CN112163350B - Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system - Google Patents
Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system Download PDFInfo
- Publication number
- CN112163350B CN112163350B CN202011202911.4A CN202011202911A CN112163350B CN 112163350 B CN112163350 B CN 112163350B CN 202011202911 A CN202011202911 A CN 202011202911A CN 112163350 B CN112163350 B CN 112163350B
- Authority
- CN
- China
- Prior art keywords
- coal
- fineness
- data
- pulverized coal
- measurement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume, or surface-area of porous materials
- G01N15/02—Investigating particle size or size distribution
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
Abstract
The system overcomes the defects of time lag of periodic sampling and measurement of the traditional pulverized coal fineness, off-line operation and difficulty in realizing automatic intelligent cleaning of a combustion process; compared with the existing pulverized coal fineness online measurement method, the method has the advantages of good reliability and better economic advantage.
Description
Technical Field
The invention relates to the field of coal powder measurement, in particular to a coal powder fineness real-time online soft measurement system and method of a double model of a coal powder production system.
Background
The combustion of the pulverized coal furnace is pulverized coal suspension combustion, and coal particles are gradually ignited and burnt from the outer surface to the inner surface. When the fineness of the pulverized coal is smaller, oxygen is difficult to fully contact with the surface of pulverized coal particles, the combustion is insufficient, the mechanical incomplete combustion heat loss is increased, and the pulverized coal with smaller granularity is easier to discharge along with flue gas, so that larger fly ash heat loss is caused; when the fineness of the pulverized coal is larger, the combustion center is deviated, the risk of coking and tube explosion is increased, under the condition that the pulverized coal is left for a certain time, the pulverized coal is not burnt completely, the carbon content of slag is increased rapidly, and larger slag discharging heat loss is caused. Therefore, proper pulverized coal fineness control and adjustment are necessary preconditions for ensuring the thermal efficiency of the boiler, in particular to a coal-fired boiler.
Currently, coal fines fineness is mainly measured by off-line periodic sampling. The screening method is the most common offline measurement method with the advantages of simplicity, easy implementation, high precision and the like, but the time delay limits the automation and the intellectualization of the boiler; the optical measurement method can realize online measurement and has high precision, but the laser optical sensor is easy to be stained and can only be used for single-point measurement; the ultrasonic method can obtain the fineness of coal particles in the two-phase flow through the relation between sound attenuation and frequency, but the method is greatly influenced by background noise signals; the electrostatic method is also a popular online measurement means, but the electrostatic signal is greatly influenced by the running environment of the power plant, and the practical application of the electrostatic method is greatly limited; the coal powder fineness is measured by the Zhejiang university team by using an image processing method, but the method is only effective for millimeter-sized coal particles, and further research and study are still needed for measuring the micron-sized coal particles for pneumatic conveying.
Disclosure of Invention
Aiming at the problems of the prior art scheme, the invention discloses an on-line coupling soft measurement system and method for the fineness of coal powder, which overcome the defects of time lag, off-line operation and difficulty in realizing automatic intelligent cleaning of a combustion process in the conventional periodic sampling measurement of the fineness of coal powder; compared with the existing pulverized coal fineness online measurement method, the method has the advantages of good reliability and better economic advantage.
The utility model provides a pulverized coal fineness real-time on-line soft measurement system of two models of powder process system, includes numerical model database, CK measurement and control module, data extraction and conversion module, parameter data preprocessing module, pulverized coal fineness calculation module, data verification feedback module, KZ control module based on the T-S model, its characterized in that:
the coal powder fineness calculation module based on the T-S model is connected with the numerical model database;
the data extraction and conversion module is connected with the CK measurement and control module, the data verification feedback module is connected with the KZ control module, and the CK measurement and control module and the KZ control module are connected with each other;
the CK measurement and control module comprises measurement and control sensors CK-01, CK-02, CK-03, CK-04 and expansion units thereof, and is arranged on a coal bunker feeder, a coal mill, a coal dust separator and a coal dust bunker to measure and give out characteristic parameters of the coal pulverizing system;
the data extraction and conversion module is used for expressing and extracting and converting parameters of the characteristic parameters of the coal pulverizing system by using an expert empirical formula containing parameters, and storing the parameters into a coal fines fineness parameter input data set u;
the parameter data preprocessing module judges and distinguishes redundant data in the pulverized coal fineness parameter input data set u
Or weakly correlated data, key parameters +.>
The pulverized coal fineness calculation module based on the T-S model is used for inputting a data set u and a pulverized coal fineness target value according to pulverized coal fineness parameters, training and obtaining a T-S fuzzy set model;
the data verification feedback module is used for carrying out verification feedback evaluation on the power generation efficiency and the boiler thermal efficiency by the coal fines;
and the KZ control module is used for collecting basic data acquired by the CK measurement and control module, and regulating and controlling the temperature and humidity of the coal bunker, the operation parameters of the coal mill and the operation parameters of the coal dust separator through the PLC.
