CN116383262B - Power plant SIS system-based energy consumption data management platform - Google Patents
Power plant SIS system-based energy consumption data management platform Download PDFInfo
- Publication number
- CN116383262B CN116383262B CN202310626302.9A CN202310626302A CN116383262B CN 116383262 B CN116383262 B CN 116383262B CN 202310626302 A CN202310626302 A CN 202310626302A CN 116383262 B CN116383262 B CN 116383262B
- Authority
- CN
- China
- Prior art keywords
- data
- power plant
- mapping
- energy consumption
- space
- 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.)
- Active
Links
- 238000005265 energy consumption Methods 0.000 title claims abstract description 127
- 238000013523 data management Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 claims abstract description 65
- 238000007405 data analysis Methods 0.000 claims abstract description 20
- 238000010276 construction Methods 0.000 claims abstract description 17
- 238000001914 filtration Methods 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims abstract description 11
- 238000013507 mapping Methods 0.000 claims description 125
- 239000011159 matrix material Substances 0.000 claims description 97
- 230000002159 abnormal effect Effects 0.000 claims description 24
- 238000000605 extraction Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 13
- 230000005856 abnormality Effects 0.000 claims description 11
- 238000012546 transfer Methods 0.000 claims description 11
- 238000013506 data mapping Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 4
- 241000003980 Pithites Species 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 21
- 238000004458 analytical method Methods 0.000 abstract description 18
- 238000012544 monitoring process Methods 0.000 abstract description 10
- 238000001514 detection method Methods 0.000 description 15
- 238000005457 optimization Methods 0.000 description 8
- 238000007726 management method Methods 0.000 description 5
- 238000012795 verification Methods 0.000 description 4
- 238000010606 normalization Methods 0.000 description 3
- 238000013500 data storage Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- 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
- G06F16/2457—Query processing with adaptation to user needs
-
- 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
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- 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
- G06F16/2477—Temporal data queries
-
- 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
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of power plant monitoring, in particular to an energy consumption data management platform based on a power plant SIS system, which comprises the following components: an SIS system and an energy consumption data management system; the SIS system is used for collecting power plant data and storing the power plant data; the energy consumption data management platform comprises: the data filter construction unit is used for acquiring power plant data in a set time range as historical data, carrying out data analysis on the historical data to find out energy consumption related data in the historical data, and constructing a data filter based on the energy consumption related data; the data filter periodically acquires power plant data from the SIS system, and performs data filtering on the acquired power plant data to obtain filtered data. The invention can efficiently process and analyze large-scale power plant data and improve the efficiency of energy consumption data processing and analysis.
Description
Technical Field
The invention belongs to the technical field of power plant monitoring, and particularly relates to an energy consumption data management platform based on a power plant SIS system.
Background
In conventional power plant energy consumption data management, a SIS (Supervisory Information System, plant-level monitoring system) system is typically used to collect and store power plant data. The SIS system is a real-time monitoring system and can acquire various information such as power plant equipment data, operation data, environment data and the like. However, the conventional SIS system has some limitations in terms of energy consumption data management, such as limited ability of data analysis and anomaly detection, and fails to provide comprehensive and accurate energy consumption related information, resulting in difficulty in optimizing power plants in terms of energy consumption and diagnosing problems.
In addition, conventional energy consumption data management methods often rely on manual data processing and analysis, lacking automated and intelligent energy consumption data management capabilities. In large-scale power plants, the data volume is large and complex, the manual processing and analysis is time-consuming and labor-consuming, and errors are easily generated. In addition, the traditional method has certain limitation on detection and identification of abnormal data, and may cause delay of problems or incapacity of timely solving the problems.
Accordingly, there remain some problems in the prior art that need to be solved. The main problems include:
the data processing and analysis efficiency is low: traditional energy consumption data management methods rely on manual processing and analysis, which is time-consuming and labor-consuming. In large-scale power plants, the data volume is huge, and is difficult to process and analyze efficiently, limiting the efficiency and accuracy of energy consumption data management.
Data anomaly detection capability is limited: the traditional abnormality detection method has a certain limitation on abnormality detection of power plant energy consumption data, and can not discover and solve abnormal conditions in time, so that normal operation and optimization of the power plant are affected.
Data storage and management are imperfect: the traditional SIS system has some defects in the aspects of data storage and management, cannot provide comprehensive and accurate energy consumption related information, and is difficult to meet the requirements of power plant energy consumption data management.
In view of the foregoing, there are some problems and challenges in energy consumption data management in the prior art, and a new energy consumption data management platform is needed to improve the efficiency of data processing and analysis, enhance the capability of anomaly detection, and provide comprehensive and accurate energy consumption related information to support the operation and energy optimization of a power plant.
Disclosure of Invention
The invention mainly aims to provide an energy consumption data management platform based on a power plant SIS system, which can efficiently process and analyze large-scale power plant data and improve the efficiency of energy consumption data processing and analysis through the construction of a data filter and a data mapper. And secondly, the energy consumption data management platform can timely discover and solve abnormal conditions in the energy consumption data through an intelligent abnormal detection algorithm, so that the running reliability and efficiency of the power plant are improved.
In order to solve the problems, the technical scheme of the invention is realized as follows:
an energy consumption data management platform based on a power plant SIS system, comprising: an SIS system and an energy consumption data management system; the SIS system is used for collecting power plant data and storing the power plant data; the energy consumption data management platform comprises: the data filter construction unit is used for acquiring power plant data in a set time range as historical data, carrying out data analysis on the historical data to find out energy consumption related data in the historical data, and constructing a data filter based on the energy consumption related data; the data filter periodically acquires power plant data from the SIS system, and performs data filtering on the acquired power plant data to obtain filtered data; the data mapping unit is used for carrying out data clustering on the filtered data, finding a plurality of data clustering centers and constructing a data mapper based on each data clustering center; the data mapper is used for mapping the extended feature space of the filtered data to obtain mapping data under each extended feature space; the feature center of each extended feature space corresponds to a data clustering center; the data analyzer is used for carrying out data analysis on the mapping data in each extended feature space so as to judge whether data abnormality occurs, and screening out abnormal data as an abnormal result under the condition that the data abnormality occurs; the data searching device is used for responding to a data searching instruction of a user, carrying out one-dimensional mapping on the mapping data of each expansion feature space in a space filling curve mode to obtain one-dimensional mapping data, and carrying out data searching based on the one-dimensional mapping data to obtain a searching result.
