to the accompanying drawings.
Example one
Specifically, the method for monitoring the state of the environmental protection equipment based on the electrical characteristics, provided by the embodiment of the present application, is used for realizing accurate sensing and timely reporting of the abnormal state of the environmental protection equipment, and includes:
step 1, acquiring historical electrical data of normally-operated environment-friendly equipment, performing cluster analysis on the historical electrical data, fitting each obtained cluster to obtain a binary Gaussian distribution model, and calculating parameters of the binary Gaussian distribution model;
step 2, calculating the average probability density of each cluster based on the obtained binary Gaussian model distribution parameters;
step 3, regularly collecting the electrical data time sequence of the environmental protection equipment, and calculating the membership degree of the electrical data time sequence and the electrical characteristics of the environmental protection equipment;
and 4, comparing the electrical characteristic membership degree with a preset threshold value, and outputting a monitoring result of the running state of the environmental protection equipment.
In implementation, the method for monitoring the state of the environmental protection equipment provided by the embodiment of the application realizes the monitoring of the state of the environmental protection equipment through the following steps:
step 1: and measuring voltage and current signals in an electrical loop of the environmental protection equipment by using the electric power data acquisition device, and further obtaining active power and reactive power data of the environmental protection equipment.
Step 2: and accumulating the historical power data of the normal operation of the environmental protection equipment, and clustering the historical power data of the environmental protection equipment by using a mean shift algorithm to obtain clusters with different numerical distribution characteristics. The specific steps of the algorithm are as follows: firstly, randomly selecting an initial sample point from a data set, iterating the initial sample point through a mean shift vector to be distributed to a nearby area with higher data density, taking the initial sample point and surrounding sample points as a cluster, then continuously randomly selecting a sample point and repeating the operation until the sample points in all the data sets are at least distributed to one cluster, and finally merging clusters with similar distances to finish the whole clustering process. The basic form of the mean shift vector is as follows:
where x represents the sample point being "assigned", N (x) represents a neighborhood around x, and xiIs any point within the neighborhood. K (x)i-x) is a kernel function, and the main function of the kernel function is to give different weights to the sample points according to the positions of the sample points in the sample space in the iterative process, so that the convergence speed of the algorithm is increased, and the clustering effect is optimized. The application selects a Gaussian kernel function, its formThe following were used:
where h is the bandwidth of the kernel function, it may be determined by searching through a heuristic algorithm based on the distance between each pair of sample points in the historical data.
Fitting a binary Gaussian distribution model for each cluster generated by clustering, wherein the probability density function form is as follows:
wherein x is [ P, Q ]]TMu and Σ are the mean vector parameter and covariance matrix parameter of the random variable x, respectively, and can be obtained by maximum likelihood estimation, as shown in the following formula:
wherein XiRepresents the observed sample of the random variable x, and N is the total number of observed samples.
And step 3: the total clustering number of the mean shift algorithm is not set to be K, and the kth cluster is marked as ck. C is tokEach sample point in the cluster is substituted into the probability density function expression of the Gaussian distribution model corresponding to the cluster to obtain different probability densities, and the average value of the different probability densities is calculated and defined as the average probability density level P of the clusterkThe expression is as follows:
and 4, step 4: the power data time sequence of the environmental protection equipment is regularly collected and recorded as S ═ S1,s2,…,sTIn which s isi=[Pi,Qi]TT is dependent on a preset input monitoring period length. Based on the mean probability density level P defined in step 3, each point s in the input sequence is first definediAt ckMembership b in a clusterk(si):
Next, B(s) is definedi) Is s isiRelative to the membership degree of the electrical characteristics of the environment-friendly equipment, the calculation method is as follows:
B(si)=max{b1(si),b2(si),...,bk(si)},
then, calculating the electrical characteristic membership B (S) of the whole input power sequence relative to the environmental protection equipment by using the following formula:
and marking the threshold value of the membership degree B (S) of the electrical characteristics as sigma, when the membership degree B (S) is more than or equal to sigma, determining that the sequence accords with the electrical characteristics when the environmental protection equipment normally operates, and judging that the state of the environmental protection equipment is normal, otherwise, judging that the environmental protection equipment is abnormal.
