CN112797402A - Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system - Google Patents
Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system Download PDFInfo
- Publication number
- CN112797402A CN112797402A CN202011616655.3A CN202011616655A CN112797402A CN 112797402 A CN112797402 A CN 112797402A CN 202011616655 A CN202011616655 A CN 202011616655A CN 112797402 A CN112797402 A CN 112797402A
- Authority
- CN
- China
- Prior art keywords
- submodule
- data
- prediction
- clustering
- model
- 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
- 238000007781 pre-processing Methods 0.000 claims abstract description 29
- 238000013500 data storage Methods 0.000 claims abstract description 16
- 238000004891 communication Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 16
- 238000010977 unit operation Methods 0.000 claims description 10
- 239000003245 coal Substances 0.000 claims description 8
- 238000010219 correlation analysis Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 8
- 239000002893 slag Substances 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 7
- 238000003064 k means clustering Methods 0.000 claims description 3
- 239000013049 sediment Substances 0.000 claims 1
- 238000000034 method Methods 0.000 abstract description 11
- 238000010248 power generation Methods 0.000 abstract description 7
- 239000000463 material Substances 0.000 description 13
- 238000002485 combustion reaction Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 3
- 239000008186 active pharmaceutical agent Substances 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 239000003344 environmental pollutant Substances 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 231100000719 pollutant Toxicity 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000007306 turnover Effects 0.000 description 1
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C10/00—Fluidised bed combustion apparatus
- F23C10/02—Fluidised bed combustion apparatus with means specially adapted for achieving or promoting a circulating movement of particles within the bed or for a recirculation of particles entrained from the bed
- F23C10/04—Fluidised bed combustion apparatus with means specially adapted for achieving or promoting a circulating movement of particles within the bed or for a recirculation of particles entrained from the bed the particles being circulated to a section, e.g. a heat-exchange section or a return duct, at least partially shielded from the combustion zone, before being reintroduced into the combustion zone
- F23C10/08—Fluidised bed combustion apparatus with means specially adapted for achieving or promoting a circulating movement of particles within the bed or for a recirculation of particles entrained from the bed the particles being circulated to a section, e.g. a heat-exchange section or a return duct, at least partially shielded from the combustion zone, before being reintroduced into the combustion zone characterised by the arrangement of separation apparatus, e.g. cyclones, for separating particles from the flue gases
- F23C10/10—Fluidised bed combustion apparatus with means specially adapted for achieving or promoting a circulating movement of particles within the bed or for a recirculation of particles entrained from the bed the particles being circulated to a section, e.g. a heat-exchange section or a return duct, at least partially shielded from the combustion zone, before being reintroduced into the combustion zone characterised by the arrangement of separation apparatus, e.g. cyclones, for separating particles from the flue gases the separation apparatus being located outside the combustion chamber
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C10/00—Fluidised bed combustion apparatus
- F23C10/18—Details; Accessories
- F23C10/22—Fuel feeders specially adapted for fluidised bed combustion apparatus
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C10/00—Fluidised bed combustion apparatus
- F23C10/18—Details; Accessories
- F23C10/24—Devices for removal of material from the bed
- F23C10/26—Devices for removal of material from the bed combined with devices for partial reintroduction of material into the bed, e.g. after separation of agglomerated parts
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C10/00—Fluidised bed combustion apparatus
- F23C10/18—Details; Accessories
- F23C10/28—Control devices specially adapted for fluidised bed, combustion apparatus
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23C—METHODS OR APPARATUS FOR COMBUSTION USING FLUID FUEL OR SOLID FUEL SUSPENDED IN A CARRIER GAS OR AIR
- F23C2206/00—Fluidised bed combustion
- F23C2206/10—Circulating fluidised bed
Abstract
The invention relates to a pants leg type circulating fluidized bed boiler unit and a bed pressure prediction system, belonging to the technical field of intelligent power generation, wherein the bed pressure prediction system comprises a data preprocessing submodule, an input selection submodule, a clustering algorithm submodule, a classified data storage submodule, a clustering model submodule and a prediction model submodule; the clustering algorithm submodule inputs data and sets weight, and performs unsupervised clustering on the data so as to distinguish various working conditions; and the prediction model submodule stores the trained prediction submodel, selects the prediction submodel according to the signal and outputs a predicted value. According to the method, the prediction models under all categories are constructed in a clustering mode, so that the prediction accuracy is improved, and the influence of multi-working-condition operation on a prediction system is avoided; the method has strong popularization and applicability, different clustering weights are set aiming at the units with different parameters, different clustering models are constructed, and the method has strong operability and flexibility.