Further, the numerical model database comprises a three-dimensional geometric model and a three-dimensional calculation model of an aerodynamic force field of a boiler of the power plant, which are constructed based on the combination of the historical operation data of the power plant and the set working condition, and a prediction model trained based on the historical simulation information of the combination of the historical operation data of the power plant and the set working condition.
Further, CK-01 is measured and a coal quality characteristic parameter is fed back to a KZ control module; CK-02 measures and gives out the structural characteristic and the operation characteristic parameter of the coal mill, and the structural characteristic and the operation characteristic parameter are fed back to the KZ control module, and the operation characteristic parameter of the coal mill is regulated according to a control signal of the KZ control module; CK-03 measures and gives out the structural characteristic of the separator and the operation characteristic parameter of the separator, and the structural characteristic and the operation characteristic parameter are fed back to the KZ control module, and the operation characteristic parameter of the separator is regulated according to a control signal of the KZ control module; CK-04 measures and gives out coal quality characteristic parameters and operation characteristic parameters thereof, and the coal quality characteristic parameters are fed back to the KZ control module; the expansion unit is ready for use, and the measured parameters include, but are not limited to, primary air quantity, boiler load, and coal volatile.
A pulverized coal fineness real-time on-line soft measurement method of a double model of a pulverizing system comprises the following steps:
step 1, establishing a numerical model database;
step 2, installing and constructing a CK measurement and control module in a power plant site, collecting pulverized coal fineness data of a sample, and determining a boundary condition of a set working condition;
step 3, inputting data to coal powder fineness data, namely coal powder fineness parameter input data, collecting data and extracting and converting the data through sample data acquired by the CK measurement and control module;
step 4, preprocessing and screening pulverized coal fineness parameter input data, and establishing a pulverized coal fineness calculation model based on a T-S model;
step 5, training a coal powder fineness calculation model based on a T-S model, acquiring real-time coal powder fineness data by a CK measurement and control module, judging whether a coal powder fineness result meets the accuracy requirement through a numerical model database, if not, entering a step 6, and if so, entering a step 7;
step 6, carrying out structure identification of a TS model, carrying out front part parameter identification through subtractive clustering, carrying out back part parameter identification through least square estimation, establishing a coal fines fineness TS prediction model, comparing coal fines fineness data of a coal bin and a coal bin through the prediction model, checking whether accuracy requirements are met, if so, entering step 7, and if not, recycling the step;
and 7, outputting pulverized coal fineness data meeting the precision requirement.
Further, in step 1, a numerical model database is established, including but not limited to, a three-dimensional geometric model of a certain boiler of the power plant and a three-dimensional calculation model of an aerodynamic force field are established based on the power plant historical operation data and the set working condition combination, and a prediction model is trained based on the power plant historical operation data and the set working condition combination historical simulation information; and the set working condition in the database model is the combination of the boundary condition and the combustion condition, and a corresponding set working condition combination table is obtained.
Further, in step 2, the built CK measurement and control module and the corresponding collected basic data include: the measurement and control sensor CK-01 comprises a coal quality analyzer, a DS5000 type coal quality component online measuring instrument, a Hardgkin grindability index measuring instrument and a programmable logic controller PLC-01, and is used for collecting volatile components V, fixed carbon C, ash content A, total moisture Mt and grindability coefficient HGI of coal to be ground; the measurement and control sensor CK-02 comprises an inductive material level gauge, a piezoelectric sensor, a thermocouple, an air quantity measuring instrument and a programmable logic controller PLC-02, and is used for collecting coal feeding quantity L, hydraulic loading force P, inlet air temperature temp. and inlet air quantity L of a coal mill for separating coal dust air The method comprises the steps of carrying out a first treatment on the surface of the Measurement and control sensorCK-03 including non-contact photoelectric sensor and programmable logic controller PLC-03 for collecting rotation speed RS of dynamic separator DynS Air quantity L of primary air wind The method comprises the steps of carrying out a first treatment on the surface of the The measurement and control sensor CK-04 comprises a non-contact photoelectric sensor and a programmable logic controller PLC-03, and is used for collecting the fineness R of pulverized coal r Real coal powder uniformity index n coal Real, pulverized coal concentration ω_real.