An empirical value of the threshold σ is obtained by: dividing historical power data into a plurality of time sequences meeting the algorithm input requirement, calculating the electrical characteristic membership degree of each sequence relative to the environment-friendly equipment, and taking the minimum value as a threshold value sigma.
In the step 1, active power P and reactive power Q are selected as electrical data of the environmental protection equipment to be analyzed, and considering that a plurality of environmental protection equipment have various working states at present and the running power in each working state is different, the historical power data of the environmental protection equipment are clustered by using a mean shift algorithm to obtain clusters with different numerical value distribution characteristics. The algorithm does not need to determine the number of clusters in advance, but performs the following operations: firstly, randomly selecting an initial sample point from a data set, iterating the initial sample point through a mean shift vector to be distributed to a nearby area with higher data density, taking the initial sample point and surrounding sample points as a cluster, then continuously randomly selecting a sample point and repeating the operation until the sample points in all the data sets are at least distributed to one cluster, and finally merging clusters with similar distances to finish the whole clustering process. The basic form of the mean shift vector is as follows:
where x represents the sample point being "assigned", N (x) represents a neighborhood around x, and xiIs any point within the neighborhood. K (x)i-x) is a kernel function, and the main function of the kernel function is to give different weights to the sample points according to the positions of the sample points in the sample space in the iterative process, so that the convergence speed of the algorithm is increased, and the clustering effect is optimized. In the present application, a gaussian kernel function is selected, which has the following form:
where h is the bandwidth of the kernel function, it may be determined by searching through a heuristic algorithm based on the distance between each pair of sample points in the historical data.
Next, a binary gaussian distribution model is fitted to each cluster from which the cluster is generated, with a probability density function of the form:
wherein x is [ P, Q ]]TMu and Σ are the mean vector parameter and covariance matrix parameter of the random variable x, respectively, and can be obtained by maximum likelihood estimation, as shown in the following formula:
wherein XiRepresents the observed sample of the random variable x, and N is the total number of observed samples.
In step 2, the total clustering number of the mean shift algorithm is not set to be K, and the kth cluster is set to be ck. The mean probability density level is defined as follows: c is tokEach sample point in the cluster is substituted into the probability density function expression of the Gaussian distribution model corresponding to the cluster to obtain different probability densities, and the average value of the different probability densities is the average probability density level P of the clusterkThe expression is as follows:
in step 3, the input power sequence of the algorithm is denoted as S ═ S1,s2,…,sTIn which s isi=[Pi,Qi]TT depends on a preset input monitoring period length. Based on the mean probability density level P defined in step 3, each point s in the input sequence is first definediAt ckMembership b in a clusterk(si):
In the formula, the denominator PkHas the effect ofk(si) To prevent the extreme case that the probability density obtained when the samples in the cluster are more dispersed is too small or the probability obtained when the samples are more concentrated is too large, b can be usedk(si) Normalized to the numerical range with the actual probability significance.
Next, to characterize siDefining B(s) according to the degree of coincidence with the overall power characteristic of the environmental protection equipmenti) Is s isiRelative to the membership degree of the electrical characteristics of the environment-friendly equipment, the calculation method is as follows:
B(si)=max{b1(si),b2(si),...,bk(si)},
the maximum value is taken in the formula, so that the sample points at all times are distributed to the most possible power clusters to which the sample points belong as far as possible, various possible running states of the environmental protection equipment are fully considered, and the risk of misjudgment is reduced.
Then, calculating the electrical characteristic membership B (S) of the whole input power sequence relative to the environmental protection equipment by using the following formula:
in step 4, the threshold of the membership degree b(s) of the electrical characteristics is recorded as σ, when b(s) is greater than or equal to σ, the sequence is determined to conform to the electrical characteristics of the environmental protection equipment during normal operation, the state of the environmental protection equipment is normal, otherwise, the environmental protection equipment is determined to be abnormal. The empirical value of the threshold σ is obtained by: dividing historical power data into a plurality of time sequences meeting the algorithm input requirement, calculating the electrical characteristic membership degree of each sequence relative to the environment-friendly equipment, and taking the minimum value as a threshold value sigma.
