CN104634706A - Neural network-based soft measurement method for pulverized coal fineness - Google Patents
Neural network-based soft measurement method for pulverized coal fineness Download PDFInfo
- Publication number
- CN104634706A CN104634706A CN201510033857.8A CN201510033857A CN104634706A CN 104634706 A CN104634706 A CN 104634706A CN 201510033857 A CN201510033857 A CN 201510033857A CN 104634706 A CN104634706 A CN 104634706A
- Authority
- CN
- China
- Prior art keywords
- coal
- training
- model
- fineness
- neural network
- 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
Landscapes
- Disintegrating Or Milling (AREA)
- Crushing And Grinding (AREA)
Abstract
The invention relates to a neural network-based soft measurement method for pulverized coal fineness. The method is characterized by comprising the following steps: S01, acquiring inlet and outlet associated data of a coal mill according to the type of the coal mill and dividing data samples into training samples and test samples; S02, establishing an artificial neural network model of the pulverized coal fineness R90; S03, determining the training precision or the maximum training number of times; S04, training the artificial neural network model by using the training samples till the training precision is satisfied or after the training of the model reaches the maximum training number of times, stopping the training of the model; S05, loading the artificial neural network model to the test samples and testing; and S06, successfully training the model if the test result shows that the model satisfies the requirements in precision and generalization ability. The method provided by the invention solves the problems caused by difficulty of sampling pulverized coal in a power plant and shortage of analysis of the pulverized coal fineness R90.
Description
Technical field
The present invention relates to a kind of fineness of pulverized coal R90 flexible measurement method, particularly a kind of fineness of pulverized coal flexible measurement method based on neural network.
Background technology
Fineness of pulverized coal R90 has consequence in the combustion adjustment of fuel-burning power plant, is boiler combustion optimization adjustment, improves the important reference of boiler efficiency.
Traditional analysis of the fineness of pulverized coal R90 based on on-site sampling all will obtain coal dust at the scene and carry out assay, and every day all will repeat same work, and workload is large, and coal dust sampling technique requires high, contaminated environment, sampling difficulty.Develop a kind of fineness of pulverized coal on-line analysis without the need to manual intervention to be necessary very much.
The method of current on-line checkingi fineness of pulverized coal R90 has microwave measuring method, laser Fluctuation Method, electrostatic measurement method, the instrument and equipment of these analytical approachs is expensive, in-site installation quantity is many, and safeguard by the impact of washing away and being bonded in sensor of on-the-spot pulverized coal particle, the data precision is poor, makes it be very restricted in actual applications.
Neural network is a Kind of Nonlinear Dynamical System, and its characteristic is distributed storage and the concurrent collaborative process of information.Although single neuronic structure is extremely simple, function is limited, and the behavior achieved by network system that a large amount of neuron is formed is extremely colourful.
The detection of fineness of pulverized coal R90 relates to the measurement of nonlinear parameter, therefore by the method for neural network, effectively control because fineness of pulverized coal R90 data lack to the difficulty that boiler operatiopn, burning optimization adjustment are brought from coal pulverizer operational factor, separator feature parameter, coal industry analysis to the foundation of the Nonlinear Mapping model of fineness of pulverized coal R90.
Summary of the invention
The object of this invention is to provide a kind of fineness of pulverized coal flexible measurement method based on neural network.
The present invention realizes by the following technical solutions: a kind of fineness of pulverized coal flexible measurement method based on neural network, it is characterized in that: comprise the following steps: step S01: according to coal pulverizer type, obtain this coal pulverizer import, outlet related data, the data sample of acquisition is divided into training sample and test sample book; Step S02: the artificial nerve network model setting up fineness of pulverized coal R90; Step S03: determine the training precision of described artificial network's model or maximum frequency of training; Step S04: utilize training sample to the training of described artificial network's model until meet training precision and require or after having reached maximum frequency of training to the training of model, stop the training of model and perform step S05; Step S05: test sample book be loaded into the artificial nerve network model of fineness of pulverized coal R90 and test; Step S06: if the result display model of test has met the requirement of precision and generalization ability two aspect, then model training success, otherwise return step S03.
In an embodiment of the present invention, described coal pulverizer related data comprises: coal pulverizer import primary air flow, coal pulverizer import wind-warm syndrome, coal pulverizer coal-supplying amount, coal pulverizer goes out one's intention as revealed in what one says powder temperature, coal pulverizer running current, separator for coal mill baffle opening or rotary gas separator rotating speed, raw coal air-dried moisture, the empty butt volatile matter of raw coal, the empty dry basis ash content of raw coal and raw coal Hardgrove grindability.
In an embodiment of the present invention, described artificial neural network comprises input layer, hidden layer and output layer; Described input layer is described coal pulverizer related data; Described hidden layer is the neuron node containing some; Namely described output layer exports coal pulverizer export coal powder fineness R90 for only having an output neuron node.