Further, in step 3, parameters are obtained according to the following empirical formula by measuring and controlling basic data collected by the sensors CK-01, CK-02 and CK-03
Extracting and converting, wherein m is more than or equal to 1, n is more than or equal to 1, and w is more than or equal to 1;
measurement and control sensor CK-01 pulverized coal fineness data set of bituminous coal:
measurement and control sensor CK-01 anthracite coal fines fineness data set:
measurement and control sensor CK-02 pulverized coal fineness data set:
measurement and control sensor CK-03 pulverized coal fineness data set:
constructing a pulverized coal fineness parameter input data set u:
wherein m, n, w are positive integers.
Further, in step 4, first, redundant data in the collected data of the measurement and control sensors CK-01, CK-02 and CK-03 are judged and distinguished
Or weakly correlated data, key parameters +.>
Then, the coal fines fineness parameter input data set u is subjected to sparsification treatment, and is stored in COO format and marked as u Spec The method comprises the steps of carrying out a first treatment on the surface of the Reading the physical index of the pulverized coal fineness dependent variable and storing the physical index in a matrix
The method comprises the steps of carrying out a first treatment on the surface of the And (3) completing the acquisition of an input variable sample of the pulverized coal fineness calculation model based on the T-S model.
Further, in step 5, pulverized coal fineness data of a pulverized coal system is obtained from a CK measurement and control module, and basic conditions of a set working condition are determined by combining a set working condition combination table; inquiring a numerical model database to obtain a coal powder fineness number, and comparing CK-04 measured data with a coal powder fineness empirical formula to judge whether the coal powder fineness data meets the accuracy requirement; if yes, outputting pulverized coal fineness data as a final value; if not, entering a T-S model to predict the fineness of the pulverized coal.
Further, in step 6, the step of predicting the fineness of the pulverized coal by the T-S model is as follows:
inputting pulverized coal fineness parameters into a data set u and a pulverized coal fineness dependent variable physical index matrix
Together form an initial input matrix->
/>
The density function of any point in the initial input matrix is
Let the cluster center be x c The cluster center density is +.>
If D is i+1 ＜TD i Removing the clustering center, and recalculating density values of other points in the field, wherein T is a threshold value set according to historical data;
each cluster center x c One dimension corresponding to a linear function:
identifying a back-piece parameter matrix heart q= [ q ] by using least square method 0 q 1 … q n ]Obtaining coal powder fineness value output of the T-S prediction model: />
Wherein E= [ D 0 D 1 … D n ]And (5) completing the coal powder fineness T-S prediction model.
Compared with the prior art, the invention has the advantages that:
(1) Technical advantages are that: the T-S model can be combined with expert experience to establish fuzzy rules, has certain physical significance, and can reflect the physical relationship between the input quantity of the parameter model and the fineness of the pulverized coal rather than simple numerical fitting.
(2) Algorithm advantage: the pulverized coal fineness online coupling soft measurement method combined with the T-S fuzzy neural network parameter model optimizes the parameters of the front part, and the obtained fuzzy neural network has simple structure and clear physical meaning.
(3) Economic benefit: the historical operation data of the depth reference equipment are fused with two pulverized coal fineness prediction methods, and the robustness of the method is guaranteed.
Drawings
FIG. 1 is a schematic diagram of a real-time on-line soft measurement system for pulverized coal fineness in an embodiment of the invention.
Fig. 2 is a schematic diagram of a real-time on-line soft measurement method for fineness of pulverized coal according to an embodiment of the present invention.
FIG. 3 is a schematic diagram of the installation position of the CK measurement and control module in an embodiment of the invention.
Detailed Description
The technical scheme of the invention is further described in detail below with reference to the attached drawings.
