CN113837539A - Coal fired boiler heating surface depth fault early warning system based on industrial internet - Google Patents
Coal fired boiler heating surface depth fault early warning system based on industrial internet Download PDFInfo
- Publication number
- CN113837539A CN113837539A CN202110953769.5A CN202110953769A CN113837539A CN 113837539 A CN113837539 A CN 113837539A CN 202110953769 A CN202110953769 A CN 202110953769A CN 113837539 A CN113837539 A CN 113837539A
- Authority
- CN
- China
- Prior art keywords
- data
- intelligent
- model
- calculation
- intelligent analysis
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
本发明公开了一种基于工业互联网的燃煤锅炉受热面深度故障预警系统，涉及工业互联网技术领域，本发明包括数据采集、数据转换与传输、大数据融合存储、数据管理、数据对象、数据计算、智能分析和智能化应用；本发明通过智能算法建立炉管温度与锅炉工况的关联关系数据模型以及炉管间温度与位置的数据模型，达到对炉管的温度进行实时分析和历史变化分析，给出问题原因及提前预警，运行中可及时调整运行方式，改善或延缓故障发展趋势，通过对相邻管屏壁温数据的比较分析，对壁温历史数据的变化分析，在机组调停、检修中可针对潜在风险进行预防性消缺，大大有利于减少锅炉四管泄漏风险，提高机组安全连续运行能力。
The invention discloses a coal-fired boiler heating surface deep fault early warning system based on the industrial Internet, and relates to the technical field of the industrial Internet. The invention includes data collection, data conversion and transmission, big data fusion storage, data management, data objects, and data calculation. , intelligent analysis and intelligent application; the invention establishes a data model of the correlation relationship between furnace tube temperature and boiler working conditions and a data model of temperature and position between furnace tubes through intelligent algorithms, so as to achieve real-time analysis and historical change analysis of furnace tube temperature , the cause of the problem and early warning are given, the operation mode can be adjusted in time during operation, and the development trend of the failure can be improved or delayed. During the maintenance, the potential risks can be eliminated preventively, which is greatly beneficial to reduce the leakage risk of the four pipes of the boiler and improve the safe and continuous operation capability of the unit.
Description
Technical Field
The invention relates to the technical field of industrial internet, in particular to a coal-fired boiler heating surface depth fault early warning system based on the industrial internet.
Background
The industrial internet is a novel infrastructure, an application mode and an industrial ecology deeply integrated by a new generation of information communication technology and industrial economy, and a brand new manufacturing and service system covering a whole industrial chain and a whole value chain is constructed by comprehensively connecting people, machines, objects, systems and the like, so that a realization approach is provided for the digitalization, networking and intelligent development of industry and even industry, and the industrial internet is an important basic stone of the fourth industrial revolution. Industrial internet is not a simple application of the internet in industry, but has a more rich connotation and extension. The system takes a network as a basis, a platform as a center, data as an element and safety as a guarantee, is not only an infrastructure for industrial digitization, networking and intelligent transformation, but also an application mode for deep integration of internet, big data, artificial intelligence and entity economy, and is also a new state and a new industry, and the form, the supply chain and the industry chain of an enterprise are reshaped. Currently, the industrial internet fusion application is widely expanded to national economic key industries, six new modes of platform design, intelligent manufacturing, networking collaboration, personalized customization, service extension and digital management are formed, the effects of enabling, intelligence and assignment are continuously shown, and the economic quality improvement, efficiency improvement, cost reduction, green and safe development of entities are powerfully promoted.
In a thermal power plant, the furnace tube is in a severe working environment with high temperature and high pressure for a long time, and the deterioration problem of the furnace tube is very serious; the deterioration of the furnace tube greatly shortens the service life of the furnace tube, and the furnace tube has potential risk of tube explosion leakage, which is one of main threats of safe operation of the thermal power boiler.
Disclosure of Invention
The invention aims to solve the defects in the prior art, and provides a coal-fired boiler heating surface depth fault early warning system based on an industrial internet.