Description
Technical Field
The invention belongs to the technical field of intelligent power generation, and particularly relates to a pants-leg type circulating fluidized bed boiler unit and a bed pressure prediction system.
Background
In recent years, the demand for environmental protection has been increased. The new energy electric power in China develops rapidly, and the wind power installation exceeds the United states and is the first place in the world; the installed photovoltaic capacity of 30.1GW is increased in China in 2019, the China continuously occupies the first place of the world for 7 years, and the shipment volume of the photovoltaic industry occupies more than 70% of the world. By 2019, the national wind and light loading amount reaches 3.9 hundred million kilowatts, and the wind and light loading device is the first in the world.
However, the instability, intermittency and uncertainty of the new energy generator set are very unfavorable for the stable operation of the power grid side. If the disadvantages of the good-system energy cannot be solved, the grid connection of the new energy source unit with large capacity is difficult to realize. Aiming at the problem of new energy power generation, the state provides a peak-load and frequency-modulation strategy to enhance the stability of new energy grid connection. Thermal power generation is used as a main power generation mode in China, and has the function of peak regulation and frequency modulation, so that instability caused by new energy power grid connection is compensated, and the stability of a power grid is ensured. Aiming at peak regulation and frequency modulation, the traditional thermal power generating unit has the problem of poor load regulation capacity, and the traditional thermal power generating unit under deep peak regulation needs to assist combustion in a furnace through oil feeding. Meanwhile, the unit can generate a large amount of pollutants in a deep peak regulation state, and is not beneficial to environmental protection and sustainable development. The circulating fluidized bed boiler is a clean coal combustion technology and has the following advantages: the fuel has high combustion efficiency, wide fuel adaptability, less pollutant discharge, better load regulation performance, high-efficiency comprehensive utilization of resources and the like, and is widely applied to the industrial fields of power generation, heat supply and the like at home and abroad.
With the technical progress, large-capacity circulating fluidized bed units are continuously developed. However, the problem of insufficient combustion is brought by the large-scale circulating fluidized bed, the penetration capability of secondary air in the combustion process is difficult to ensure by the traditional furnace type, most of the inside of the furnace is in a negative oxygen combustion state, and economic loss is caused. In order to solve this problem, a pant-leg type boiler is born from this. The boiler with the underpants legs solves the problem of combustion, but also causes the problems of transverse transportation and fluctuation of materials on two sides of the boiler. When large deviation occurs to materials on two sides of a hearth, positive feedback is easily formed by transverse transfer of the materials, so that the phenomenon that the material on one side is rapidly raised occurs, and a unit is stopped when the phenomenon is serious.
Therefore, the balance of the materials on the two sides of the hearth is an important premise for ensuring the stable and safe operation of the unit. In the underpants leg type circulating fluidized bed unit, the single-side bed pressure can directly reflect the quantity of single-side materials, and the concentration state of the materials can be visually shown through the prediction of the bed pressure.
The bed pressure of a pants-leg type circulating fluidized bed boiler is mainly influenced by materials distributed in a hearth, the balance of the materials on two sides of the hearth is very important in the running process of a unit, the control mode adopted at present is mainly controlled according to field measuring point data, and when the bed pressure is unstable or a turnover accident occurs, the traditional control mode cannot control the bed pressure very timely. Therefore, the instability of the bed pressure is easy to become a potential safety hazard in the operation of the unit. If the bed pressure can be predicted, the method is very favorable for the safe and stable operation of the unit, and is also a very important link in the intelligent power generation technology.
Disclosure of Invention
The invention provides a short leg type circulating fluidized bed boiler unit and a bed pressure prediction system for solving the technical problems in the prior art, which predict and monitor the bed pressure of the unit through the bed pressure prediction system, and play a role in guiding the operation of field personnel, thereby improving the safety and stability of the unit operation.