The pulverized coal fineness real-time on-line soft measurement system of the double models of the pulverizing system comprises the following 7 components: (1) a model database. The method comprises the steps of constructing a three-dimensional geometric model of a boiler of the power plant and a three-dimensional calculation model of an aerodynamic field based on power plant historical operation data and set working condition combinations, and constructing a prediction model trained based on historical simulation information of the power plant historical operation data and set working condition combinations. (2) The CK measurement and control module (comprising CK-01, CK-02, CK-03 and an expansion unit thereof) is installed on site in a coal bunker feeder, a coal mill and a coal dust separator of a power plant, and measures and gives out coal quality characteristics, coal mill structural characteristics and operation characteristics, separator structure and operation characteristics and other coal manufacturing system characteristic parameters, wherein the CK-01 measures and gives out the coal quality characteristic parameters to be fed back to the control module; CK-02 measures and gives structural characteristics of the coal mill and operating characteristic parameters thereof, and the structural characteristics and the operating characteristic parameters are fed back to the control module, and the operating characteristic parameters of the coal mill are adjusted according to control signals of the control module; CK-03 measures and gives out the structural characteristic of the separator and the operation characteristic parameter feedback to the control module, and the operation characteristic parameter of the separator is regulated according to the control signal of the control module; the expansion unit is ready for use, including but not limited to primary air volume, boiler load, coal volatiles, etc. A schematic of this module is shown in fig. 2. (3) The data extraction and conversion module expresses the coal quality characteristics, coal mill structural characteristics and operation characteristics thereof, separator structure and operation characteristics thereof and other pulverizing system characteristic parameters by using expert empirical formulas containing parameters and for each parameter
A detailed flow of the pulverized coal fineness real-time on-line soft measurement method of the double models of the pulverizing system is shown in fig. 1 and 2, wherein a model database, a data extraction and conversion module, a data preprocessing module, a pulverized coal fineness calculation module based on a T-S model and a verification feedback module jointly form a control calculation core of the method, and the control module and a measurement and control module jointly form a hardware system platform of the method. A hardware system platform of a pulverized coal fineness online coupling soft measurement method comprises the following measuring equipment:
a measurement and control sensor CK-01 (a coal quality analyzer, a DS5000 type coal quality component online measuring instrument, a Hardgkin grindability index determinator and a distributed programmable logic controller PLC) is arranged on a coal conveying pipeline of a coal bunker coal feeder, and full moisture Mt and grindability coefficient HGI of coal to be ground are collected;
the coal feeding amount L, the hydraulic loading force P, the inlet air temperature temp. Temp and the inlet air quantity L of the coal mill for separating the coal dust are collected by installing a measurement and control sensor CK-02 (an inductance type level gauge, a piezoelectric sensor, a thermocouple, an air quantity measuring instrument and a distributed programmable logic controller PLC) on the coal conveying pipeline of the coal mill air ；
The pulverized coal separator is provided with a measurement and control sensor CK-03 (a non-contact photoelectric sensor and a distributed programmable logic controller PLC) for collecting the rotating speed RS of the dynamic separator DynS 。
The control device comprises: distributed programmable logic controller PLC.
The on-line coupling soft measurement method for the fineness of the pulverized coal comprises the following steps: (1) establishing a model database; (2) The method comprises the steps of installing and building a CK measurement and control module and collecting basic data in a power plant site; (3) Extracting and converting pulverized coal fineness parameter input data set data; (4) preprocessing and screening parameter data; (5) Training of a coal fines fineness calculation model based on a T-S model.
(1) Establishing a model database:
a pulverized coal fineness real-time online soft measurement method of a double model of a pulverizing system is applied to a certain power plant boiler and is used for carrying out numerical simulation analysis aiming at a set working condition of the plant to establish a model database, and comprises the steps of constructing a three-dimensional geometric model and a three-dimensional calculation model of an aerodynamic force field of the certain power plant boiler based on the combination of historical operation data of the power plant and the set working condition, and constructing a prediction model trained based on historical simulation information of the combination of the historical operation data of the power plant and the set working condition. The given operating conditions in the database model are combinations of boundary conditions and combustion conditions, but include, but are not limited to, the following tables.
(2) And (3) installing and building a CK measurement and control module and collecting basic data in the power plant site:
the measurement and control sensor CK-01 (coal quality analyzer, DS5000 type coal quality component on-line measuring instrument, hardgkin grindability index measuring instrument and distributed programmable logic controller PLC) collects volatile components V, fixed carbon C, ash content A, total moisture Mt and grindability coefficient HGI of coal to be ground.
Measurement and control sensor CK-02 (inductance type level gauge, piezoelectric sensor, thermocouple, air quantity measuring instrument and distributed programmable logic controller PLC) is used for collecting coal feeding quantity L, hydraulic loading force P, inlet air temperature temp. and inlet air quantity L of coal mill for separating coal dust air 。
Measurement and control sensor CK-03 (non-contact photoelectric sensor and distributed programmable logic controller PLC) for collecting rotation speed RS of dynamic separator DynS Air quantity L of primary air wind 。
Measurement and control sensor CK-04 (non-contact photoelectric sensor and distributed programmable logic controller PLC) for collecting pulverized coal fineness R r Real coal powder uniformity index n coal Real, pulverized coal concentration ω_real.