Further, the power plant data acquired by the SIS system includes: plant data DE, plant operation data OP and plant environment data EN; the power plant equipment data are factory data of the power plant equipment; the power plant equipment operation data are operation data of the power plant equipment during operation; the power plant environment data are environment data of a power plant; when the SIS system stores power plant data, the following data tuples are constructed for storage:
Q i ＝{DE i |{T 1 ，OP 1 ，EN 1 }，{T 2 ，OP 2 ，EN 2 }，...，{T n ，OP n ，EN n }}；
wherein Q is i I is a data tuple, i is a serial number of power plant equipment; t (T) n To obtain the time of the power plant data, n is the number of times of obtaining the power plant data.
Further, the method for obtaining the power plant data in the set time range as the historical data and performing data analysis on the historical data to find the energy consumption related data in the historical data includes: the obtained history data is represented by an input matrix M, and each element in the input matrix M is a data tuple Q i The method comprises the steps of carrying out a first treatment on the surface of the Extracting matrix components of the input matrix M to obtain a component matrix C for representing the energy consumption related data, wherein the method comprises the following specific processes: selecting a set of scale parameters sigma 1 ，σ 2 ，...，σ k The method comprises the steps of carrying out a first treatment on the surface of the Then, a Gaussian filter is applied to the input matrix M to generate a set of scale-space matrices
L＝L 1 ，L 2 ，...，L k ；
Wherein each scale space matrixExpressed at the scale sigma i A lower fuzzy matrix; the formula for constructing the scale space is as follows:
where (x, y) is the pixel coordinates in the matrix,is a gaussian kernel function for performing a smoothing operation on the input matrix M; for each scale space matrix L i Detecting keypoints and computing feature descriptors using a feature extraction algorithm; assume that a set of feature points is obtained:
P＝p 1 ，p 2 ，...，p q ；
wherein each feature point p j ＝(x j ，y j ，σ j ) Including location and scale information; generating a component matrix according to the position and the scale of the characteristic points of the pre-established energy consumption related dataWherein each column represents a component of a feature point; for each feature point p j Each row of the component matrix represents a location, scale, direction, and feature descriptor of the feature point; assuming that the component matrix C has q columns, representing q feature points; then the j-th column of C is represented as a vector v j Wherein:
v j ＝[x j ，y j ，σ j ，θ j ，d j1 ，d j2 ，...，d jn ]；
wherein x is j ，y j ，σ j Respectively represent characteristic points p j And the position and scale, θ j Representing the characteristic point p j Direction d of (d) j1 ，d j2 ，...，d jn Representing the characteristic point p j N components in the feature descriptor of (2); the size of the component matrix C is p multiplied by q, wherein p represents the attribute dimension of the feature points, and q represents the number of the extracted feature points; the position and the scale of the characteristic points of the energy consumption related data are obtained by acquiring the existing data And calculating the energy consumption related data.
Further, the method for calculating the feature descriptor comprises the following steps: constructing a matrix area around the feature points, and dividing the matrix area into a plurality of sub-areas; then, for each sub-region, its local gradient direction operator is calculated to count the frequency with which the gradient direction falls in each direction interval. The local gradient direction operators of the subregions are connected to form a feature descriptor.
Further, the data filter construction unit, based on the energy consumption related data, constructs the data filter by the method comprising: each column in the component matrix C is considered as an array, and a hash operator is assigned to each array to construct a data filter.
Further, the data filter periodically acquires power plant data from the SIS system, and performs data filtering on the acquired power plant data, so as to obtain filtered data, which comprises the following steps: when power plant data are periodically acquired from an SIS system, each acquired power plant data are randomly inserted into an array of a data filter to serve as an element, a hash arithmetic unit of each array in the data filter is used for calculating a hash value again, if the normalized mean value of the calculated hash value is equal to the normalized mean value of the hash value before the power plant data are inserted, the power plant data are filtered, otherwise, the power plant data are passed, and when the acquisition of the power plant data from the SIS system is finished in the period, the passed power plant data are reserved to serve as filtered data.
Further, the data mapping unit constructs a data mapper, and the method for mapping the extended feature space of the filtered data by the data mapper to obtain the mapped data under each extended feature space includes: clustering the filtered data X by using a hierarchical clustering algorithm to obtain K clustering clusters; for each cluster k, its cluster center c is calculated k The method comprises the steps of carrying out a first treatment on the surface of the Initializing spatial mapping vectorsAnd a space mapping covariance matrix P 0 The method comprises the steps of carrying out a first treatment on the surface of the For each of the filtered data XData sample xi: initializing the current spatial map +.> And a current spatial mapping covariance matrix Pi0; for the mapping direction t: transfer function using spatial mapping>And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping:
transferring noise covariance matrix Q using spatial mapping t And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping covariance matrix Pit:
Pit＝FtPit-Ft T +Q t ；
using an observed noise covariance matrix R t And an observation jacobian matrix Ht, calculating an observation prediction
Using an observed noise covariance matrix R t And observing the Jacobian matrix Hit, and calculating the Kalman gain K it ：
Kit＝PitHit T (HitPitHit T +R t ) -1 ；
Using the observation vector zit, the residual yit is calculated:
updating spatial mapping
Finally, the mapping data under each expansion feature space is obtained as
Further, the data analyzer performs data analysis on the mapping data in each extended feature space to determine whether a data anomaly occurs, and the method includes: mapping data for each extended feature spaceCalculate its and cluster center c k Is a Euclidean distance d; comparing the Euclidean distance d with a preset judging value, judging that the data is abnormal if the Euclidean distance d is larger than the judging value, otherwise, judging that the data is not abnormal.