The invention comprises the following technical scheme:
a pants leg type circulating fluidized bed boiler unit comprises a coal feeder, a slag cooler, a material return valve and an air gauge; the coal feeder and the slag cooler are both provided with motors, and the motors are provided with rotating speed sensors; a valve position sensor is arranged on the feed back valve; the air gauge, the rotating speed sensor and the valve position sensor are all connected with the DCS; the DCS is connected with a plug-in system through a communication system, and the plug-in system comprises a bed pressure prediction system based on an intelligent algorithm; the DCS provides a data interface for the bed pressure prediction system.
Further, the bed pressure prediction system comprises a data preprocessing submodule, an input selection submodule, a clustering algorithm submodule, a classified data storage submodule, a clustering model submodule and a prediction model submodule; the clustering algorithm submodule inputs data and sets weight, and performs unsupervised clustering on the data so as to distinguish various working conditions; the classified data storage submodule stores a data classification result output by the clustering model submodule, and the data classification result is used for training and verifying a prediction submodel; the clustering model submodule stores the existing clustering model, distinguishes the type of the current working condition according to the input of online data, is connected with the prediction model submodule and provides a model selection signal for the prediction model submodule; and the prediction model submodule stores the trained prediction submodel, selects the prediction submodel according to the signal of the clustering model submodule, outputs a predicted value and realizes the prediction of online data.
Further, the data preprocessing submodule performs default value and dead pixel screening and data normalization processing on the unit operation data provided by the DCS. The operation data comprises a historical database and online data, and the data preprocessing submodule selects input variables when processing the online data. The bed pressure prediction system is trained and verified through a historical database to obtain a prediction submodel.
The input selection submodule performs variable correlation analysis on data, selects the input of the clustering algorithm submodule by combining mechanism priori knowledge, updates the result in the data preprocessing submodule and reduces the operation amount of online data processing.
A bed pressure prediction system of a pants leg type circulating fluidized bed boiler unit comprises a data preprocessing submodule, an input selection submodule, a clustering algorithm submodule, a classified data storage submodule, a clustering model submodule and a prediction model submodule; the clustering algorithm submodule inputs data and sets weight, and performs unsupervised clustering on the data so as to distinguish various working conditions; the classified data storage submodule stores a data classification result output by the clustering model submodule, and the data classification result is used for training and verifying a prediction submodel; the clustering model submodule stores the existing clustering model, distinguishes the type of the current working condition according to the input of online data, is connected with the prediction model submodule and provides a model selection signal for the prediction model submodule; and the prediction model submodule stores the trained prediction submodel, selects the prediction submodel according to the signal of the clustering model submodule, outputs a predicted value and realizes the prediction of online data. The bed pressure prediction system is used for providing bed pressure prediction values under different operation parameters.
Further, the bed pressure prediction system and the DCS are in data communication through a communication system, and the DCS provides a data interface for the bed pressure prediction system. The communication system hardware is connected with the server through a network cable; in communication transmission, communication between the DCS system and the server is realized through an http protocol mode, a PI WEB API mode, a pyodbc mode and the like.
Further, the data preprocessing submodule performs default value and dead pixel screening and data normalization processing on the unit operation data provided by the DCS.
Furthermore, the unit operation data comprises a historical database and online data, and the data preprocessing submodule selects input variables when processing the online data.
Further, the bed pressure prediction system is trained and verified through the historical database to obtain a prediction sub-model.
Furthermore, the input selection submodule performs variable correlation analysis on the data, selects the input of the clustering algorithm submodule by combining mechanism prior knowledge, and updates the result in the data preprocessing submodule, thereby reducing the operation amount of online data processing.
Further, the sub-module of the clustering algorithm adopts a K-means clustering mode, wherein the distance formula adopts an Euclidean distance formula with weight, and the formula is expressed as follows:
wherein
Represents the weight KmThe corresponding variable.
The internal calculation process of the bed pressure prediction system comprises the following steps: collecting historical operation data from a historical database of a DCS (distributed control system), enabling the data to enter a bed pressure prediction system through a communication system, preprocessing the data by a data preprocessing submodule, including screening and cleaning of default and bad point data, normalizing the data, and transmitting the processed data to an input selection submodule; the input selection submodule performs correlation analysis on the variables according to the data and mechanism priori knowledge, screens proper input variables and transmits the input variables to the clustering algorithm submodule; the clustering algorithm submodule carries out clustering according to input data, and corrects the algorithm according to clustering algorithm weight setting to form a clustering model submodule; dividing historical data according to the clustering model submodule, and storing each class of data in the classification data storage submodule; dividing data into a training set and a test set according to the proportion according to the data of each category, and establishing an independent prediction sub-model under each category; and various prediction submodels are stored in the prediction model submodule, and the prediction model submodule selects a corresponding prediction submodel according to the model selection signal of the clustering model submodule to realize the prediction of the online data.