(3) Extracting and converting pulverized coal fineness parameter input data set data:
measurement and control sensor CK-01 pulverized coal fineness data set of bituminous coal:
measurement and control sensor CK-01 anthracite coal fines fineness data set:
measurement and control sensor CK-02 pulverized coal fineness data set:
measurement and control sensor CK-03 pulverized coal fineness data set:
constructing a pulverized coal fineness parameter input data set u:
wherein m, n, w are positive integers. />
(4) Parameter data preprocessing and screening:
the acquired data of the measurement and control sensor is influenced by environmental conditions, and the acquired data of the measurement and control sensor can be subjected to conditions of overrun, distortion, frame loss and the like, so that the acquired data of the measurement and control sensor is screened, and redundant data in the acquired data of the measurement and control sensors (CK-01, CK-02 and CK-03) are judged and distinguished
(or weakly correlated data) and key parameters +.>
The coal fines fineness parameter input data set u is subjected to sparsification treatment and is stored in COO (Coordinate Format) format, and is recorded as u Spec . Reading the physical index of the pulverized coal fineness dependent variable and storing the physical index in a matrix
And (3) completing the acquisition of an input variable sample of the pulverized coal fineness calculation model based on the T-S model.
And collecting input and output variable samples of a pulverized coal fineness calculation model based on a T-S model.
(5) Training of a coal fines fineness calculation model based on a T-S model:
basic data of the pulverized coal system are obtained from CK measurement and control modules (CK-01, CK-02 and CK-03), and basic conditions of the set working conditions are determined by combining a combination table of the set working conditions.
And inquiring a numerical model database to obtain a coal powder fineness value, and comparing the measured value of CK-04 with the CK-01 coal powder fineness data set to judge whether the coal powder fineness value meets the precision requirement. If yes, outputting the fineness value of the pulverized coal as a final value; if not, entering a T-S model to predict the fineness of the pulverized coal.
Coal fines fineness parameter input data set u and coal fines fineness dependent variable physical index matrix
Together form an initial input matrix->
For any point, the density function is
The clustering center is x c The density of the cluster centers is as follows
If D is i+1 ＜TD i (T is a threshold value set according to historical data, the threshold value range is set to be 0.8-0.9), the clustering center is removed, and the density values of other points in the field are recalculated.
Each cluster center x c One dimension corresponding to a linear function:
identifying a back-piece parameter matrix heart q= [ q ] by using least square method 0 q 1 … q n ]Obtaining the coal powder fineness value output of the T-S fuzzy model:
wherein E= [ D 0 D 1 … D n ]. And (5) completing the coal powder fineness T-S prediction model.
And (3) comparing the measured value of CK-04 with the formulas (1) and (2) to judge whether the coal powder fineness value meets the precision requirement. If yes, outputting the fineness value of the pulverized coal as a final value; if not, re-entering a step of predicting the fineness of the pulverized coal by the T-S model. The pulverized coal fineness calculation model based on the T-S model is an adaptive algorithm model to be trained, and for each given process flow and equipment, equipment parameters (i.e. training process figure 2) are continuously adjusted by referring to the optimal pulverized coal fineness required by the boiler so as to produce the optimal pulverized coal fineness required by the boiler. Wherein, the optimal pulverized coal fineness required by the boiler is a fixed value or range given by an expert system according to the working condition required by the power plant; the fineness value of the pulverized coal in the pulverized coal bin tested by CK-04 is used as the process quantity in the training process, and the fineness value is finally approximate to a fixed value or range given by an expert system according to the working condition required by a power plant through the adjustment training.
The above description is merely of preferred embodiments of the present invention, and the scope of the present invention is not limited to the above embodiments, but all equivalent modifications or variations according to the present disclosure will be within the scope of the claims.
Claims (10)
1. The utility model provides a pulverized coal fineness real-time on-line soft measurement system of two models of powder process system, includes numerical model database, CK measurement and control module, data extraction and conversion module, parameter data preprocessing module, pulverized coal fineness calculation module, data verification feedback module, KZ control module based on the T-S model, its characterized in that:
the coal powder fineness calculation module based on the T-S model is connected with the numerical model database;
the data extraction and conversion module is connected with the CK measurement and control module, the data verification feedback module is connected with the KZ control module, and the CK measurement and control module and the KZ control module are connected with each other;
the CK measurement and control module comprises measurement and control sensors CK-01, CK-02, CK-03, CK-04 and expansion units thereof, and is arranged on a coal bunker feeder, a coal mill, a coal dust separator and a coal dust bunker to measure and give out characteristic parameters of the coal pulverizing system;
the data extraction and conversion module is used for expressing and extracting and converting parameters of the characteristic parameters of the coal pulverizing system by using an expert empirical formula containing parameters, and storing the parameters into a coal fines fineness parameter input data set u;
the parameter data preprocessing module judges and distinguishes redundant data in the pulverized coal fineness parameter input data set u
Or weakly correlated data, key parameters +.>
The pulverized coal fineness calculation module based on the T-S model is used for inputting a data set u and a pulverized coal fineness target value according to pulverized coal fineness parameters, training and obtaining a T-S fuzzy set model;
the data verification feedback module is used for carrying out verification feedback evaluation on the power generation efficiency and the boiler thermal efficiency by the coal fines;
and the KZ control module is used for collecting basic data acquired by the CK measurement and control module, and regulating and controlling the temperature and humidity of the coal bunker, the operation parameters of the coal mill and the operation parameters of the coal dust separator through the PLC.