Optionally, step 1 includes:
selecting active power P and reactive power Q as electrical data of the environmental protection equipment to be analyzed, and clustering historical power data of the environmental protection equipment by using a mean shift algorithm to obtain clusters with different numerical value distribution characteristics;
fitting a binary Gaussian distribution model for each cluster from which the cluster is generated, wherein the probability density function is in the form of:
wherein x is [ P, Q ═ Q]TMu and Σ are the mean vector parameter and covariance matrix parameter of the random variable x, respectively, and can be obtained by maximum likelihood estimation, as shown in the following formula:
wherein XiRepresents the observed sample of the random variable x, and N is the total number of observed samples.
Optionally, step 2 includes:
let K be the total number of clusters of the mean shift algorithm, and c be the kth clusterk;
The mean probability density level is defined as: c is tokSubstituting each sample point into the probability density function expression of the Gaussian distribution model corresponding to the cluster to obtain different probability densities and calculating the average value of the probability densities, wherein the average value is the average probability density level P of the clusterkThe expression is as follows:
optionally, step 3 includes:
let the input power sequence of the algorithm be S ═ S1,s2,…,sTIn which s isi=[Pi,Qi]TT is the input monitoring period length;
defining each point s in the input sequence based on the mean probability density level P defined in step 3iAt ckMembership b in a clusterk(si):
Definition of B(s)i) Is s isiRelative to the membership degree of the electrical characteristics of the environment-friendly equipment, the calculation method is as follows:
B(si)=max{b1(si),b2(si),...,bk(si)},
then, calculating the electrical characteristic membership B (S) of the whole input power sequence relative to the environmental protection equipment by using the following formula:
optionally, the step 4 includes:
and marking the threshold value of the membership degree B (S) of the electrical characteristics as sigma, when the membership degree B (S) is more than or equal to sigma, determining that the sequence accords with the electrical characteristics when the environmental protection equipment normally operates, and judging that the state of the environmental protection equipment is normal, otherwise, judging that the environmental protection equipment is abnormal.
Optionally, the method for obtaining the threshold σ includes:
dividing historical power data into a plurality of time sequences meeting algorithm input requirements;
and calculating the membership degree of the electrical characteristics of each sequence relative to the environmental protection equipment, and taking the minimum value of the membership degrees as a threshold value sigma.
Fig. 1 shows a specific process of the environmental protection equipment status monitoring algorithm provided by the present application. In this embodiment, an electric power data acquisition device is installed in a power supply loop of a large-scale dust removal device in a certain factory, the data sampling frequency is one point per minute, the monitoring time period is set to be 7:30 in the morning to 6:30 in the evening according to the production time arrangement of the factory, and 1320 original data are acquired in total for two days. And a 64-bit Tencent cloud server based on a Linux CentOS 7.6 operating system is deployed at the cloud end and used for receiving, analyzing and uploading data to the database. The reading of data, format conversion and the algorithm mentioned in the specification are all realized by Python 3.
The embodiment sets the monitoring period of the environmental protection equipment to be 1 hour, namely, each input power sequence comprises 60 sampling data. 660 data of the first day are used as a historical data set, and 660 data of the second day are used as test data. In order not to affect the normal operation of the factory production and the environmental protection equipment, the embodiment simulates the abnormal data of the environmental protection equipment in an off-line simulation mode. The specific method comprises the following steps: the test data is divided into 11 time sequences with the length of 60, and then 50% of continuous data in each sequence are randomly set to zero, so that the working condition that the environmental protection equipment is continuously closed for half an hour is simulated.