The invention has the advantages and positive effects that:
1. according to the method, the prediction models under all categories are constructed in a clustering mode, so that the prediction accuracy is improved, and the influence of multi-working-condition operation on a prediction system is avoided.
2. The method has strong popularization and applicability, different clustering weights are set aiming at the units with different parameters, different clustering models are constructed, and the method has strong operability and flexibility.
Drawings
Fig. 1 is a schematic view of the structure of a circulating fluidized bed boiler unit according to the present invention.
Fig. 2 is a schematic flow diagram of the bed pressure prediction system of the present invention.
In the figure, 1-coal feeder; 2-a slag cooler; 3-a material return valve; 4-a wind gauge; 5-DCS system; 6-a communication system; 7-plug-in system; 8-a history database; 9-online data;
11-bed pressure prediction system; 11-1-data preprocessing submodule; 11-2-input selection submodule; 11-3-clustering algorithm submodule; 11-4-clustering model submodule; 11-5-classify the data storage sub-module; 11-6-predictor model; 11-7-the prediction model submodule.
Detailed Description
To further clarify the disclosure of the present invention, its features and advantages, reference is made to the following examples taken in conjunction with the accompanying drawings.
Example 1: referring to the attached drawings 1-2, a underpants leg type circulating fluidized bed boiler unit comprises a coal feeder 1, a slag cooler 2, a material return valve 3 and an air gauge 4; the coal feeder 1 and the slag cooler 2 are both provided with motors, and the motors are provided with rotating speed sensors; a valve position sensor is arranged on the material return valve 3; the air gauge 4, the rotating speed sensor and the valve position sensor are all connected with the DCS system 5; the DCS system 5 is connected with a plug-in system 7 through a communication system 6, and the plug-in system 7 is provided with a bed pressure prediction system 11 based on an intelligent algorithm; the DCS system 5 provides a data interface for the bed pressure prediction system 11. The slag cooler 2 adopts a FH-30-150 type composite slag cooler.
Referring to fig. 2, the bed pressure prediction system 11 includes a data preprocessing sub-module 11-1, an input selection sub-module 11-2, a clustering algorithm sub-module 11-3, a classified data storage sub-module 11-5, a clustering model sub-module 11-4, and a prediction model sub-module 11-7; the clustering algorithm submodule 11-3 inputs data and sets weights, and performs unsupervised clustering on the data so as to distinguish the working conditions; the classified data storage submodule 11-5 stores a data classification result output by the clustering model submodule 11-4, and the data classification result is used for training and verifying the prediction submodel 11-6; the clustering model submodule 11-4 stores the existing clustering model, distinguishes the current working condition type according to the input of the online data 9, is connected with the prediction model submodule 11-6 and provides a model selection signal for the prediction model submodule 11-6; the prediction model submodule 11-7 stores the trained prediction submodel 11-6, selects the prediction submodel 11-6 according to the signal of the clustering model submodule 11-4, outputs a predicted value and realizes the prediction of online data.
The data preprocessing submodule 11-1 performs default value, dead pixel screening and data normalization processing on the unit operation data provided by the DCS system 5. The operating data comprises a historical database 8 and online data 9, and the data preprocessing submodule 11-1 selects input variables when processing the online data 9. The bed pressure prediction system 11 is trained and verified through the historical database 8 to obtain a prediction sub-model 11-6. The input selection submodule 11-2 performs variable correlation analysis on data, selects the input of the clustering algorithm submodule 11-3 by combining mechanism prior knowledge, updates the result in the data preprocessing submodule 11-1, and reduces the operation amount of online data processing.