Further, the data finder performs one-dimensional mapping on mapping data of each extended feature space by using a space filling curve mode to obtain one-dimensional mapping data, and performs data finding based on the one-dimensional mapping data to obtain a finding result, and the method comprises the following steps: mapping data in each extended feature spaceMapping the space filling curve into a one-dimensional space by using a space filling curve mode to obtain one-dimensional mapping data dit; repeating the above steps, mapping data +_for each sample>Obtaining a one-dimensional mapping data set di0, d i1 ，...，d i(t+1) The method comprises the steps of carrying out a first treatment on the surface of the For given query data q, mapping it to obtain one-dimensional query value d q The method comprises the steps of carrying out a first treatment on the surface of the From one-dimensional maps in a one-dimensional map datasetShooting data, searching and inquiring the value d in a one-dimensional space by using binary search q The closest mapping data is used as the search result.
Further, the space filling curve is one of a Hilbert curve or a Z curve.
The energy consumption data management platform based on the power plant SIS system has the following beneficial effects:
the energy consumption data processing and analysis efficiency is improved: conventional energy consumption data management methods typically rely on manual processing and analysis, which is time-consuming and labor-intensive. The invention introduces the concepts of the data filter and the data mapper, and can efficiently process large-scale power plant data through an automatic data processing and analyzing algorithm. The data filter can rapidly screen out the data related to energy consumption, and avoids the complexity and errors of manual processing. The data mapper may then map the data into an extended feature space, providing a more comprehensive and accurate representation of the data. Therefore, the energy consumption data management platform can greatly improve the efficiency of data processing and analysis, and provides timely and accurate data support for power plant decision-making.
Enhancing the energy consumption data anomaly detection capability: abnormal conditions in the energy consumption data may lead to power plant operating problems or energy waste. The energy consumption data management platform can better identify and detect abnormal conditions in the energy consumption data through the construction of the data filter and the data mapper. The data filter can rapidly and accurately identify the data related to the energy consumption through filtering and clustering algorithms, and the data mapper can timely find abnormal changes in the data through updating and analyzing the mapping data. Through the comprehensive abnormality detection mechanism, the energy consumption data management platform can monitor the energy consumption condition of the power plant in real time, discover and solve abnormality problems in time, and improve the operation efficiency and the energy utilization rate of the power plant.
Intelligent energy consumption data management is realized: the energy consumption data management platform of the invention realizes intelligent energy consumption data management by utilizing advanced data processing, analysis and anomaly detection algorithms. By constructing the data filter and the data mapper, the power plant data can be automatically processed and analyzed, and key energy consumption related information can be extracted from the power plant data. The intelligent data management mode can reduce the manual work load, improve the accuracy and consistency of data processing, and simultaneously provide more reliable and timely support for power plant decision.
Drawings
Fig. 1 is a schematic structural diagram of an energy consumption data management platform based on a power plant SIS system according to an embodiment of the present invention.
Detailed Description
The energy consumption data management platform based on the power plant SIS system can improve the efficiency of energy consumption data processing and analysis, enhance the capability of energy consumption data anomaly detection, provide comprehensive and accurate energy consumption related information, support the energy optimization and energy conservation and emission reduction of the power plant, and realize intelligent energy consumption data management. The method provides better energy management and decision basis for the power plant, and promotes sustainable development of the power plant and improvement of energy utilization efficiency.
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
The following will describe in detail.
An energy consumption data management platform based on a power plant SIS system, comprising: an SIS system and an energy consumption data management system; the SIS system is used for collecting power plant data and storing the power plant data; the energy consumption data management platform comprises: the data filter construction unit is used for acquiring power plant data in a set time range as historical data, carrying out data analysis on the historical data to find out energy consumption related data in the historical data, and constructing a data filter based on the energy consumption related data; the data filter periodically acquires power plant data from the SIS system, and performs data filtering on the acquired power plant data to obtain filtered data; the data mapping unit is used for carrying out data clustering on the filtered data, finding a plurality of data clustering centers and constructing a data mapper based on each data clustering center; the data mapper is used for mapping the extended feature space of the filtered data to obtain mapping data under each extended feature space; the feature center of each extended feature space corresponds to a data clustering center; the data analyzer is used for carrying out data analysis on the mapping data in each extended feature space so as to judge whether data abnormality occurs, and screening out abnormal data as an abnormal result under the condition that the data abnormality occurs; the data searching device is used for responding to a data searching instruction of a user, carrying out one-dimensional mapping on the mapping data of each expansion feature space in a space filling curve mode to obtain one-dimensional mapping data, and carrying out data searching based on the one-dimensional mapping data to obtain a searching result.
The SIS system is a factory-level monitoring system; a plant-level monitoring system is a system for monitoring and controlling the operation of an entire plant. The system integrates a plurality of subsystems and devices to acquire data and information of each part of the power plant in real time and perform centralized management and control on the data and information. These subsystems and devices include generators, transmission lines, transformers, switching devices, and the like. The plant-level monitoring system may provide monitoring and management of plant operating conditions, energy consumption data, equipment failures, and the like.
In the energy consumption data management platform, plant-level monitoring systems (including SIS systems) function to collect and store plant data. These data may include energy consumption data, operating state data, and safety-related data. The energy consumption data management platform helps a user to optimize and monitor the energy consumption by analyzing, processing and managing the data, and can detect data abnormality and provide a data searching function.
The energy consumption data management platform of the invention comprises two main components, namely an SIS system and an energy consumption data management system. The SIS system is responsible for collecting various data of the power plant, such as energy consumption data, operating state data, etc., and storing these data for subsequent processing and analysis. The energy consumption data management system is an integrated software system for processing, analyzing and managing the collected data.
The main function of the data filter construction unit is to acquire power plant data within a set time range as historical data and to perform data analysis on the historical data. Through data analysis, it can find the data related to energy consumption in the historical data, and construct the data filter based on the data related to energy consumption. The data filter may be used to screen and extract specific data for subsequent processing and analysis.
The data filter is constructed based on the energy consumption related data and is used for filtering the acquired data when the power plant data is periodically acquired from the SIS system. The data obtained after filtering is called filtered data, and the data is subjected to pretreatment and screening, so that the data amount can be reduced, and meanwhile, important information related to energy consumption is reserved.
The data mapping unit is mainly used for carrying out data clustering on the filtered data. It can categorize similar data points into different data cluster centers, thereby finding multiple data cluster centers. Based on each data cluster center, a data mapper may be constructed.