Fig. 2 shows the clustering result of the historical number based on the mean shift algorithm, and it can be seen that the historical power data of the environment-friendly device in this embodiment can be divided into 3 clusters. And then, calculating the mean vector and covariance matrix parameters of the Gaussian model corresponding to each cluster by using maximum likelihood estimation. And then, sequentially substituting all sample points in each cluster into corresponding probability density function expressions, solving function values corresponding to each sample point and averaging to obtain the average probability density level of the cluster. Table 1 shows the sets of numerical characteristic parameters (μ, Σ, F) of the respective clusters obtained by clustering.
Table 1: numerical characteristic parameter of each cluster
The historical data are also divided into 11 sequences with the length of 60, and the membership degree of each sequence relative to the electrical characteristics of the environmental protection equipment is calculated according to the algorithm steps in the specification, and the result is as follows (two decimal places are reserved):
table 2: degree of membership of historical power sequence to electrical characteristics of environmental protection equipment
From table 2 it can be determined that the empirical value of the membership threshold σ is around 0.76. The abnormal data obtained by simulation according to the method is spliced with the test data of the next day to form a test set of the algorithm, and the sigma is sequentially set to be 0.72,0.73,0.74,0.75,0.76,0.77,0.78,0.79 and 0.80 to test the recognition accuracy of the algorithm under different thresholds, wherein the test results are as follows:
table 3: test result of environment-friendly equipment state monitoring method
From the above results, it can be seen that when the threshold σ is set in the vicinity of the obtained empirical value, any abnormal situation is not missed, and when σ is set higher than 0.77, only at most 1 erroneous judgment is generated. In an actual scene, the environmental protection equipment may be shut down for a short time due to various reasons, and the algorithm should give a certain tolerance to the situation, so that sigma can be slightly lower than an empirical value, and the identification accuracy of a long-time abnormal state is improved under the condition that misjudgment is not generated.
This example fully demonstrates the effectiveness of the monitoring method proposed in this application. The method can help environmental protection departments to effectively attack the bad behaviors of enterprises in environmental protection work, and is helpful to promote the smooth progress of government environmental protection supervision.
The sequence numbers in the above embodiments are merely for description, and do not represent the sequence of the assembly or the use of the components.
The above description is only exemplary of the present application and should not be taken as limiting the present application, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (6)
1. The method for monitoring the state of the environmental protection equipment based on the electrical characteristics is characterized by comprising the following steps:
step 1, acquiring historical electrical data of normally-operated environment-friendly equipment, performing cluster analysis on the historical electrical data, fitting each obtained cluster to obtain a binary Gaussian distribution model, and calculating parameters of the binary Gaussian distribution model;
step 2, calculating the average probability density of each cluster based on the obtained binary Gaussian model distribution parameters;
step 3, regularly collecting the electrical data time sequence of the environmental protection equipment, and calculating the membership degree of the electrical data time sequence and the electrical characteristics of the environmental protection equipment;
and 4, comparing the electrical characteristic membership degree with a preset threshold value, and outputting a monitoring result of the running state of the environmental protection equipment.
2. The electrical characteristic-based environmental protection equipment state monitoring method according to claim 1, wherein the step 1 comprises:
selecting active power P and reactive power Q as electrical data of the environmental protection equipment to be analyzed, and clustering historical power data of the environmental protection equipment by using a mean shift algorithm to obtain clusters with different numerical value distribution characteristics;
fitting a binary Gaussian distribution model for each cluster from which the cluster is generated, wherein the probability density function is in the form of:
wherein x is [ P, Q ═ Q]TMu and Σ are the mean vector parameter and covariance matrix parameter of the random variable x, respectively, and can be obtained by maximum likelihood estimation, as shown in the following formula:
wherein XiRepresents the observed sample of the random variable x, and N is the total number of observed samples.