The internal calculation process of the bed pressure prediction system comprises the following steps: historical operation data are collected from a historical database 8 of a DCS (distributed control system) 5, the data enter a bed pressure prediction system 11 through a communication system 6, a data preprocessing submodule 11-1 firstly preprocesses the data, including screening and cleaning of default and bad point data and data normalization, and transmits the processed data to an input selection submodule 11-2; the input selection submodule 11-2 performs correlation analysis on the variables according to the data and mechanism prior knowledge, screens proper input variables and transmits the input variables to the clustering algorithm submodule 11-3; the clustering algorithm sub-module 11-3 carries out clustering according to input data, and corrects the algorithm according to clustering algorithm weight setting to form a clustering model sub-module 11-4; dividing historical data according to the clustering model submodule 11-4, and storing each class of data in the classification data storage submodule 11-5; dividing data into a training set and a test set according to the proportion according to the data of each category, and establishing an independent prediction sub-model 11-6 under each category; the various prediction submodels 11-6 are stored in the prediction model submodule 11-7, and the prediction model submodule 11-7 selects the corresponding prediction submodel 11-6 according to the model selection signal of the clustering model submodule 11-4, so that the online data 9 can be predicted.
Example 2: referring to the attached figure 2, the bed pressure prediction system of the pants leg type circulating fluidized bed boiler unit comprises a data preprocessing submodule 11-1, an input selection submodule 11-2, a clustering algorithm submodule 11-3, a classification data storage submodule 11-5, a clustering model submodule 11-4 and a prediction model submodule 11-7; the clustering algorithm submodule 11-3 inputs data and sets weights, and performs unsupervised clustering on the data so as to distinguish the working conditions; the classified data storage submodule 11-5 stores a data classification result output by the clustering model submodule 11-4, and the data classification result is used for training and verifying the prediction submodel 11-6; the clustering model submodule 11-4 stores the existing clustering model, distinguishes the current working condition type according to the input of the online data 9, is connected with the prediction model submodule 11-6 and provides a model selection signal for the prediction model submodule 11-6; the prediction model submodule 11-7 stores the trained prediction submodel 11-6, selects the prediction submodel 11-6 according to the signal of the clustering model submodule 11-4, outputs a predicted value and realizes the prediction of online data.
The bed pressure prediction system 11 and the DCS 5 are in data communication through the communication system 6, and the DCS 5 provides a data interface for the bed pressure prediction system 11. The hardware of the communication system 6 is connected with a server through a network cable; in communication transmission, communication between the DCS system 5 and the server is realized through an http protocol mode, a PI mode, a WEB mode, an API mode, a pyodbc mode and the like.
The data preprocessing submodule 11-1 performs default value, dead pixel screening and data normalization processing on the unit operation data provided by the DCS system 5. The unit operation data comprises a historical database 8 and online data 9, and the data preprocessing submodule 11-1 selects input variables when processing the online data 9. The bed pressure prediction system 11 is trained and verified through the historical database 8 to obtain a prediction sub-model 11-6.
The input selection submodule 11-2 performs variable correlation analysis on data, selects the input of the clustering algorithm submodule 11-3 by combining mechanism prior knowledge, updates the result in the data preprocessing submodule 11-1, and reduces the operation amount of online data processing.
The clustering algorithm sub-module 11-3 adopts a K-means clustering mode, wherein a distance formula adopts an Euclidean distance formula with weight.
The formula is expressed as follows:
wherein
Represents the weight KmThe corresponding variable. The specific weight can be adjusted according to the field condition, expert experience and test effect. The working procedure of this example is the same as in example 1.
While the preferred embodiments of the present invention have been illustrated and described, it will be appreciated by those skilled in the art that the foregoing embodiments are illustrative and not limiting, and that many changes may be made in the form and details of the embodiments of the invention without departing from the spirit and scope of the invention as defined in the appended claims. All falling within the scope of protection of the present invention.
Claims (10)
1. The utility model provides a pants leg type circulating fluidized bed boiler unit, includes coal feeder, cold sediment ware, feed back valve and anemograph, its characterized in that: the coal feeder and the slag cooler are both provided with motors, and the motors are provided with rotating speed sensors; a valve position sensor is arranged on the feed back valve; the air gauge, the rotating speed sensor and the valve position sensor are all connected with the DCS; the DCS is connected with a plug-in system through a communication system, and the plug-in system comprises a bed pressure prediction system based on an intelligent algorithm; the DCS provides a data interface for the bed pressure prediction system.