In order to achieve the purpose, the invention adopts the following technical scheme:
a coal-fired boiler heating surface depth fault early warning system based on an industrial internet comprises data acquisition, data conversion and transmission, big data fusion storage, data management, data objects, data calculation, intelligent analysis and intelligent application;
the data acquisition acquires various data of the power plant equipment through a DCS/SIS/KDM interface;
the data conversion and transmission module transmits various equipment data acquired by the data acquisition to a data storage module, a service module, a display module and the like of a monitoring platform;
the big data fusion storage stores structured and unstructured data;
the data management classifies different data into different databases for effective management according to the service of monitoring data of different equipment;
the data object records and displays process parameters, equipment states, working condition information and knowledge items;
the data calculation is carried out on the equipment diagnosis health prediction through a big data engine, and the data are calculated and combed through different calculation modules according to the intelligent analysis requirements of the business;
the intelligent analysis uses various artificial intelligence and machine learning algorithms to carry out intelligent analysis on services such as intelligent monitoring platform equipment diagnosis and the like;
the intelligent application main functions comprise measuring point abnormity detection, overrun statistics, operation assessment and degradation trend analysis.
Preferably, the data acquisition is mainly responsible for acquiring various types of data of equipment from a KDM system.
Preferably, the data conversion and transmission is performed by filtering and cleaning the device state information and the working condition information through a TCP/IP protocol, and then performing high-speed transmission on the network.
Preferably, the data management stores data such as real-time data, model analysis results, historical hotspot data, models and knowledge, and the data management comprises a real-time database and a historical database;
the real-time database stores data of each measuring point transmitted in real time; and the historical database stores historical data of each measuring point.
Preferably, the intelligent analysis adopts different intelligent analysis algorithms to provide intelligent analysis capability for fault monitoring, diagnosis, prediction and the like, the algorithm of the intelligent analysis can be used by other programs after being initialized, and the initialization stage is divided into four steps: data preprocessing, feature extraction, model training and algorithm deployment.
Preferably, the historical database stores historical data in the historical database by adopting hot spot data caching and pre-reading technologies according to the current data storage situation and by combining the characteristics of equipment data, and only keeps the historical data of the recent period of time and other historical hot spot data;
the hot spot data caching stores the hot spot data into a cache, stores the hot spot data into a Redis distributed database, a calculation or application server obtains the hot spot data from the hot spot database, and directly obtains the data through a KDM data center data sharing API for the data which is not in the database;
the pre-reading technology is used for pre-judging data required by a computer program or an application in the future, and reading and caching the data in advance.
Preferably, the data preprocessing is to meet the requirements of the algorithm of intelligent analysis by processing the data into the required format and distribution type;
the feature extraction is to process the preprocessed data, extract the data needed by intelligent analysis from the preprocessed data, and simultaneously smooth the data;
the algorithm training generates a data database query condition according to a data driving model to be trained, a database server is queried so as to read equipment state data stored in a data warehouse, data calculation tasks such as data conversion and feature extraction are generated, the result of the data calculation task calculation is used as training data of the data driving model and input into an intelligent calculation server, and the intelligent calculation server calls various intelligent algorithms to perform data driving modeling on the equipment state;
the algorithm deployment obtains a trained data driving model stored in a database from the database, the intelligent analysis server loads the data driving model and performs necessary initialization work on the model, and the intelligent analysis server calls an interface provided by a WebService server and externally releases functions provided by the data driving model into a restful WebService form.
Preferably, unsupervised learning is carried out by the intelligent analysis and the data object, a boiler wall temperature intelligent analysis model is established, new data are classified by using the model after the new data arrive, and the abnormal fluctuation of the furnace tube temperature measuring points in time and the change of temperature correlation with the surrounding furnace tube measuring points are captured by the intelligent application depth, so that a deteriorated furnace tube evaluation standard model is established.
Preferably, the algorithm of the boiler wall temperature intelligent analysis model is as follows:
the calculation formula of the metal temperature of the pipe wall is as follows:
the calculation formula of the outer wall temperature is as follows:
in the formula: t is tqCalculating the temperature in the point pipe; beta is the ratio of the outer diameter to the inner diameter of the tube; mu is a heat dissipation coefficient; delta is the thickness of the tube wall; q is a meterCalculating the point at which the tube absorbs the thermal load; lambda is the heat conductivity coefficient of the metal of the tube wall; alpha is alpha2Heat release coefficient for the vapor side;
the calculation formula of the heat absorption load of the pipe is as follows:
in the formula: theta is the calculated point flue gas temperature; alpha is alpha1Calculating the convection heat release coefficient of the smoke side of the point; alpha is alpha3Calculating the radiation heat release coefficient of the ignition smoke; ε is the contamination coefficient of the tube at the point of calculation.
Preferably, the flow of the deterioration furnace tube evaluation standard model is as follows: reading a measuring point information file, acquiring a measuring point list needing to be calculated and detected, and grouping the measuring points; extracting historical data from kdm for each group of measuring points, and extracting data of the current time period; and asynchronously calling the main function of the algorithm program.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the invention, an incidence relation data model of the furnace tube temperature and the boiler working condition and a data model of the temperature and the position between the furnace tubes are established through an intelligent algorithm, so that the temperature of the furnace tubes is analyzed in real time and historical change is analyzed, the cause of the problem and early warning are given, the operation mode can be adjusted in time during operation, the fault development trend is improved or delayed, the wall temperature historical data is analyzed through comparison and analysis of the wall temperature data of adjacent tube panels, a data model for evaluating the safety of the furnace tube temperature and the heating surface operation working condition is established, the safety of the tube walls is evaluated in sequence, the potential risks can be eliminated in the set dispatching and maintenance, the leakage risks of four tubes of the boiler are greatly reduced, and the safe and continuous operation capability of the set is improved.
2. According to the invention, a time series machine learning-based furnace tube degradation targeted algorithm model is developed through comprehensive analysis of working conditions and furnace tube temperatures in a power plant generator set; the working state of the unit can be identified, the condition of false alarm of a furnace tube degradation model caused by data fluctuation of start and stop of the unit is eliminated, a big data mining technology is applied, abnormal fluctuation of furnace tube temperature data in time and change of correlation with surrounding furnace tube temperature are deeply captured, a degradation furnace tube evaluation standard model is established, and full life cycle state monitoring of the furnace tube heating state is realized; the model is automatically updated on line along with the operation, and is free from maintenance; the workload of starting the machine and the pressure of safety accidents are reduced.
Drawings
FIG. 1 is a schematic diagram of a system overall framework of a coal-fired boiler heating surface depth fault early warning system based on an industrial internet, which is provided by the invention;
FIG. 2 is a diagram of a historical data storage architecture of a coal-fired boiler heating surface depth fault early warning system based on the industrial Internet according to the present invention;
FIG. 3 is a working flow chart of a deteriorated furnace tube evaluation standard model of a coal-fired boiler heating surface depth fault early warning system based on the industrial internet.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Referring to fig. 1-2, a coal-fired boiler heating surface deep fault early warning system based on an industrial internet comprises data acquisition, data conversion and transmission, big data fusion storage, data management, data objects, data calculation, intelligent analysis and intelligent application;
the data acquisition acquires various data of the power plant equipment through a DCS/SIS/KDM interface;
the data conversion and transmission module transmits various equipment data acquired by data acquisition to a data storage module, a service module, a display module and the like of the monitoring platform;
the big data fusion storage stores structured data and unstructured data;
the data management classifies different data into different databases for effective management according to the service of monitoring data of different equipment;
the data object records and displays the process parameters, the equipment state, the working condition information and the knowledge items;
the data calculation is carried out on the equipment diagnosis health prediction through a big data engine, and the data are calculated and combed through different calculation modules according to the intelligent analysis requirements of the business;
the intelligent analysis uses various artificial intelligence and machine learning algorithms to carry out intelligent analysis on services such as intelligent monitoring platform equipment diagnosis and the like;
the intelligent application main functions comprise measuring point abnormity detection, overrun statistics, operation assessment and deterioration trend analysis.
The method comprises the steps of acquiring various data of all equipment of the power plant in real time through data acquisition, completely reserving historical data of the whole life cycle of the equipment, sending the data for other services in a data sharing mode, and mainly taking charge of acquiring various data of the equipment from a KDM system through the data acquisition.
The data conversion and transmission processes pretreatment such as filtering and cleaning of equipment state information and working condition information through TCP/IP protocol, and then high speed transmission is carried out on the network.
The data management system comprises a data management module, a data processing module and a data processing module, wherein the data management module is used for storing data such as real-time data, model analysis results, historical hotspot data, models and knowledge;
the real-time database stores the data of each measuring point transmitted in real time; and the historical database stores historical data of each measuring point.
The intelligent analysis adopts different intelligent analysis algorithms to provide intelligent analysis capability for fault monitoring, diagnosis, prediction and the like, the algorithm of the intelligent analysis can be used by other programs after being initialized, and the initialization stage is divided into four steps: data preprocessing, feature extraction, model training and algorithm deployment.
The historical database stores historical data in a Redis historical database by adopting hot data caching and pre-reading technologies according to the current data storage situation and by combining the data characteristics of equipment, and only keeps the historical data and other historical hot data in the recent period of time;
the hot spot data caching stores the hot spot data into a cache, stores the hot spot data into a Redis distributed database, a calculation or application server obtains the hot spot data from the hot spot database, and directly obtains the data through a KDM data center data sharing API for the data which is not in the database;
the pre-reading technology is used for pre-judging data required by a computer program or an application in the future, and reading and caching the data in advance.
The data preprocessing is to process the data into a required format and a required distribution type so as to meet the requirement of an intelligent analysis algorithm;
the feature extraction is to process the preprocessed data, extract the data needed by intelligent analysis from the preprocessed data, and simultaneously smooth the data;
the algorithm training generates a data database query condition according to a data driving model to be trained, a database server is queried so as to read equipment state data stored in a data warehouse, data calculation tasks such as data conversion and feature extraction are generated, the result of the data calculation task calculation is used as training data of the data driving model and input into an intelligent calculation server, and the intelligent calculation server calls various intelligent algorithms to perform data driving modeling on the equipment state;
the algorithm deployment obtains a trained data driving model stored in the database from the database, the intelligent analysis server loads the data driving model and performs necessary initialization work on the model, and the intelligent analysis server calls an interface provided by the WebService server and externally releases functions provided by the data driving model into a restful WebService form.
The method comprises the steps of carrying out unsupervised learning by intelligent analysis and combining with data objects, establishing an intelligent analysis model of the wall temperature of the boiler, classifying new data by using the model after the new data arrives, capturing abnormal fluctuation of furnace tube temperature measuring points in time and changes of temperature correlation with surrounding furnace tube measuring points by intelligent application depth, and establishing a deteriorated furnace tube evaluation standard model.
The algorithm of the boiler wall temperature intelligent analysis model is as follows:
the calculation formula of the metal temperature of the pipe wall is as follows:
the calculation formula of the outer wall temperature is as follows:
in the formula: t is tqCalculating the temperature in the point pipe; beta is the ratio of the outer diameter to the inner diameter of the tube; mu is a heat dissipation coefficient; delta is the thickness of the tube wall; q is the calculated point heat load absorbed by the pipe; lambda is the heat conductivity coefficient of the metal of the tube wall; alpha is alpha2Heat release coefficient for the vapor side;
the calculation formula of the heat absorption load of the pipe is as follows:
in the formula: theta is the calculated point flue gas temperature; alpha is alpha1Calculating the convection heat release coefficient of the smoke side of the point; alpha is alpha3Calculating the radiation heat release coefficient of the ignition smoke; ε is the contamination coefficient of the tube at the point of calculation.
Wherein, the flow of the deteriorated furnace tube evaluation standard model is as follows: reading a measuring point information file, acquiring a measuring point list needing to be calculated and detected, and grouping the measuring points; extracting historical data from kdm for each group of measuring points, and extracting data of the current time period; and asynchronously calling the main function of the algorithm program.
Referring to fig. 3, when the deteriorated furnace tube evaluation standard model works, firstly reading a measuring point information file, acquiring a measuring point list needing to be calculated and detected, and grouping measuring points; extracting historical data from kdm for each group of measuring points, and extracting data of the current time period; asynchronously calling a main function of an algorithm program;
then judging the working state of the unit and the abnormal state of the measuring point, if the working state and the abnormal state of the measuring point exist, returning to False, and jumping out of the algorithm program; loading a prediction model corresponding to the measuring point; aligning temperature measuring point data and unit load data; using a ridge regression model, calculating the station degradation state:
1) processing the load data by using PolynomialFeatures to obtain a characteristic item;
2) obtaining a measuring point temperature predicted value based on a ridge regression prediction model;
3) judging whether the measuring points are degraded or not by the algorithm model;
if the model judges that the measuring point is degraded, updating the model information file pkl, and putting the current window time period into the model information file pkl; updating a current measuring point prediction model; generating alarm information according to the format requirement of a database, and returning the result to an interface program;
if the model judges that the measuring point is updated to be in a normal state (maintenance occurs) from the degradation state, updating the model information file pkl, and putting the current window time period into the model information file pkl; updating a current measuring point prediction model;
if the model judges that the measuring point is not degraded, returning to False;
in the interface program, if furnace tube degradation alarm information exists in a result returned by the degradation algorithm program, the formatted alarm information is pushed to kafka;
acquiring measurement point data of the reference time and data of a current window based on the reference model time and the current window time in the furnace tube degradation alarm information, and realizing temperature and load data alignment in the function (according to the front end requirement);
and pushing the measuring point data acquired in the previous step to a background.
After completing all computing tasks, the interface program waits for timing tasks: the main function is re-executed periodically.
The working process of the intelligent analysis model for the wall temperature of the boiler is as follows:
1) calibrating the working condition of each time through the temperature change of the key measuring point;
2) analyzing distribution conditions (highest point, lowest point, percentile and the like) of the temperature of each measuring point under different historical working conditions by combining with working condition information, setting a threshold value of an absolute value of the temperature according to the distribution conditions, and alarming when new data exceeds the threshold value;
3) analyzing the distribution (standard deviation, range difference and the like) of all tube temperatures of the same screen at the same time under different historical working conditions by combining with working condition information, setting a threshold value according to the distribution, and alarming when new data exceeds the threshold value;
4) analyzing the distribution conditions (highest point, lowest point, quantile and the like) of the change rate of the temperature of each screen measuring point under different historical working conditions by combining with the working condition information, setting a threshold value of the temperature change rate according to the distribution conditions, and alarming when new data exceed the threshold value;
5) aiming at the temperature of a certain measuring point at a certain moment, modeling the temperature of other pipe measuring points on the same screen at the same time, obtaining a predicted value through a model after new data comes, comparing the predicted value with a real value, and alarming if the difference is large;
6) aiming at the temperature of a certain measuring point at a certain moment, modeling the temperature of the certain measuring point at a plurality of moments before the certain measuring point, obtaining a predicted value through a model after new data comes, comparing the predicted value with a real value, and alarming if the difference is large;
7) combining working condition information, using the temperatures of all measuring points in a certain screen at the same time, performing unsupervised learning and modeling, and classifying (normal/abnormal) new data by using the model after the new data comes;
8) for the temperature of a certain measuring point, a section of data is intercepted by using a sliding window, self-encoding-decoding learning is carried out by using an AutoEncoder technology in deep learning, after new data arrives, the same self-encoding-decoding is carried out, and if the difference between the obtained data and the original data is larger, an alarm is given;
9) according to the fault record, the data are labeled, then a classification model is trained by using supervision algorithms such as random forests, SVM, xgboost and the like, and the new data are classified by using the model after arriving.
In the description of the present invention, it is to be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate orientations and positional relationships based on those shown in the drawings, and are used only for convenience of description and simplicity of description, and do not indicate or imply that the equipment or element being referred to must have a particular orientation, be constructed and operated in a particular orientation, and thus, should not be considered as limiting the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (10)
1. A coal-fired boiler heating surface depth fault early warning system based on an industrial internet is characterized by comprising data acquisition, data conversion and transmission, big data fusion storage, data management, data objects, data calculation, intelligent analysis and intelligent application;
the data acquisition acquires various data of the power plant equipment through a DCS/SIS/KDM interface;
the data conversion and transmission module transmits various equipment data acquired by the data acquisition to a data storage module, a service module, a display module and the like of a monitoring platform;
the big data fusion storage stores structured and unstructured data;
the data management classifies different data into different databases for effective management according to the service of monitoring data of different equipment;
the data object records and displays process parameters, equipment states, working condition information and knowledge items;
the data calculation is carried out on the equipment diagnosis health prediction through a big data engine, and the data are calculated and combed through different calculation modules according to the intelligent analysis requirements of the business;
the intelligent analysis uses various artificial intelligence and machine learning algorithms to carry out intelligent analysis on services such as intelligent monitoring platform equipment diagnosis and the like;
the intelligent application main functions comprise measuring point abnormity detection, overrun statistics, operation assessment and degradation trend analysis.
2. The industrial internet-based deep fault early warning system for the heating surface of the coal-fired boiler according to claim 1, wherein the data acquisition is mainly responsible for acquiring various types of data of equipment from a KDM system, various types of data of all equipment of a power plant are acquired in real time, historical data of the whole life cycle of the equipment are completely reserved, and the data are transmitted for other services in a data sharing mode.
3. The system of claim 1, wherein the data conversion and transmission is performed by filtering and cleaning the device status information and the operating condition information via TCP/IP protocol, and then performing high-speed transmission via the network.
4. The deep fault early warning system for the heating surface of the coal-fired boiler based on the industrial internet as claimed in claim 1, wherein the data management stores data such as real-time data, model analysis results, historical hot data, models and knowledge, and comprises a real-time database and a historical database;
the real-time database stores data of each measuring point transmitted in real time; and the historical database stores historical data of each measuring point.
5. The industrial internet-based deep fault early warning system for the heating surface of the coal-fired boiler, according to claim 1, is characterized in that intelligent analysis adopts different intelligent analysis algorithms to provide intelligent analysis capability for fault monitoring, diagnosis, prediction and the like, the algorithm of the intelligent analysis can be used by other programs after being initialized, and the initialization stage is divided into four steps: data preprocessing, feature extraction, model training and algorithm deployment.
6. The coal-fired boiler heating surface depth fault early warning system based on the industrial internet as claimed in claim 4, characterized in that the historical database stores historical data in Redis the historical database only retaining the historical data of the latest period of time and other historical hot data by adopting hot data caching and pre-reading technology according to the current data storage situation and combining the characteristics of equipment data;
the hot spot data caching stores the hot spot data into a cache, stores the hot spot data into a Redis distributed database, a calculation or application server obtains the hot spot data from the hot spot database, and directly obtains the data through a KDM data center data sharing API for the data which is not in the database;
the pre-reading technology is used for pre-judging data required by a computer program or an application in the future, and reading and caching the data in advance.
7. The industrial internet-based deep fault early warning system for the heating surface of the coal-fired boiler according to claim 5, characterized in that the data preprocessing is to process the data into a required format and distribution type to meet the requirements of an intelligent analysis algorithm;
the feature extraction is to process the preprocessed data, extract the data needed by intelligent analysis from the preprocessed data, and simultaneously smooth the data;
the algorithm training generates a data database query condition according to a data driving model to be trained, a database server is queried so as to read equipment state data stored in a data warehouse, data calculation tasks such as data conversion and feature extraction are generated, the result of the data calculation task calculation is used as training data of the data driving model and input into an intelligent calculation server, and the intelligent calculation server calls various intelligent algorithms to perform data driving modeling on the equipment state;
the algorithm deployment obtains a trained data driving model stored in a database from the database, the intelligent analysis server loads the data driving model and performs necessary initialization work on the model, and the intelligent analysis server calls an interface provided by a WebService server and externally releases functions provided by the data driving model into a restful WebService form.
8. The system of claim 1, wherein an intelligent analysis model of boiler wall temperature is established by performing unsupervised learning through the intelligent analysis and combining with the data objects, new data are classified by using the model after arriving, and a deteriorated furnace tube evaluation standard model is established by capturing abnormal fluctuation of furnace tube temperature measuring points in time and changes of temperature correlation with surrounding furnace tube measuring points through the intelligent application depth.
9. The industrial internet-based coal-fired boiler heating surface depth fault early warning system according to claim 8, wherein the algorithm of the boiler wall temperature intelligent analysis model is as follows:
the calculation formula of the metal temperature of the pipe wall is as follows:
the calculation formula of the outer wall temperature is as follows:
in the formula: t is tqCalculating the temperature in the point pipe; beta is the ratio of the outer diameter to the inner diameter of the tube; mu is a heat dissipation coefficient; delta is the thickness of the tube wall; q is the calculated point heat load absorbed by the pipe; lambda is the heat conductivity coefficient of the metal of the tube wall; alpha is alpha2Heat release coefficient for the vapor side;
the calculation formula of the heat absorption load of the pipe is as follows:
in the formula: theta is the calculated point flue gas temperature; alpha is alpha1Calculating the convection heat release coefficient of the smoke side of the point; alpha is alpha3Calculating the radiation heat release coefficient of the ignition smoke; ε is the contamination coefficient of the tube at the point of calculation.
10. The industrial internet-based deep fault early warning system for the heating surface of the coal-fired boiler according to claim 1, wherein the evaluation standard model process of the deteriorated furnace tube is as follows: reading a measuring point information file, acquiring a measuring point list needing to be calculated and detected, and grouping the measuring points; extracting historical data from kdm for each group of measuring points, and extracting data of the current time period; and asynchronously calling the main function of the algorithm program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110953769.5A CN113837539A (en) | 2021-08-19 | 2021-08-19 | Coal fired boiler heating surface depth fault early warning system based on industrial internet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110953769.5A CN113837539A (en) | 2021-08-19 | 2021-08-19 | Coal fired boiler heating surface depth fault early warning system based on industrial internet |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113837539A true CN113837539A (en) | 2021-12-24 |
Family
ID=78960822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110953769.5A Pending CN113837539A (en) | 2021-08-19 | 2021-08-19 | Coal fired boiler heating surface depth fault early warning system based on industrial internet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113837539A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115033830A (en) * | 2022-06-01 | 2022-09-09 | 华北电力科学研究院有限责任公司 | Wall temperature evaluation method and device |
| CN116186625A (en) * | 2023-02-22 | 2023-05-30 | 太原理工大学 | A method for early warning of boiler heating surface overtemperature |
| CN116821104A (en) * | 2022-08-18 | 2023-09-29 | 南通泽烁信息科技有限公司 | Industrial Internet data processing method and system based on big data |
| CN118211725A (en) * | 2024-04-08 | 2024-06-18 | 江苏省特种设备安全监督检验研究院 | Industrial equipment risk analysis system and method based on data reconstruction |
| CN119202693A (en) * | 2024-07-24 | 2024-12-27 | 成都贝斯特数码科技有限责任公司 | A method and system for diagnosing and predicting heating surface faults based on boiler flow field analysis |
| CN119920363A (en) * | 2025-03-27 | 2025-05-02 | 北京北科欧远科技有限公司 | Soft measurement method, system, medium and program product for flue gas flow of coal-fired boiler |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832543A (en) * | 2010-05-14 | 2010-09-15 | 上海发电设备成套设计研究院 | Method for intelligently managing wall temperature of final superheater and final reheater of boiler of power station |
| CN104766138A (en) * | 2015-03-27 | 2015-07-08 | 大唐淮南洛河发电厂 | Thermal power plant equipment property evaluation and early warning method and system based on industrial internet |
| CN108388149A (en) * | 2018-03-30 | 2018-08-10 | 福建省特种设备检验研究院 | A kind of Industrial Boiler analog simulation and remote supervision system |
| CN110750872A (en) * | 2019-09-24 | 2020-02-04 | 若普自动化技术（北京）有限公司 | Boiler metal wall temperature data analysis system and method |
| CN111259562A (en) * | 2020-02-10 | 2020-06-09 | 南京依盛信息技术有限公司 | Dynamic virtual system for analyzing and controlling failure area of heating surface of boiler |
| EP3663671A1 (en) * | 2018-12-04 | 2020-06-10 | bCheck NV | Boiler performance monitoring and breakdown prediction system |
| US20200231466A1 (en) * | 2017-10-09 | 2020-07-23 | Zijun Xia | Intelligent systems and methods for process and asset health diagnosis, anomoly detection and control in wastewater treatment plants or drinking water plants |
| CN112859822A (en) * | 2021-04-25 | 2021-05-28 | 北京邮电大学 | Equipment health analysis and fault diagnosis method and system based on artificial intelligence |
| KR102283884B1 (en) * | 2020-08-21 | 2021-08-03 | 주식회사 데카엔지니어링 | Artificial intelligence based fire and power anomaly prediction system and method |
-
2021
- 2021-08-19 CN CN202110953769.5A patent/CN113837539A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101832543A (en) * | 2010-05-14 | 2010-09-15 | 上海发电设备成套设计研究院 | Method for intelligently managing wall temperature of final superheater and final reheater of boiler of power station |
| CN104766138A (en) * | 2015-03-27 | 2015-07-08 | 大唐淮南洛河发电厂 | Thermal power plant equipment property evaluation and early warning method and system based on industrial internet |
| US20200231466A1 (en) * | 2017-10-09 | 2020-07-23 | Zijun Xia | Intelligent systems and methods for process and asset health diagnosis, anomoly detection and control in wastewater treatment plants or drinking water plants |
| CN108388149A (en) * | 2018-03-30 | 2018-08-10 | 福建省特种设备检验研究院 | A kind of Industrial Boiler analog simulation and remote supervision system |
| EP3663671A1 (en) * | 2018-12-04 | 2020-06-10 | bCheck NV | Boiler performance monitoring and breakdown prediction system |
| CN110750872A (en) * | 2019-09-24 | 2020-02-04 | 若普自动化技术（北京）有限公司 | Boiler metal wall temperature data analysis system and method |
| CN111259562A (en) * | 2020-02-10 | 2020-06-09 | 南京依盛信息技术有限公司 | Dynamic virtual system for analyzing and controlling failure area of heating surface of boiler |
| KR102283884B1 (en) * | 2020-08-21 | 2021-08-03 | 주식회사 데카엔지니어링 | Artificial intelligence based fire and power anomaly prediction system and method |
| CN112859822A (en) * | 2021-04-25 | 2021-05-28 | 北京邮电大学 | Equipment health analysis and fault diagnosis method and system based on artificial intelligence |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115033830A (en) * | 2022-06-01 | 2022-09-09 | 华北电力科学研究院有限责任公司 | Wall temperature evaluation method and device |
| CN115033830B (en) * | 2022-06-01 | 2024-09-20 | 华北电力科学研究院有限责任公司 | A wall temperature evaluation method and device |
| CN116821104A (en) * | 2022-08-18 | 2023-09-29 | 南通泽烁信息科技有限公司 | Industrial Internet data processing method and system based on big data |
| CN116186625A (en) * | 2023-02-22 | 2023-05-30 | 太原理工大学 | A method for early warning of boiler heating surface overtemperature |
| CN118211725A (en) * | 2024-04-08 | 2024-06-18 | 江苏省特种设备安全监督检验研究院 | Industrial equipment risk analysis system and method based on data reconstruction |
| CN119202693A (en) * | 2024-07-24 | 2024-12-27 | 成都贝斯特数码科技有限责任公司 | A method and system for diagnosing and predicting heating surface faults based on boiler flow field analysis |
| CN119920363A (en) * | 2025-03-27 | 2025-05-02 | 北京北科欧远科技有限公司 | Soft measurement method, system, medium and program product for flue gas flow of coal-fired boiler |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113837539A (en) | Coal fired boiler heating surface depth fault early warning system based on industrial internet | |
| CN113283344A (en) | Mining conveying belt deviation detection method based on semantic segmentation network | |
| WO2021070201A2 (en) | Method and system for realtime monitoring and forecasting of fouling of air preheater equipment | |
| CN118797293B (en) | Container stacking machine real-time fault monitoring and early warning system | |
| CN107631754A (en) | Slope monitoring method and system based on big data platform | |
| CN119179297B (en) | Industrial big data processing method and system based on Internet of things | |
| CN111666652A (en) | Steam heating network steam trap inspection emission operation scheduling method and operation scheduling system | |
| CN108540310A (en) | A kind of behavior prediction method based on wireless network collaborative perception | |
| CN110852484A (en) | A system and method for early warning of wind turbine failure | |
| CN117366661A (en) | Heating system fault data analysis method, system, equipment and storage medium | |
| CN112988843B (en) | SMT chip mounter fault management and diagnosis system based on SQL Server database | |
| CN107301506A (en) | A kind of initial failure risk analysis method for considering mass deviation and failure correlation | |
| CN108508860A (en) | A kind of process industry production system data monitoring method based on coupled relation | |
| CN217423214U (en) | Safety diagnosis system for heating surface of boiler system | |
| CN120101172B (en) | Boiler combustion automatic detection control system based on big data model | |
| CN119858287B (en) | A method and system for monitoring the operation status of injection molding equipment | |
| CN120626320A (en) | Method and system for monitoring emissions of an automobile internal combustion engine | |
| CN116702950B (en) | Prediction method for enterprise illegal production behavior in coal industry based on electric power big data | |
| CN117893014B (en) | Safety risk assessment method, system, medium and equipment for long-term service power station boilers | |
| CN118644942A (en) | A method for identifying smoke in ship engine rooms based on deep learning | |
| CN106643885A (en) | Building safety monitoring system and monitoring method | |
| CN117454177A (en) | Abnormal biomass power generation monitoring system and method thereof | |
| CN114440205A (en) | A safety diagnosis system and method for the heating surface of a boiler system | |
| CN120744403B (en) | A method and system for intelligent management of equipment operation and maintenance | |
| CN114446027A (en) | A device fault alarm method, system, device and medium based on the Internet of Things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211224 |
|
| RJ01 | Rejection of invention patent application after publication |