3. The electrical characteristic-based eco-equipment status monitoring method according to claim 1 or 2, wherein the step 2 comprises:
let K be the total number of clusters of the mean shift algorithm, and c be the kth clusterk;
The mean probability density level is defined as: c is tokSubstituting each sample point into the probability density function expression of the Gaussian distribution model corresponding to the cluster to obtain different probability densities and calculating the average value of the probability densities, wherein the average value is the average probability density level P of the clusterkThe expression is as follows:
4. the electrical characteristic-based eco-equipment status monitoring method according to claim 1 or 3, wherein the step 3 comprises:
let the input power sequence of the algorithm be S ═ S1,s2,…,sTIn which s isi=[Pi,Qi]TT is the input monitoring period length;
defining each point s in the input sequence based on the mean probability density level P defined in step 3iAt ckMembership b in a clusterk(si):
Definition of B(s)i) Is s isiRelative to the membership degree of the electrical characteristics of the environment-friendly equipment, the calculation method is as follows:
B(si)=max{b1(si),b2(si),...,bk(si)},
then, calculating the electrical characteristic membership B (S) of the whole input power sequence relative to the environmental protection equipment by using the following formula:
5. the electrical characteristic-based eco-equipment status monitoring method according to claim 1 or 4, wherein the step 4 comprises:
and marking the threshold value of the membership degree B (S) of the electrical characteristics as sigma, when the membership degree B (S) is more than or equal to sigma, determining that the sequence accords with the electrical characteristics when the environmental protection equipment normally operates, and judging that the state of the environmental protection equipment is normal, otherwise, judging that the environmental protection equipment is abnormal.
6. The electrical characteristic-based eco-equipment status monitoring method according to claim 1 or 5, wherein the method of obtaining the threshold value σ comprises:
dividing historical power data into a plurality of time sequences meeting algorithm input requirements;
and calculating the membership degree of the electrical characteristics of each sequence relative to the environmental protection equipment, and taking the minimum value of the membership degrees as a threshold value sigma.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011230180.4A CN112561230B (en) | 2020-11-06 | 2020-11-06 | Environment-friendly equipment state monitoring method based on electrical characteristics |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011230180.4A CN112561230B (en) | 2020-11-06 | 2020-11-06 | Environment-friendly equipment state monitoring method based on electrical characteristics |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112561230A true CN112561230A (en) | 2021-03-26 |
CN112561230B CN112561230B (en) | 2024-07-26 |
Family
ID=75041667
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011230180.4A Active CN112561230B (en) | 2020-11-06 | 2020-11-06 | Environment-friendly equipment state monitoring method based on electrical characteristics |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112561230B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117743877A (en) * | 2024-02-20 | 2024-03-22 | 江苏雷博微电子设备有限公司 | Intelligent detection method for component faults of glue spraying machine |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201108357D0 (en) * | 2011-05-18 | 2011-06-29 | Onzo Ltd | Event identification |
CN203759523U (en) * | 2014-03-18 | 2014-08-06 | 福建望诚电子有限公司 | Data acquisition device for remote and automatic monitoring of an environmental protection apparatus |
US20150227837A1 (en) * | 2012-09-03 | 2015-08-13 | Isis Innovation Limited | System monitoring |
CN106569056A (en) * | 2016-10-21 | 2017-04-19 | 广州供电局有限公司 | Power transformer fault diagnosis method and diagnosis device |
CN110580502A (en) * | 2019-08-20 | 2019-12-17 | 上海纺织节能环保中心 | Factor hidden Markov load decomposition method based on Gaussian mixture |
CN110738255A (en) * | 2019-10-15 | 2020-01-31 | 和尘自仪(嘉兴)科技有限公司 | device state monitoring method based on clustering algorithm |
CN111553444A (en) * | 2020-05-15 | 2020-08-18 | 东南大学 | Load identification method based on non-invasive load terminal data |
CN111639842A (en) * | 2020-05-20 | 2020-09-08 | 湖北博华自动化系统工程有限公司 | Equipment health evaluation method, evaluation system and equipment health prediction method |