2. The pant leg type circulating fluidized bed boiler unit according to claim 1, characterized in that: the bed pressure prediction system comprises a data preprocessing submodule, an input selection submodule, a clustering algorithm submodule, a classified data storage submodule, a clustering model submodule and a prediction model submodule; the clustering algorithm submodule inputs data and sets weight, and performs unsupervised clustering on the data so as to distinguish various working conditions; the classified data storage submodule stores a data classification result output by the clustering model submodule, and the data classification result is used for training and verifying a prediction submodel; the clustering model submodule stores the existing clustering model, distinguishes the type of the current working condition according to the input of online data, is connected with the prediction model submodule and provides a model selection signal for the prediction model submodule; and the prediction model submodule stores the trained prediction submodel, selects the prediction submodel according to the signal of the clustering model submodule, outputs a predicted value and realizes the prediction of online data.
3. The pant-leg type circulating fluidized bed boiler unit according to claim 2, wherein: the data preprocessing submodule carries out default value, dead pixel screening and data normalization processing on the unit operation data provided by the DCS; the operation data comprises a historical database and online data, and the data preprocessing submodule selects an input variable when processing the online data; the bed pressure prediction system is trained and verified through a historical database to obtain a prediction sub-model; the input selection submodule performs variable correlation analysis on data, selects the input of the clustering submodule by combining mechanism priori knowledge, and updates the result in the data preprocessing submodule.
4. The utility model provides a pants leg type circulating fluidized bed boiler unit bed pressure prediction system which characterized in that: the system comprises a data preprocessing submodule, an input selection submodule, a clustering algorithm submodule, a classification data storage submodule, a clustering model submodule and a prediction model submodule; the clustering algorithm submodule inputs data and sets weight, and performs unsupervised clustering on the data so as to distinguish various working conditions; the classified data storage submodule stores a data classification result output by the clustering model submodule, and the data classification result is used for training and verifying a prediction submodel; the clustering model submodule stores the existing clustering model, distinguishes the type of the current working condition according to the input of online data, is connected with the prediction model submodule and provides a model selection signal for the prediction model submodule; and the prediction model submodule stores the trained prediction submodel, selects the prediction submodel according to the signal of the clustering model submodule, outputs a predicted value and realizes the prediction of online data.
5. The system for predicting the bed pressure of a pant-leg type circulating fluidized bed boiler unit according to claim 4, wherein: and the bed pressure prediction system and the DCS are in data communication through a communication system, and the DCS provides a data interface for the bed pressure prediction system.
6. The system for predicting bed pressure of a pant-leg type circulating fluidized bed boiler unit according to claim 5, wherein: and the data preprocessing submodule is used for carrying out default value, dead pixel screening and data normalization processing on the unit operation data provided by the DCS.
7. The system for predicting the bed pressure of a pant-leg type circulating fluidized bed boiler unit according to claim 6, wherein: the unit operation data comprises a historical database and online data, and the data preprocessing submodule selects input variables when processing the online data.
8. The system for predicting bed pressure of a pant-leg type circulating fluidized bed boiler unit according to claim 7, wherein: and the bed pressure prediction system is trained and verified through the historical database to obtain a prediction submodel.
9. The system for predicting the bed pressure of a pants-leg type circulating fluidized bed boiler unit according to any one of claims 4 to 8, wherein: the input selection submodule performs variable correlation analysis on data, selects the input of the clustering algorithm submodule by combining mechanism priori knowledge, and updates the result in the data preprocessing submodule.