The data mapper is used for performing extended feature space mapping on the filtered data. It maps the filtered data into each extended feature space and gets the mapped data under each extended feature space. The feature center of each extended feature space corresponds to a data clustering center, so that the features of the data can be better represented and described.
The data analyzer is a module for data analysis of the mapping data at each extended feature space. The method can judge whether the mapping data has abnormal conditions, and when the data is abnormal, the abnormal data can be screened out as an abnormal result.
The data finder is used for responding to the data finding instruction of the user. The method adopts a space filling curve mode to map the mapping data of each expansion feature space in one dimension, and maps the multidimensional data into one dimension space. This can simplify the data search and retrieval process. Based on the one-dimensional mapping data, the data finder can perform efficient data finding operation and return finding results to the user.
On the basis of the above embodiment, the power plant data acquired by the SIS system includes: plant data DE, plant operation data OP and plant environment data EN; the power plant equipment data are factory data of the power plant equipment; the power plant equipment operation data are operation data of the power plant equipment during operation; the power plant environment data are environment data of a power plant; when the SIS system stores power plant data, the following data tuples are constructed for storage:
Q i ＝{DE i |{T 1 ,OP 1 ,EN 1 },{T 2 ,OP 2 ,EN 2 },…,{T n ,OP n ,EN n }}；
wherein Q is i I is a data tuple, i is a serial number of power plant equipment; t (T) n To obtain the time of the power plant data, n is the number of times of obtaining the power plant data.
By constructing such a data tuple, the SIS system can organize and store plant, plant operating, and environmental data for the plant. The storage mode can realize the association and the tracing of the data at different time points, and is convenient for subsequent data analysis and processing. In the energy consumption data management platform, the stored data can be further utilized and analyzed to realize the functions of energy consumption optimization, anomaly detection, data search and the like.
On the basis of the above embodiment, the method for obtaining the power plant data in the set time range as the historical data and performing data analysis on the historical data to find the energy consumption related data in the historical data includes: the obtained history data is represented by an input matrix M, and each element in the input matrix M is a data tuple Q i The method comprises the steps of carrying out a first treatment on the surface of the Extracting matrix components of the input matrix M to obtain a component matrix C for representing the energy consumption related data, wherein the method comprises the following specific processes: selecting a set of scale parameters sigma 1 ，σ 2 ，...，σ k The method comprises the steps of carrying out a first treatment on the surface of the Then, for input moment The matrix M applies a gaussian filter to generate a set of scale space matrices:
L＝L 1 ，L 2 ，...，L k ；
wherein each scale space matrixExpressed at the scale sigma i A lower fuzzy matrix; the formula for constructing the scale space is as follows:
where (x, y) is the pixel coordinates in the matrix,is a gaussian kernel function for performing a smoothing operation on the input matrix M; for each scale space matrix L i Detecting keypoints and computing feature descriptors using a feature extraction algorithm; assume that a set of feature points is obtained:
P＝p 1 ，p 2 ，...，p q ；
wherein each feature point p j ＝(x j ，y j ，σ j ) Including location and scale information; generating a component matrix according to the position and the scale of the characteristic points of the pre-established energy consumption related dataWherein each column represents a component of a feature point; for each feature point p j Each row of the component matrix represents a location, scale, direction, and feature descriptor of the feature point; assuming that the component matrix C has q columns, representing q feature points; then the j-th column of C is represented as a vector v j Wherein:
v j ＝[x j ，y j ，σ j ，θ j ，d j1 ，d j2 ，...，d jn ]；
wherein x is j ，y j ，σ j Respectively represent characteristic points p j And the position and scale, θ j Representing the characteristic point p j Direction d of (d) j1 ，d j2 ，...，d jn Representing the characteristic point p j N components in the feature descriptor of (2); the size of the component matrix C is p multiplied by q, wherein p represents the attribute dimension of the feature points, and q represents the number of the extracted feature points; the position and the scale of the characteristic points of the energy consumption related data are obtained by acquiring the existing energy consumption related data and calculating.
Specifically, the data filter construction unit may acquire historical data of the power plant within a set time range, and analyze the historical data to find the energy consumption related data. The process involves steps of scale space construction, feature extraction, selection of energy consumption related data, construction of a data filter, and the like, to help achieve extraction and management of the energy consumption related data. The method can improve the efficiency and accuracy of the energy consumption data management platform and support the decision and optimization work of users in the aspect of power plant energy consumption.
Selecting a set of suitable scale parameters sigma 1 ，σ 2 ，...，σ k The purpose of (2) is to be able to capture features and variations of different scales in the data during the scale space construction and feature extraction process. This allows a more comprehensive view to be obtained in the analysis, understanding the characteristics of the data from multiple scales, and better enabling the information related to energy consumption to be distinguished and extracted.
The process and steps for selecting the scale parameters are as follows:
understanding data characteristics: first, the characteristics of the data need to be analyzed and understood. Knowledge of the scale range, trend, and important features of the data is the basis for selecting scale parameters.
Setting initial scale parameters: an initial set of scale parameters is selected based on an understanding of the data characteristics. These initial parameters should cover the different scale ranges that may exist in the data.
And (3) verification and adjustment: and carrying out scale space construction and feature extraction on the data by using the selected initial scale parameters, and then evaluating the result. The scale parameters may be adjusted based on the quality of the results and the degree to which they are expected.
Repeating the iteration: the adjustment and optimization of the scale parameters can be performed in an iterative manner as required. And (3) repeating the application step, and further adjusting the scale parameters according to the evaluation result until the optimal result is reached.
Evaluating and selecting optimal parameters: after a series of iterations is completed, the best combination of scale parameters is selected based on the evaluation results. The optimal parameters should be able to best capture the features and variations in the data while meeting the needs of a particular application.
The spacing, number, and extent of dimensions may need to be considered in selecting the scale parameters. One common approach is to use an equal set of scale parameters, e.g., σ 1 ，σ 2 ，...，σ k ＝2 i Where i is an integer, which can be appropriately adjusted according to the characteristics and range of the data.
For each scale space matrix L i The method and steps for detecting key points and calculating feature descriptors using a feature extraction algorithm are as follows:
and (3) key point detection: in the scale space matrix L i Feature extraction algorithms, such as SIFT, SURF, or ORB, are applied above to detect keypoints. The algorithms detect the local features, gradient information, corner points and the like of the matrix and determine the positions and scales of the key points.
Positioning and refining key points: and (5) carrying out accurate positioning and refinement on the detected key points. This step aims to improve the accuracy and stability of the key points. Common methods include techniques such as sub-pixel accurate positioning, scale space interpolation, etc. to improve the accuracy of the location and scale of the keypoints.
Feature descriptor calculation: in the local matrix area around each keypoint, feature descriptors of the keypoint are calculated using a feature extraction algorithm. A feature descriptor is a compact representation that describes the features of a matrix in the vicinity of a keypoint. The specific calculation method will vary depending on the feature extraction algorithm used, but generally comprises the steps of:
the matrix area around the keypoints is divided into sub-areas or blocks.
Gradient calculations are performed on pixels within each sub-region or block to capture the local directionality and texture characteristics of the matrix. Based on the result of the gradient calculation, a feature vector or descriptor is generated, which contains local feature information of the region around the key point.
Normalization of feature descriptors: to ensure invariance and comparability of feature descriptors, the computed feature descriptors are typically normalized or normalized. This may be achieved by performing an average zeroing or unitizing operation on the feature vector.
Through the above steps, for each scale space matrix L i The feature extraction algorithm detects the keypoints and calculates feature descriptors for the keypoints.
The method and the step of pre-establishing the position and the scale of the characteristic points of the energy consumption related data comprise the following steps:
data acquisition and preparation: and collecting energy consumption data of the power plant, and preprocessing and preparing according to actual requirements. The quality and integrity of the data is ensured for subsequent feature point analysis.
Defining characteristic points: and defining the concept and the attribute of the energy consumption related characteristic points according to the characteristics and the targets of the energy consumption data of the power plant. This may be based on a law of change in energy consumption, an abnormal pattern of energy consumption, or other characteristics related to energy consumption.
Feature point location selection: and determining the position of the characteristic point in the power plant data according to the definition of the energy consumption related characteristic point. This may be selected based on the location of peaks, troughs, changing discontinuities, or other key feature points of the energy consumption curve.
Feature point scale selection: and selecting a proper scale for each energy consumption related characteristic point. The scale may be expressed as a size of a time window, a time span of energy consumption changes, or a scale measure of other energy consumption related features.
Calibration and verification: calibration and verification is performed to ensure that the pre-established location and scale of the energy consumption related feature points match the characteristics of the actual data. This may include discussion with domain experts, comparison with actual data, and verification of energy consumption analysis results in specific scenarios.
On the basis of the above embodiment, the method for calculating the feature descriptor includes: constructing a matrix area around the feature points, and dividing the matrix area into a plurality of sub-areas; then, for each sub-region, its local gradient direction operator is calculated to count the frequency with which the gradient direction falls in each direction interval. The local gradient direction operators of the subregions are connected to form a feature descriptor.
Constructing a matrix area around the feature points:
a matrix region of a fixed size is defined based on the detected positions of the feature points, the region encompassing a local image region in the vicinity of the feature points.
Dividing subareas:
the matrix area is divided into a number of sub-areas. Typically, the sub-regions are divided into grid-like tiles. The number and size of the sub-regions may be adjusted according to specific requirements.
Calculating a local gradient direction operator:
for each sub-region, its local gradient direction operator is calculated. This can be achieved by calculating the gradient magnitude and gradient direction for each pixel. Common gradient operators include Sobel, prewitt or Scharr operators.
Counting gradient direction frequency:
and counting the local gradient direction operators of each sub-region into a direction interval. The direction intervals are typically divided into a fixed number of cells, for example between 0 ° and 180 ° into 8 direction intervals.
Feature descriptor generation:
the local gradient direction frequencies of the sub-regions are connected together to form a feature descriptor vector. The dimension of this vector depends on the number of sub-regions and the number of directional intervals per sub-region.
On the basis of the above embodiment, the data filter construction unit, based on the energy consumption related data, constructs a data filter by a method including: each column in the component matrix C is considered as an array, and a hash operator is assigned to each array to construct a data filter.
On the basis of the above embodiment, the method for periodically obtaining power plant data from the SIS system by the data filter and performing data filtering on the obtained power plant data to obtain filtered data includes: when power plant data are periodically acquired from an SIS system, each acquired power plant data are randomly inserted into an array of a data filter to serve as an element, a hash arithmetic unit of each array in the data filter is used for calculating a hash value again, if the normalized mean value of the calculated hash value is equal to the normalized mean value of the hash value before the power plant data are inserted, the power plant data are filtered, otherwise, the power plant data are passed, and when the acquisition of the power plant data from the SIS system is finished in the period, the passed power plant data are reserved to serve as filtered data.
Preparation of the component matrix C:
it is ensured that the component matrix C has been correctly generated, wherein each column represents the components of a feature point, such as information of position, scale, orientation and feature descriptors.
Constructing an array:
each column in the component matrix C is considered as an array, where each array element corresponds to a component value of a feature point in the column.
Hash operator allocation:
a hash operator is assigned to each array. A hash operator is a data structure for mapping input data to hash values of a fixed size.
And (3) constructing a data filter:
a data filter is built for each of the number sets using a hash operator. A specific construction process may involve taking each array element as input and calculating by a hash operator a corresponding hash value. These hash values will be used to construct a data filter to enable filtering and extraction of plant data.
Through the steps, the data filter can acquire power plant data from the SIS system and filter the acquired data to obtain filtered energy consumption related data. Such data filters may be used for subsequent data analysis, anomaly detection, or other tasks for energy consumption management and optimization.