The present invention, by the method for neural network, controls because fineness of pulverized coal R90 data lack to the difficulty that boiler operatiopn, burning optimization adjustment are brought from coal pulverizer operational factor, separator feature parameter, coal industry analysis effectively to the foundation of the Nonlinear Mapping model of fineness of pulverized coal R90.
Accompanying drawing explanation
Fig. 1 is FB(flow block) of the present invention.
Embodiment
Fig. 1 is process flow diagram of the present invention, and key step of the present invention is: step S01: according to coal pulverizer type, obtains this coal pulverizer associated outlet data, the data sample of acquisition is divided into training sample and test sample book; Step S02: the artificial nerve network model setting up fineness of pulverized coal R90; Step S03: determine the training precision of described artificial network's model or maximum frequency of training; Step S04: utilize training sample to the training of described artificial network's model until meet training precision and require or after having reached maximum frequency of training to the training of model, stop the training of model and perform step S05; Step S05: test sample book be loaded into the artificial nerve network model of fineness of pulverized coal R90 and test; Step S06: if the result display model of test has met the requirement of precision and generalization ability two aspect, then model training success, otherwise return step S03.
Step S01 is further comprising the steps of: first collect coal pulverizer and import and export parameter (comprise import primary air flow, import wind-warm syndrome, a coal-supplying amount and go out one's intention as revealed in what one says powder temperature, coal pulverizer running current) and separator for coal mill baffle opening or rotary gas separator rotating speed; Sampling obtains raw coal sample and pulverized coal sample carries out raw coal technical analysis again, obtains the data sample of raw coal air-dried moisture, empty butt volatile matter, empty dry basis ash content, raw coal Hardgrove grindability and corresponding coal pulverizer export coal powder fineness R90; Secondly data sample is divided into training sample and test sample book.
Artificial neural network includes input layer, hidden layer and output layer.In the specific embodiment of the invention, the input layer of network is: coal pulverizer import primary air flow, coal pulverizer import wind-warm syndrome, coal pulverizer coal-supplying amount, coal pulverizer go out one's intention as revealed in what one says powder temperature, coal pulverizer running current, separator for coal mill baffle opening or rotary gas separator rotating speed, raw coal air-dried moisture, the empty butt volatile matter of raw coal, raw coal empty dry basis ash content, raw coal Hardgrove grindability, are the input vectors of one ten dimension; The hidden layer of network is the neuron node containing some; The output layer of network only has an output neuron node namely to export coal pulverizer export coal powder fineness R90.
Flexible measurement method based on neural network mainly utilizes coal pulverizer to import and export parameter (coal pulverizer import primary air flow, coal pulverizer import wind-warm syndrome, coal pulverizer coal-supplying amount, coal pulverizer go out one's intention as revealed in what one says powder temperature) and separator for coal mill baffle opening or rotary gas separator rotating speed, samples to obtain that raw coal sample and pulverized coal sample carry out raw coal technical analysis (air-dried moisture, empty butt volatile matter, empty dry basis ash content), raw coal Hardgrove grindability, fineness of pulverized coal R90 etc. the numerical value of Accurate Determining can carry out modeling analysis of neural network.The present invention, without the need to manual intervention, saves a large amount of manpower and more accurate than the data that obtain of instrumentation sampling assay.
The foregoing is only preferred embodiment of the present invention, all equalizations done according to the present patent application the scope of the claims change and modify, and all should belong to covering scope of the present invention.
Claims (3)
1., based on a fineness of pulverized coal flexible measurement method for neural network, it is characterized in that: comprise the following steps:
Step S01: according to coal pulverizer type, obtains this coal pulverizer import, outlet related data, the data sample of acquisition is divided into training sample and test sample book;
Step S02: the artificial nerve network model setting up fineness of pulverized coal R90;
Step S03: determine the training precision of described artificial network's model or maximum frequency of training;
Step S04: utilize training sample to the training of described artificial network's model until meet training precision and require or after having reached maximum frequency of training to the training of model, stop the training of model and perform step S05;
Step S05: test sample book be loaded into the artificial nerve network model of fineness of pulverized coal R90 and test;
Step S06: if the result display model of test has met the requirement of precision and generalization ability two aspect, then model training success, otherwise return step S03.
2. the fineness of pulverized coal flexible measurement method based on neural network according to claim 1, it is characterized in that: described in described step S01, coal pulverizer related data comprises: coal pulverizer import primary air flow, coal pulverizer import wind-warm syndrome, coal pulverizer coal-supplying amount, coal pulverizer goes out one's intention as revealed in what one says powder temperature, coal pulverizer running current, separator for coal mill baffle opening or rotary gas separator rotating speed, raw coal air-dried moisture, the empty butt volatile matter of raw coal, the empty dry basis ash content of raw coal and raw coal Hardgrove grindability.