2. The pulverized coal fineness real-time online soft measurement system of a double model of a pulverizing system according to claim 1, wherein the pulverized coal fineness real-time online soft measurement system is characterized in that: the numerical model database comprises a three-dimensional geometric model and a three-dimensional calculation model of an aerodynamic field of a boiler of a power plant constructed based on the historical operation data of the power plant and the set working condition combination, and a prediction model trained based on the historical simulation information of the historical operation data of the power plant and the set working condition combination.
3. The pulverized coal fineness real-time online soft measurement system of a double model of a pulverizing system according to claim 1, wherein the pulverized coal fineness real-time online soft measurement system is characterized in that: CK-01 is measured and gives a coal quality characteristic parameter which is fed back to a KZ control module; CK-02 measures and gives out the structural characteristic and the operation characteristic parameter of the coal mill, and the structural characteristic and the operation characteristic parameter are fed back to the KZ control module, and the operation characteristic parameter of the coal mill is regulated according to a control signal of the KZ control module; CK-03 measures and gives out the structural characteristic of the separator and the operation characteristic parameter of the separator, and the structural characteristic and the operation characteristic parameter are fed back to the KZ control module, and the operation characteristic parameter of the separator is regulated according to a control signal of the KZ control module; CK-04 measures and gives out coal quality characteristic parameters and operation characteristic parameters thereof, and the coal quality characteristic parameters are fed back to the KZ control module; the expansion unit is ready for use, and the measured parameters include, but are not limited to, primary air quantity, boiler load, and coal volatile.
4. A real-time on-line soft measurement method for pulverized coal fineness of a double model of a pulverizing system is characterized by comprising the following steps of: the method comprises the following steps:
step 1, establishing a numerical model database;
step 2, installing and constructing a CK measurement and control module in a power plant site, collecting pulverized coal fineness data of a sample, and determining boundary conditions of a set working condition;
step 3, inputting data to coal powder fineness data, namely coal powder fineness parameter input data, collecting data and extracting and converting the data through sample data acquired by the CK measurement and control module;
step 4, preprocessing and screening pulverized coal fineness parameter input data, and establishing a pulverized coal fineness calculation model based on a T-S model;
step 5, training a coal powder fineness calculation model based on a T-S model, acquiring real-time coal powder fineness data by a CK measurement and control module, judging whether a coal powder fineness result meets the accuracy requirement through a numerical model database, if not, entering a step 6, and if so, entering a step 7;
step 6, carrying out structure identification of a TS model, carrying out front part parameter identification through subtractive clustering, carrying out back part parameter identification through least square estimation, establishing a coal fines fineness TS prediction model, comparing coal fines fineness data of a coal bin and a coal bin through the prediction model, checking whether accuracy requirements are met, if so, entering step 7, and if not, recycling the step;
and 7, outputting pulverized coal fineness data meeting the precision requirement.
5. The real-time on-line soft measurement method for pulverized coal fineness of double models of a pulverizing system according to claim 4, wherein the method comprises the following steps: in the step 1, a numerical model database is established, including but not limited to a three-dimensional geometric model of a boiler of a power plant and a three-dimensional calculation model of an aerodynamic force field are established based on the historical operation data of the power plant and the combination of the set working conditions, and a prediction model trained based on the historical simulation information of the combination of the historical operation data of the power plant and the set working conditions is established; and the set working condition in the database model is the combination of the boundary condition and the combustion condition, and a corresponding set working condition combination table is obtained.