On the basis of the above embodiment, the method for obtaining the mapping data under each extended feature space by the data mapping unit, constructing a data mapper, and performing extended feature space mapping on the filtered data by the data mapper includes: clustering the filtered data X by using a hierarchical clustering algorithm to obtain K clustering clusters; for each cluster k, its cluster center c is calculated k The method comprises the steps of carrying out a first treatment on the surface of the Initializing spatial mapping vectorsAnd a space mapping covariance matrix P 0 The method comprises the steps of carrying out a first treatment on the surface of the For each data sample xi in the filtered data X: initializing the current spatial map +.>And a current spatial mapping covariance matrix Pi0; for the mapping direction t: transfer function using spatial mapping>And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping:
transferring noise covariance matrix Q using spatial mapping t And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping covariance matrix Pit:
Pit＝FtPit-Ft T +Q t ；
using an observed noise covariance matrix R t And an observation jacobian matrix Ht, calculating an observation prediction
Using an observed noise covariance matrix R t And observing the Jacobian matrix Hit, and calculating the Kalman gain K it ：
Kit＝PitHit T (HitPitHit T +R t ) -1 ；
Using the observation vector zit, the residual yit is calculated:
updating spatial mapping
Finally, the mapping data under each expansion feature space is obtained as
Hierarchical clustering algorithm: similar data samples are grouped into the same cluster by clustering the filtered data. The purpose of this is to identify the intrinsic structure and pattern in the data for subsequent analysis and processing.
Calculating a clustering center: for each cluster, its cluster center, i.e., the average or other statistic of the data samples within the cluster, is calculated. The cluster center represents the characteristics of the cluster and can be used as an important reference in the data mapper.
Spatial mapping transfer: and updating mapping data according to the prior information and the observation information by using a method of spatial mapping transfer function and Kalman filtering. The purpose of this is to obtain more accurate and complete mapping data for each extended feature space.
Updating the spatial mapping: according to the Kalman gain and the observation residual, the formula is updated:
and carrying out iterative updating on the space mapping. In this way, the accuracy and stability of the mapping data can be continuously optimized.
Through these steps, more valuable features can be extracted from the filtered data and more comprehensive and accurate mapping data generated under each extended feature space. The mapping data can be used for subsequent tasks such as data analysis, anomaly detection, data search and the like, helps understand the energy consumption behavior of the power plant, discovers potential problems, and supports data-driven decision making and optimization.
On the basis of the above embodiment, the data analyzer performs data analysis on the mapping data in each extended feature space to determine whether a data anomaly has occurred, and the method includes: mapping data for each extended feature spaceCalculating the Euclidean distance d between the clustering center ck and the clustering center; comparing the Euclidean distance d with a preset judging value, judging that the data is abnormal if the Euclidean distance d is larger than the judging value, otherwise, judging that the data is not abnormal.
On the basis of the above embodiment, the data finder performs one-dimensional mapping on the mapping data of each extended feature space by using a space filling curve mode to obtain one-dimensional mapping data, and performs data finding based on the one-dimensional mapping data to obtain a finding result, and the method comprises the following steps: mapping data in each extended feature spaceMapping the space filling curve into a one-dimensional space by using a space filling curve mode to obtain one-dimensional mapping data dit; repeating the above steps, mapping data +_for each sample>Obtaining a one-dimensional mapping data set di0, d i1 ，...，d i(t+1) The method comprises the steps of carrying out a first treatment on the surface of the For given query data q, mapping it to obtain one-dimensional query value d q The method comprises the steps of carrying out a first treatment on the surface of the Searching and inquiring the value d in a one-dimensional space by using binary search according to one-dimensional mapping data in the one-dimensional mapping data set q The closest mapping data is used as the search result.
On the basis of the above embodiment, the space filling curve is one of a Hilbert curve or a Z curve.
The Hilbert curve is a commonly used space-filling curve for mapping a multi-dimensional space into a one-dimensional space. The method has the characteristics of order preservation and continuity, and can be used for maintaining the spatial relationship of adjacent data points. The following will be describedSpecific steps of mapping into one-dimensional space:
according toAnd determining the coordinate range of the one-dimensional space. Assume that the range of one-dimensional space is [0, L]Where L is the maximum in one dimension.
Will beThe value of each dimension of (1) is mapped to [0, L ]]Within a range of (2). Normalization can be performed using linear transformation or normalization, etc., to ensure that the data is within a specified range.
Will normalizeConsider the coordinates of a point in D-dimensional space.
And mapping the coordinate points in the D-dimensional space to the coordinate points in the one-dimensional space according to an algorithm of the Hilbert curve. The specific Hilbert curve algorithm may refer to the relevant literature or to the open source code, for example, how to calculate the sequential values of the Hilbert curve.
Obtaining a mapped one-dimensional coordinate point d it Its value is at [0, L ]Within a range of (2).
Repeating the above steps for mapping data for each sampleObtaining a one-dimensional mapping data set di0, d i1 ，...，d i(t+1) 。
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. An energy consumption data management platform based on a power plant SIS system, which is characterized by comprising: an SIS system and an energy consumption data management system; the SIS system is used for collecting power plant data and storing the power plant data; the energy consumption data management platform comprises: the data filter construction unit is used for acquiring power plant data in a set time range as historical data, carrying out data analysis on the historical data to find out energy consumption related data in the historical data, and constructing a data filter based on the energy consumption related data; the data filter periodically acquires power plant data from the SIS system, and performs data filtering on the acquired power plant data to obtain filtered data; the data mapping unit is used for carrying out data clustering on the filtered data, finding a plurality of data clustering centers and constructing a data mapper based on each data clustering center; the data mapper is used for mapping the extended feature space of the filtered data to obtain mapping data under each extended feature space; the feature center of each extended feature space corresponds to a data clustering center; the data analyzer is used for carrying out data analysis on the mapping data in each extended feature space so as to judge whether data abnormality occurs, and screening out abnormal data as an abnormal result under the condition that the data abnormality occurs; the data searching device is used for responding to a data searching instruction of a user, carrying out one-dimensional mapping on the mapping data of each expansion feature space in a space filling curve mode to obtain one-dimensional mapping data, and carrying out data searching based on the one-dimensional mapping data to obtain a searching result;
The data filter construction unit is used for acquiring power plant data in a set time range as historical data, and performing data analysis on the historical data to find out energy consumption related data in the historical data, and the method comprises the following steps: the obtained history data is represented by an input matrix M, and each element in the input matrix M is a data tuple Q i The method comprises the steps of carrying out a first treatment on the surface of the Extracting matrix components of the input matrix M to obtain a component matrix C for representing the energy consumption related data, wherein the method comprises the following specific processes: selecting a set of scale parameters sigma 1 ，σ 2 ，...，σ k The method comprises the steps of carrying out a first treatment on the surface of the Then, a gaussian filter is applied to the input matrix M, generating a set of scale-space matrices:
L＝L 1 ，L 2 ，...，L k ；
wherein each scale space matrixExpressed at the scale sigma i A lower fuzzy matrix; the formula for constructing the scale space is as follows:
where (x, y) is the pixel coordinates in the matrix,is a gaussian kernel function for performing a smoothing operation on the input matrix M; for each scale space matrix L i Detecting keypoints and computing feature descriptors using a feature extraction algorithm; assuming that a set is obtainedCharacteristic points:
P＝p 1 ，p 2 ，...，p q ；
wherein each feature point p j ＝(x j ，y j ，σ j ) Including location and scale information; generating a component matrix according to the position and the scale of the characteristic points of the pre-established energy consumption related data Wherein each column represents a component of a feature point; for each feature point p j Each row of the component matrix represents a location, scale, direction, and feature descriptor of the feature point; assuming that the component matrix C has q columns, representing q feature points; then the j-th column of C is represented as a vector v j Wherein:
v j ＝[x j ，y j ，σ j ，θ j ，d j1 ，d j2 ，...，d jn ]；
wherein x is j ，y j ，σ j Respectively represent characteristic points p j And the position and scale, θ j Representing the characteristic point p j Direction d of (d) j1 ，d j2 ，...，d jn Representing the characteristic point p j N components in the feature descriptor of (2); the size of the component matrix C is p multiplied by q, wherein p represents the attribute dimension of the feature points, and q represents the number of the extracted feature points; the position and the scale of the characteristic points of the energy consumption related data are obtained by acquiring the existing energy consumption related data and calculating;
the data search device performs one-dimensional mapping on mapping data of each expansion feature space by using a space filling curve mode to obtain one-dimensional mapping data, and performs data search based on the one-dimensional mapping data to obtain a search result, wherein the method comprises the following steps: mapping data in each extended feature spaceMapping the space filling curve into a one-dimensional space by using a space filling curve mode to obtain one-dimensional mapping data dit; repeating the above steps, mapping data +_for each sample >Obtaining a one-dimensional mapping data set di0, d i1 ，...，d i(t+1) The method comprises the steps of carrying out a first treatment on the surface of the For given query data q, mapping it to obtain one-dimensional query value d q The method comprises the steps of carrying out a first treatment on the surface of the Searching and inquiring the value d in a one-dimensional space by using binary search according to one-dimensional mapping data in the one-dimensional mapping data set q The closest mapping data is used as the search result.
2. The power plant SIS system-based energy consumption data management platform of claim 1, wherein the power plant data acquired by the SIS system comprises: plant data DE, plant operation data OP and plant environment data EN; the power plant equipment data are factory data of the power plant equipment; the power plant equipment operation data are operation data of the power plant equipment during operation; the power plant environment data are environment data of a power plant; when the SIS system stores power plant data, the following data tuples are constructed for storage:
Q i ＝{DE i |{T 1 ，OP 1 ，EN 1 }，{T 2 ，OP 2 ，EN 2 }，...，{T n ，OP n ，EN n }}；
wherein Q is i I is a data tuple, i is a serial number of power plant equipment; t (T) n To obtain the time of the power plant data, n is the number of times of obtaining the power plant data.
3. The power plant SIS system-based energy consumption data management platform according to claim 2, wherein the feature descriptor calculation method comprises: constructing a matrix area around the feature points, and dividing the matrix area into a plurality of sub-areas; then, for each sub-region, calculating a local gradient direction operator of the sub-region, and connecting the local gradient direction operators of the sub-regions to form a feature descriptor according to the frequency of the statistical gradient direction falling in each direction interval.
4. A power plant SIS system based energy consumption data management platform according to claim 3, wherein the data filter construction unit, based on the energy consumption related data, constructs a data filter by a method comprising: each column in the component matrix C is considered as an array, and a hash operator is assigned to each array to construct a data filter.
5. The power plant SIS system-based energy consumption data management platform according to claim 4, wherein the data filter periodically acquires power plant data from the SIS system and performs data filtering on the acquired power plant data, and the method for obtaining filtered data comprises the following steps: when power plant data are periodically acquired from an SIS system, each acquired power plant data are randomly inserted into an array of a data filter to serve as an element, a hash arithmetic unit of each array in the data filter is used for calculating a hash value again, if the normalized mean value of the calculated hash value is equal to the normalized mean value of the hash value before the power plant data are inserted, the power plant data are filtered, otherwise, the power plant data are passed, and when the acquisition of the power plant data from the SIS system is finished in the period, the passed power plant data are reserved to serve as filtered data.
6. The power plant SIS system-based energy consumption data management platform according to claim 5, wherein the data mapping unit constructs a data mapper, and the method of the data mapper performing extended feature space mapping on the filtered data to obtain mapped data under each extended feature space comprises: clustering the filtered data X by using a hierarchical clustering algorithm to obtain K clustering clusters; for each cluster k, its cluster center c is calculated k The method comprises the steps of carrying out a first treatment on the surface of the Initializing spatial mapping vectorsAnd a space mapping covariance matrixP 0 The method comprises the steps of carrying out a first treatment on the surface of the For each data sample xi in the filtered data X: initializing the current spatial map +.>And a current spatial mapping covariance matrix Pi0; for the mapping direction t: transfer function using spatial mapping> And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping:
transferring noise covariance matrix Q using spatial mapping t And a space mapping transfer jacobian matrix Ft, calculating a priori space mapping covariance matrix Pit:
Pit＝FtPit-Ft T +Q t ；
using an observed noise covariance matrix R t And an observation jacobian matrix Ht, calculating an observation prediction
Using an observed noise covariance matrix R t And observing the Jacobian matrix Hit, and calculating the Kalman gain K it ：
Kit＝PitHit T (HitPitHit T +R t ) -1 ；
Using the observation vector zit, the residual yit is calculated:
Updating spatial mapping
Finally, the mapping data under each expansion feature space is obtained as
7. The power plant SIS system based energy consumption data management platform according to claim 6, wherein the data analyzer performs data analysis on the mapping data in each extended feature space to determine whether a data anomaly has occurred, the method comprising: mapping data for each extended feature spaceCalculating the Euclidean distance d between the clustering center ck and the clustering center; comparing the Euclidean distance d with a preset judging value, judging that the data is abnormal if the Euclidean distance d is larger than the judging value, otherwise, judging that the data is not abnormal.
8. The power plant SIS system based energy consumption data management platform of claim 7, wherein the space filling curve is one of a Hilbert curve or a Z curve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310626302.9A CN116383262B (en) | 2023-05-31 | 2023-05-31 | Power plant SIS system-based energy consumption data management platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310626302.9A CN116383262B (en) | 2023-05-31 | 2023-05-31 | Power plant SIS system-based energy consumption data management platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116383262A CN116383262A (en) | 2023-07-04 |
| CN116383262B true CN116383262B (en) | 2023-08-11 |
Family
ID=86971336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310626302.9A Active CN116383262B (en) | 2023-05-31 | 2023-05-31 | Power plant SIS system-based energy consumption data management platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116383262B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678590A (en) * | 2013-12-12 | 2014-03-26 | 用友软件股份有限公司 | Report collecting device and report collecting method based on OLAP |
| KR20160022713A (en) * | 2014-08-20 | 2016-03-02 | 한전케이디엔주식회사 | Power factor management system for power station |
| CN111624931A (en) * | 2020-06-18 | 2020-09-04 | 山东山大世纪科技有限公司 | Industrial park electricity utilization internet intelligent operation and maintenance management and control system and method |
| CN112528448A (en) * | 2021-01-01 | 2021-03-19 | 谭世克 | Topology and data management maintenance system |
| CN113986892A (en) * | 2021-09-14 | 2022-01-28 | 国电南京自动化股份有限公司 | Coal consumption prediction and data cleaning method and system for thermal power generating unit |
| CN114065967A (en) * | 2021-11-15 | 2022-02-18 | 西安热工研究院有限公司 | Monitoring and diagnosing system and method for whole life cycle of power plant SIS system data |
| CN114154842A (en) * | 2021-11-30 | 2022-03-08 | 华能湖南岳阳发电有限责任公司 | Operation guidance system and method based on real-time data analysis of power plant |
-
2023
- 2023-05-31 CN CN202310626302.9A patent/CN116383262B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678590A (en) * | 2013-12-12 | 2014-03-26 | 用友软件股份有限公司 | Report collecting device and report collecting method based on OLAP |
| KR20160022713A (en) * | 2014-08-20 | 2016-03-02 | 한전케이디엔주식회사 | Power factor management system for power station |
| CN111624931A (en) * | 2020-06-18 | 2020-09-04 | 山东山大世纪科技有限公司 | Industrial park electricity utilization internet intelligent operation and maintenance management and control system and method |
| CN112528448A (en) * | 2021-01-01 | 2021-03-19 | 谭世克 | Topology and data management maintenance system |
| CN113986892A (en) * | 2021-09-14 | 2022-01-28 | 国电南京自动化股份有限公司 | Coal consumption prediction and data cleaning method and system for thermal power generating unit |
| CN114065967A (en) * | 2021-11-15 | 2022-02-18 | 西安热工研究院有限公司 | Monitoring and diagnosing system and method for whole life cycle of power plant SIS system data |
| CN114154842A (en) * | 2021-11-30 | 2022-03-08 | 华能湖南岳阳发电有限责任公司 | Operation guidance system and method based on real-time data analysis of power plant |
Non-Patent Citations (1)
| Title |
|---|
| 基于火电厂热力设备特征的大数据应用技术研究;吕晓磊;《CNKI优秀硕士学位论文全文库》;C042-329 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116383262A (en) | 2023-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110990461A (en) | Big data analysis model algorithm model selection method and device, electronic equipment and medium | |
| CN110021425A (en) | A kind of relatively detector and its construction method and cervical cancer cell detection method | |
| CN107316053A (en) | A kind of cloth image Rapid matching search method | |
| CN108416373A (en) | A kind of unbalanced data categorizing system based on regularization Fisher threshold value selection strategies | |
| CN113807418A (en) | Injection molding machine energy consumption abnormity detection method and system based on Gaussian mixture model | |
| CN111626360A (en) | Method, device, equipment and storage medium for detecting fault type of boiler | |
| CN114700587A (en) | Missing welding defect real-time detection method and system based on fuzzy reasoning and edge calculation | |
| Liu et al. | Improved image matching algorithm based on LK optical flow and grid motion statistics | |
| Sun et al. | Flow measurement-based self-adaptive line segment clustering model for leakage detection in water distribution networks | |
| CN116383262B (en) | Power plant SIS system-based energy consumption data management platform | |
| Zhou et al. | Analyzing nitrogen effects on rice panicle development by panicle detection and time-series tracking | |
| Ni et al. | Multi-meter intelligent detection and recognition method under complex background | |
| CN108399413B (en) | Picture shooting area identification and geographical positioning method and device | |
| CN115378000A (en) | Power distribution network operation state evaluation method based on interval two-type fuzzy clustering analysis | |
| CN112801028A (en) | Spectrum and space map hyperspectral image classification method based on inductive representation learning network | |
| CN109189775B (en) | Industrial monitoring platform mass data processing system and method | |
| Yu et al. | Abnormal crowdsourced data detection using remote sensing image features | |
| Kundur et al. | Pest Detection and Recognition: An approach using Deep Learning Techniques | |
| CN113705920B (en) | Method for generating water data sample set for thermal power plant and terminal equipment | |
| CN110852355B (en) | KNN-based label data classification method | |
| Li | Practice of machine learning algorithm in data mining field | |
| CN113723835B (en) | Water consumption evaluation method and terminal equipment for thermal power plant | |
| Cao et al. | Improved YOLOv3 model based on ResNeXt for target detection | |
| Gao et al. | An input module of deep learning for the analysis of time series with unequal length | |
| CN117541832B (en) | Abnormality detection method, abnormality detection system, electronic device, and storage medium |
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 |