3. the fineness of pulverized coal flexible measurement method based on neural network according to claim 1, is characterized in that: described artificial neural network comprises input layer, hidden layer and output layer; Described input layer is described coal pulverizer related data; Described hidden layer is the neuron node containing some; Namely described output layer exports coal pulverizer export coal powder fineness R90 for only having an output neuron node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510033857.8A CN104634706A (en) | 2015-01-23 | 2015-01-23 | Neural network-based soft measurement method for pulverized coal fineness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510033857.8A CN104634706A (en) | 2015-01-23 | 2015-01-23 | Neural network-based soft measurement method for pulverized coal fineness |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104634706A true CN104634706A (en) | 2015-05-20 |
Family
ID=53213700
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510033857.8A Pending CN104634706A (en) | 2015-01-23 | 2015-01-23 | Neural network-based soft measurement method for pulverized coal fineness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104634706A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106124371A (en) * | 2016-08-03 | 2016-11-16 | 西安理工大学 | A kind of Dual-Phrase Distribution of Gas olid fineness measurement device based on electrostatic method and measuring method |
| CN106124373A (en) * | 2016-06-17 | 2016-11-16 | 中国大唐集团科学技术研究院有限公司华东分公司 | A kind of measuring method of coal powder density |
| CN106251089A (en) * | 2016-08-15 | 2016-12-21 | 江苏方天电力技术有限公司 | A kind of ature of coal fluctuation status online soft sensor method |
| CN106529671A (en) * | 2016-10-28 | 2017-03-22 | 国网福建省电力有限公司 | Neural network-based raw coal total moisture soft measurement method |
| CN109541168A (en) * | 2018-11-26 | 2019-03-29 | 江苏方天电力技术有限公司 | A kind of economic fineness of pulverized coal on-line monitoring and method of adjustment |
| CN109615084A (en) * | 2017-09-30 | 2019-04-12 | 中电华创电力技术研究有限公司 | Positive-pressure type medium-speed pulverizer fuel pulverizing plant fineness of pulverized coal real-time monitoring system |
| CN111612211A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | Predictive modeling method for coal powder fineness of coal mill |
| CN111612210A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | Online optimization method for coal powder fineness of coal mill |
| CN111612212A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | On-line optimization model updating method for coal powder fineness of coal mill |
| CN112800995A (en) * | 2021-02-04 | 2021-05-14 | 广州甄好数码科技有限公司 | Intelligent particle size detection method using multi-scale feature weighting |
| CN113533622A (en) * | 2021-07-19 | 2021-10-22 | 华能国际电力股份有限公司上海石洞口第二电厂 | Coal quality prediction method for coal mill based on neural network |
| CN113792255A (en) * | 2021-11-17 | 2021-12-14 | 西安热工研究院有限公司 | Method for calculating coal mill outlet coal powder fineness on line based on power plant big data |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6014653A (en) * | 1996-01-26 | 2000-01-11 | Thaler; Stephen L. | Non-algorithmically implemented artificial neural networks and components thereof |
| CN2752755Y (en) * | 2004-10-15 | 2006-01-18 | 南京大陆中电科技股份有限公司 | Powder coal fineness on-line detector |
| CN101038277A (en) * | 2007-04-19 | 2007-09-19 | 东北大学 | Soft measurement method for coal power fineness in powdering producer |
| CN103995775A (en) * | 2014-05-20 | 2014-08-20 | 浪潮电子信息产业股份有限公司 | Testing data generating method based on neural network |
-
2015
- 2015-01-23 CN CN201510033857.8A patent/CN104634706A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6014653A (en) * | 1996-01-26 | 2000-01-11 | Thaler; Stephen L. | Non-algorithmically implemented artificial neural networks and components thereof |
| CN2752755Y (en) * | 2004-10-15 | 2006-01-18 | 南京大陆中电科技股份有限公司 | Powder coal fineness on-line detector |
| CN101038277A (en) * | 2007-04-19 | 2007-09-19 | 东北大学 | Soft measurement method for coal power fineness in powdering producer |
| CN103995775A (en) * | 2014-05-20 | 2014-08-20 | 浪潮电子信息产业股份有限公司 | Testing data generating method based on neural network |
Non-Patent Citations (2)
| Title |
|---|
| 周品: "《MATLAB神经网络设计与应用》", 31 March 2013 * |
| 王茂贵: "第六章 煤粉细度的优化选择及其软测量", 《中国优秀博硕士学位论文全文数据库（硕士） 工程科技Ⅱ辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106124373A (en) * | 2016-06-17 | 2016-11-16 | 中国大唐集团科学技术研究院有限公司华东分公司 | A kind of measuring method of coal powder density |
| CN106124373B (en) * | 2016-06-17 | 2018-08-31 | 中国大唐集团科学技术研究院有限公司华东分公司 | A kind of measurement method of coal powder density |
| CN106124371B (en) * | 2016-08-03 | 2019-03-26 | 西安理工大学 | A kind of Dual-Phrase Distribution of Gas olid fineness measurement device and measurement method based on electrostatic method |