-
2020
- 2020-11-06 CN CN202011230180.4A patent/CN112561230B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201108357D0 (en) * | 2011-05-18 | 2011-06-29 | Onzo Ltd | Event identification |
US20150227837A1 (en) * | 2012-09-03 | 2015-08-13 | Isis Innovation Limited | System monitoring |
CN203759523U (en) * | 2014-03-18 | 2014-08-06 | 福建望诚电子有限公司 | Data acquisition device for remote and automatic monitoring of an environmental protection apparatus |
CN106569056A (en) * | 2016-10-21 | 2017-04-19 | 广州供电局有限公司 | Power transformer fault diagnosis method and diagnosis device |
CN110580502A (en) * | 2019-08-20 | 2019-12-17 | 上海纺织节能环保中心 | Factor hidden Markov load decomposition method based on Gaussian mixture |
CN110738255A (en) * | 2019-10-15 | 2020-01-31 | 和尘自仪(嘉兴)科技有限公司 | device state monitoring method based on clustering algorithm |
CN111553444A (en) * | 2020-05-15 | 2020-08-18 | 东南大学 | Load identification method based on non-invasive load terminal data |
CN111639842A (en) * | 2020-05-20 | 2020-09-08 | 湖北博华自动化系统工程有限公司 | Equipment health evaluation method, evaluation system and equipment health prediction method |
Non-Patent Citations (1)
Title |
---|
王娜;李号彩;张德利;: "混合聚类分析算法在发电设备故障模式识别中的应用", 电力信息与通信技术, no. 12, pages 32 - 35 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117743877A (en) * | 2024-02-20 | 2024-03-22 | 江苏雷博微电子设备有限公司 | Intelligent detection method for component faults of glue spraying machine |
CN117743877B (en) * | 2024-02-20 | 2024-05-03 | 江苏雷博微电子设备有限公司 | Intelligent detection method for component faults of glue spraying machine |
Also Published As
Publication number | Publication date |
---|---|
CN112561230B (en) | 2024-07-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110889088A (en) | Enterprise pollution discharge supervision method assisted by electric model | |
CN110222991B (en) | Metering device fault diagnosis method based on RF-GBDT | |
CN111369168B (en) | Associated feature selection method suitable for multiple regulation and control operation scenes of power grid | |
CN111160791A (en) | Abnormal user identification method based on GBDT algorithm and factor fusion | |
CN112434962B (en) | Enterprise user state evaluation method and system based on power load data | |
CN116937575A (en) | Energy monitoring management system for grid system | |
CN113990054A (en) | Energy storage power station data analysis and early warning system | |
CN116797404A (en) | Intelligent building operation and maintenance supervision system based on big data and data processing | |
CN116976707B (en) | User electricity consumption data anomaly analysis method and system based on electricity consumption data acquisition | |
CN108776276A (en) | Multiplexing electric abnormality detection method and system | |
Xue et al. | Adaptive ultra-short-term wind power prediction based on risk assessment | |
CN117691722B (en) | Lithium battery charging safety monitoring and early warning method and system | |
CN111177128B (en) | Metering big data batch processing method and system based on improved outlier detection algorithm | |
CN118094262B (en) | Load abnormal value detection method based on spatial clustering | |
CN115291006A (en) | Method for identifying environmental protection working condition abnormity of polluted enterprise based on electricity utilization data | |
CN112561230B (en) | Environment-friendly equipment state monitoring method based on electrical characteristics | |
CN116756505B (en) | Photovoltaic equipment intelligent management system and method based on big data | |
CN114049033B (en) | Pollution discharge enterprise monitoring method based on electricity consumption data distribution | |
CN115879726A (en) | Method for mutually optimizing and screening power consumption and emission data of enterprise | |
CN115809805A (en) | Power grid multi-source data processing method based on edge calculation | |
CN113822587B (en) | Factory capacity evaluation method based on bus current data | |
CN115545240A (en) | Method, system, equipment and medium for diagnosing abnormal line loss of low-voltage distribution network transformer area | |
CN101923605B (en) | Wind pre-warning method for railway disaster prevention | |
Xie et al. | Energy System Time Series Data Quality Maintenance System Based on Data Mining Technology | |
Lingang et al. | Research on integrated calculation method of theoretical line loss of MV and LV distribution Network based on Adaboost integrated learning |
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 |