6. The real-time on-line soft measurement method for pulverized coal fineness of double models of a pulverizing system according to claim 4, wherein the method comprises the following steps: in step 2, the CK measurement and control module and the corresponding collected basic data are constructed by: the measurement and control sensor CK-01 comprises a coal quality analyzer, a DS5000 type coal quality component online measuring instrument, a Hardgkin grindability index measuring instrument and a programmable logic controller PLC-01, and is used for collecting volatile components V, fixed carbon C, ash content A, total moisture Mt and grindability coefficient HGI of coal to be ground; the measurement and control sensor CK-02 comprises an inductive material level gauge, a piezoelectric sensor, a thermocouple, an air quantity measuring instrument and a programmable logic controller PLC-02, and is used for collecting coal feeding quantity L, hydraulic loading force P, inlet air temperature temp. and inlet air quantity L of a coal mill for separating coal dust air The method comprises the steps of carrying out a first treatment on the surface of the The measurement and control sensor CK-03 comprises a non-contact photoelectric sensor and a programmable logic controller PLC-03, and is used for collecting the rotating speed RS of the dynamic separator DynS Air quantity L of primary air wind The method comprises the steps of carrying out a first treatment on the surface of the The measurement and control sensor CK-04 comprises a non-contact photoelectric sensor and a programmable logic controller PLC-03, and is used for collecting the fineness R of pulverized coal r Real, coal fines uniformity index n coal Real, pulverized coal concentration ω_real.
7. The pulverized coal fineness real-time online soft measurement of a double model of a pulverizing system according to claim 6The measuring method is characterized in that: in step 3, parameters are obtained according to the following empirical formula through measuring and controlling basic data acquired by the sensors CK-01, CK-02 and CK-03
Extracting and converting, wherein m is more than or equal to 1, n is more than or equal to 1, and w is more than or equal to 1;
measurement and control sensor CK-01 pulverized coal fineness data set of bituminous coal:
measurement and control sensor CK-01 anthracite coal fines fineness data set:
measurement and control sensor CK-02 pulverized coal fineness data set:
measurement and control sensor CK-03 pulverized coal fineness data set:
constructing a pulverized coal fineness parameter input data set u:
wherein m, n, w are positive integers. />
8. A dual mode milling system as claimed in claim 4The real-time on-line soft measurement method for the fineness of the pulverized coal is characterized by comprising the following steps of: in step 4, first, redundant data in collected data of measurement and control sensors CK-01, CK-02 and CK-03 are judged and distinguished
Or weakly correlated data, key parameters +.>
Then, the coal fines fineness parameter input data set u is subjected to sparsification treatment, and is stored in COO format and marked as u Spec The method comprises the steps of carrying out a first treatment on the surface of the Reading the physical index of the pulverized coal fineness dependent variable and storing the physical index in a matrix +.>
And (3) completing the acquisition of an input variable sample of the pulverized coal fineness calculation model based on the T-S model.
9. The real-time on-line soft measurement method for pulverized coal fineness of double models of a pulverizing system according to claim 4, wherein the method comprises the following steps: in step 5, coal powder fineness data of a coal powder system are obtained from the CK measurement and control module, and basic conditions of set working conditions are determined by combining a set working condition combination table; inquiring a numerical model database to obtain a coal powder fineness number, and comparing CK-04 measured data with a coal powder fineness empirical formula to judge whether the coal powder fineness data meets the accuracy requirement; if yes, outputting pulverized coal fineness data as a final value; if not, entering a T-S model to predict the fineness of the pulverized coal.
10. The real-time on-line soft measurement method for pulverized coal fineness of double models of a pulverizing system according to claim 4, wherein the method comprises the following steps: in the step 6, the step of predicting the fineness of the pulverized coal by the T-S model is as follows:
inputting pulverized coal fineness parameters into a data set u and a pulverized coal fineness dependent variable physical index matrix
Together form an initial input matrix->
The density function of any point in the initial input matrix is
Let the cluster center be x c The cluster center density is +.>
If D is i+1 <TD i Removing the clustering center, and recalculating density values of other points in the field, wherein T is a threshold value set according to historical data;
each cluster center x c One dimension corresponding to a linear function:
identifying a back-piece parameter matrix q= [ q ] by using a least square method 0 q 1 … q n ]Obtaining coal powder fineness value output of the T-S prediction model: />
Wherein E= [ D 0 D 1 … D n ]And (5) completing the coal powder fineness T-S prediction model. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011202911.4A CN112163350B (en) | 2020-11-02 | 2020-11-02 | Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011202911.4A CN112163350B (en) | 2020-11-02 | 2020-11-02 | Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112163350A CN112163350A (en) | 2021-01-01 |