| CN106124371A (en) * | 2016-08-03 | 2016-11-16 | 西安理工大学 | A kind of Dual-Phrase Distribution of Gas olid fineness measurement device based on electrostatic method and measuring method |
| CN106251089A (en) * | 2016-08-15 | 2016-12-21 | 江苏方天电力技术有限公司 | A kind of ature of coal fluctuation status online soft sensor method |
| CN106529671A (en) * | 2016-10-28 | 2017-03-22 | 国网福建省电力有限公司 | Neural network-based raw coal total moisture soft measurement method |
| CN109615084A (en) * | 2017-09-30 | 2019-04-12 | 中电华创电力技术研究有限公司 | Positive-pressure type medium-speed pulverizer fuel pulverizing plant fineness of pulverized coal real-time monitoring system |
| CN109541168A (en) * | 2018-11-26 | 2019-03-29 | 江苏方天电力技术有限公司 | A kind of economic fineness of pulverized coal on-line monitoring and method of adjustment |
| CN111612211A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | Predictive modeling method for coal powder fineness of coal mill |
| CN111612210A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | Online optimization method for coal powder fineness of coal mill |
| CN111612212A (en) * | 2020-04-10 | 2020-09-01 | 杭州电子科技大学 | On-line optimization model updating method for coal powder fineness of coal mill |
| CN112800995A (en) * | 2021-02-04 | 2021-05-14 | 广州甄好数码科技有限公司 | Intelligent particle size detection method using multi-scale feature weighting |
| CN113533622A (en) * | 2021-07-19 | 2021-10-22 | 华能国际电力股份有限公司上海石洞口第二电厂 | Coal quality prediction method for coal mill based on neural network |
| CN113792255A (en) * | 2021-11-17 | 2021-12-14 | 西安热工研究院有限公司 | Method for calculating coal mill outlet coal powder fineness on line based on power plant big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104634706A (en) | Neural network-based soft measurement method for pulverized coal fineness | |
| Rotter et al. | New techniques for the characterization of refuse-derived fuels and solid recovered fuels | |
| CN102226904A (en) | Soft measurement method for air quantity of primary air of large-scale boiler in power station | |
| Ahmed | Wind tunnel designs and their diverse engineering applications | |
| CN103823971A (en) | Oxygen content detecting method for boiler fuel gas of coal-fired power station | |
| CN112525977A (en) | Unorganized VOCsLeakage online monitoring and tracing method and system | |
| Song et al. | Nonlinear creep model and parameter identification of mudstone based on a modified fractional viscous body | |
| CN103617364A (en) | Large slewing bearing residual service life prediction method based on small samples | |
| CN112967764A (en) | Multi-technology coupled pollutant source analysis method and device | |
| Benaberrahmane et al. | Analytical evaluation of frequencies of bidirectional FG thick beams in thermal environment and resting on different foundation | |
| CN103698540B (en) | Rapid determination method and device for ash content in lignocellulosic biomass fuel | |
| CN106529671A (en) | Neural network-based raw coal total moisture soft measurement method | |
| CN104198430A (en) | Detecting method of substituent content of CAB (cellulose acetate butyrate) | |
| Bietresato et al. | Engine test stand layout and post processing tools for the detection of many engine performance parameters | |
| CN104142385A (en) | Evaluating method and evaluating system for power station boiler fire coal characteristics | |
| Banerjee et al. | Advanced air quality monitoring system using raspberry pi | |
| Zhao et al. | Research on the application of computer intelligent detection in civil engineering technology | |
| Jin et al. | Compilation of drilling load spectrum based on the characteristics of drilling force | |
| Steiner et al. | Particulate emissions from biomass power plants: a practical review and measurement uncertainty issues | |
| Polivyanchuk et al. | Mathematical modeling of diesel engine operation mode influence on mass emission of particulate matter with exhaust gases using microtunnel | |
| Tao et al. | Evaluation study on ecosystem based on system dynamics: A case study of Gongga Mountain forest ecosystem | |
| CN118051742B (en) | Method and system for measuring calorific value of biomass fuel based on multicomponent parameter analysis | |
| Ko¨ ppel et al. | On the development and application of the fast-response aerodynamic probe system in turbomachines—Part 3: Comparison of averaging methods applied to centrifugal compressor measurements | |
| CN116879513B (en) | Verification method, device, equipment and storage medium of gas analysis system | |
| Xie et al. | SVR based prediction modeling for carbon content of pulverized coal in primary air pipe of thermal power plant boiler |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150520 |
|
| RJ01 | Rejection of invention patent application after publication |