10. The system for predicting the bed pressure of a pants-leg type circulating fluidized bed boiler unit according to any one of claims 4 to 8, wherein: the clustering algorithm sub-module adopts a K-means clustering mode, wherein the distance formula adopts an Euclidean distance formula with weight, and the formula expression is as follows:
wherein
Represents the weight KmThe corresponding variable.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011616655.3A CN112797402A (en) | 2020-12-31 | 2020-12-31 | Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011616655.3A CN112797402A (en) | 2020-12-31 | 2020-12-31 | Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112797402A true CN112797402A (en) | 2021-05-14 |
Family
ID=75804732
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011616655.3A Pending CN112797402A (en) | 2020-12-31 | 2020-12-31 | Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112797402A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693452A (en) * | 2012-05-11 | 2012-09-26 | 上海交通大学 | Multiple-model soft-measuring method based on semi-supervised regression learning |
| CN104296131A (en) * | 2014-10-23 | 2015-01-21 | 东南大学 | Multivariable cooperative control method for double-hearth circulating fluidized bed unit |
| CN105066121A (en) * | 2015-07-29 | 2015-11-18 | 华北电力大学 | Dynamic bed temperature prediction system and method of circulating fluidized bed boiler |
| CN106224939A (en) * | 2016-07-29 | 2016-12-14 | 浙江大学 | Circulating fluid bed domestic garbage burning boiler bed temperature Forecasting Methodology and system |
| CN107145981A (en) * | 2017-05-05 | 2017-09-08 | 张勇 | A kind of optimization method and device of boiler of power plant exhaust gas temperature |
-
2020
- 2020-12-31 CN CN202011616655.3A patent/CN112797402A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102693452A (en) * | 2012-05-11 | 2012-09-26 | 上海交通大学 | Multiple-model soft-measuring method based on semi-supervised regression learning |
| CN104296131A (en) * | 2014-10-23 | 2015-01-21 | 东南大学 | Multivariable cooperative control method for double-hearth circulating fluidized bed unit |
| CN105066121A (en) * | 2015-07-29 | 2015-11-18 | 华北电力大学 | Dynamic bed temperature prediction system and method of circulating fluidized bed boiler |
| CN106224939A (en) * | 2016-07-29 | 2016-12-14 | 浙江大学 | Circulating fluid bed domestic garbage burning boiler bed temperature Forecasting Methodology and system |
| CN107145981A (en) * | 2017-05-05 | 2017-09-08 | 张勇 | A kind of optimization method and device of boiler of power plant exhaust gas temperature |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN203147907U (en) | Urban centralized heating linkage optimal control system | |
| CN108490790A (en) | A kind of overheating steam temperature active disturbance rejection cascade control method based on multiple-objection optimization | |
| CN107358345A (en) | The distributed triple-generation system optimizing operation method of meter and dsm | |
| CN112146156A (en) | Multi-mode flexible operation method and system for power plant with electric boiler | |
| CN110244568B (en) | Energy hub model of industrial enterprise microgrid and multi-energy complementary optimization control method thereof | |
| CN115688448A (en) | Optimal scheduling method for multi-region comprehensive energy system considering shared energy storage | |
| CN106873518A (en) | A kind of environmentally friendly fired power generating unit system | |
| CN114358994A (en) | Smart Internet of things monitoring method for carbon emission based on green code | |
| CN112797402A (en) | Underpants leg type circulating fluidized bed boiler unit and bed pressure prediction system | |
| CN112696667A (en) | Bed temperature early warning system of circulating fluidized bed boiler unit | |
| CN116433046A (en) | Multi-main-game-considered network storage collaborative planning method for multi-energy power distribution system | |
| CN109802447A (en) | A kind of CCHP system call evaluation method | |
| CN109635341A (en) | A kind of pressure containing part life-span prediction method of three flue double reheat boiler of tail portion | |
| CN115759507A (en) | Sensitivity factor analysis method for comprehensive low-carbon benefit of offshore wind power system | |
| CN210264931U (en) | Biogas generator set control system based on intelligent cloud | |
| CN113822598A (en) | Energy efficiency analysis method and system of comprehensive energy system considering energy flow link | |
| CN113191083A (en) | Optimization design method of flue gas waste heat recovery system considering all-working-condition external parameter change | |
| CN112488435A (en) | Thermal power generation enterprise big data real-time positive balance power supply coal consumption calculation and control method | |
| CN114779722B (en) | Intelligent combustion optimization control system and method for coal-fired power station boiler | |
| CN111798044B (en) | RIES operation planning simulation method based on improved minimum cross entropy | |
| CN215175450U (en) | Heating plant system capable of automatically adjusting heat supply and demand matching | |
| CN112803418B (en) | Optimal configuration method of distributed power flow controller | |
| CN114992520B (en) | Control method of water-doped heat tracing system of oil transfer station in unattended mode | |
| CN215107728U (en) | Land-saving type thermal power plant | |
| CN113341716B (en) | Large-scale coal-fired power plant CO based on artificial intelligence 2 Optimized scheduling method for trapping system |
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: 20210514 |