| CN112163350B true CN112163350B (en) | 2023-05-23 |
Family
ID=73865447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011202911.4A Active CN112163350B (en) | 2020-11-02 | 2020-11-02 | Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112163350B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113743672B (en) * | 2021-09-09 | 2024-04-05 | 西安热工研究院有限公司 | Online analysis and operation optimization guiding method for economic coal fines fineness of coal mill |
| CN113792255B (en) * | 2021-11-17 | 2022-02-22 | 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 |
| CN114733640B (en) * | 2022-03-03 | 2023-09-12 | 江苏丰尚智能科技有限公司 | Method and device for adjusting processing parameters of pulverizer and computer equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105640925A (en) * | 2010-08-30 | 2016-06-08 | 普马特里克斯营业公司 | Dry powder formula and method thereof for treating lung diseases |
| CN109174423A (en) * | 2018-09-18 | 2019-01-11 | 国电龙源节能技术有限公司 | Pulverizer outlet temperature control system and method based on fineness of pulverized coal on-line tuning |
| 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 |
| CN110210100A (en) * | 2019-05-27 | 2019-09-06 | 中国科学院南海海洋研究所 | A kind of high-precision sediment deposit velocity of sound prediction technique |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11327003B2 (en) * | 2018-08-30 | 2022-05-10 | Becton, Dickinson And Company | Characterization and sorting for particle analyzers |
-
2020
- 2020-11-02 CN CN202011202911.4A patent/CN112163350B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105640925A (en) * | 2010-08-30 | 2016-06-08 | 普马特里克斯营业公司 | Dry powder formula and method thereof for treating lung diseases |
| 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 |
| CN109174423A (en) * | 2018-09-18 | 2019-01-11 | 国电龙源节能技术有限公司 | Pulverizer outlet temperature control system and method based on fineness of pulverized coal on-line tuning |
| CN110210100A (en) * | 2019-05-27 | 2019-09-06 | 中国科学院南海海洋研究所 | A kind of high-precision sediment deposit velocity of sound prediction technique |
Non-Patent Citations (5)
| Title |
|---|
| "Advanced diagnostics of industrial pulverized coal burner using optical methods and artificial intelligence";Andrzej Smolarz等;《Experimental Thermal and Fluid Science》;20120409;第43卷;第82–89页 * |
| 基于最小二乘-支持向量机的制粉过程煤粉细度软测量模型;张立岩等;《清华大学学报(自然科学版)》;20071015;第47卷;第1932-1935页 * |
| 基于磨出口温度定值优化的制粉系统多变量预测控制;高耀岿等;《动力工程学报》;20171015(第10期);第47-53页 * |
| 换热器污垢特性的建模与预测;文孝强等;《化工机械》;20141215;第41卷(第06期);第699-704页 * |
| 联合进汽阀内部流场三维数值模拟及阀碟受力分析;王红娟等;《西安交通大学学报》;20161231;第50卷(第12期);第148-154页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112163350A (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112163350B (en) | Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system | |
| CN101038277B (en) | Soft measurement method for coal power fineness in powdering producer | |
| CN102654286B (en) | Intelligent dynamic combustion atmosphere controller | |
| CN100538316C (en) | A kind of thermal power plant ball mill load soft-sensing method | |
| CN103336107B (en) | Soft measurement method for f-CaO content of cement clinker | |
| CN105268536B (en) | A kind of thermal power plant ball mill load soft-sensing method of adaptive on-line correction | |
| CN107133460A (en) | A kind of online dynamic prediction method of boiler flyash carbon content | |
| CN102778538A (en) | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash | |
| CN112001527A (en) | Industrial production process target data prediction method of multi-feature fusion deep neural network | |
| Wang et al. | A study on a new algorithm to optimize ball mill system based on modeling and GA | |
| CN111754093A (en) | Fly ash carbon content prediction method based on coal quality prediction and PSO-SVM | |
| CN104750131A (en) | Fluidized bed temperature control method based on parameter identification | |
| CN111024921B (en) | System and method for monitoring coal quality in furnace | |
| CN106563537A (en) | Mill load detection method based on vibration signals of throwing-down area and sliding area of surface of barrel | |
| CN104331736A (en) | RBF (Radial Basis Function) neural network based supercritical boiler nitric oxide discharging dynamic predication method | |
| CN102226772B (en) | Device for measuring carbon content of fly ash in boiler of electric power plant | |
| CN108197723B (en) | Optimized energy-saving scheduling method for coal consumption and pollutant discharge of coal-electricity unit power supply | |
| CN104536396A (en) | Soft measurement modeling method used in cement raw material decomposing process in decomposing furnace | |
| CN116578832A (en) | Device for predicting surface roughness of machined part | |
| CN113533622A (en) | Coal quality prediction method for coal mill based on neural network | |
| CN111024920B (en) | Real-time on-line monitoring system and method for coal quality in furnace | |
| CN110954536B (en) | Online measuring device and method for carbon content of fly ash | |
| Yan et al. | Soft sensor for ball mill fill level based on uncertainty reasoning of cloud model | |
| CN116399400B (en) | Method and system for detecting operation faults of boiler unit | |
| CN108709426B (en) | Sintering machine air leakage fault online diagnosis method based on frequency spectrum characteristic bilateral detection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |