CN113358811A - Soft measurement method and device for fly ash carbon content based on firing method calibration - Google Patents
Soft measurement method and device for fly ash carbon content based on firing method calibration Download PDFInfo
- Publication number
- CN113358811A CN113358811A CN202110576286.8A CN202110576286A CN113358811A CN 113358811 A CN113358811 A CN 113358811A CN 202110576286 A CN202110576286 A CN 202110576286A CN 113358811 A CN113358811 A CN 113358811A
- Authority
- CN
- China
- Prior art keywords
- soft measurement
- fly ash
- carbon content
- soft
- prediction 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N31/00—Investigating or analysing non-biological materials by the use of the chemical methods specified in the subgroup; Apparatus specially adapted for such methods
- G01N31/12—Investigating or analysing non-biological materials by the use of the chemical methods specified in the subgroup; Apparatus specially adapted for such methods using combustion
Landscapes
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Combustion & Propulsion (AREA)
- Molecular Biology (AREA)
- Engineering & Computer Science (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Non-Biological Materials By The Use Of Chemical Means (AREA)
Abstract
The invention relates to a soft measurement method and a soft measurement device for fly ash carbon content based on ignition method calibration, wherein the method comprises the following steps: periodically sampling fly ash and soft measurement parameters corresponding to sampling time; measuring the carbon content of the fly ash by adopting a burning method according to the fly ash obtained by sampling, and taking the carbon content and corresponding soft measurement parameters as a group of training data; training and updating the soft measurement prediction model by using the last N groups of training data; and collecting soft measurement parameters, and inputting the soft measurement parameters into a soft measurement prediction model to obtain the carbon content of the fly ash. Compared with the prior art, the accuracy of the fly ash carbon content measured by the burning method is higher, the output data of the burning method is used as the basic sample data of soft measurement, a soft measurement prediction model is obtained by training, and the accuracy of the calculation result is high.
Description
Technical Field
The invention relates to a soft measurement method for carbon content in fly ash, in particular to a soft measurement method and a soft measurement device for carbon content in fly ash based on ignition method calibration.
Background
In order to match with the national energy development strategy, the energy conservation and emission reduction tasks of the thermal power generating unit are urgent, the power generation efficiency is improved, and the power generation cost is reduced, so that the thermal power generating unit becomes one of the important tasks of the thermal power plant. In coal fired boilers, the mechanical incomplete combustion heat loss is the second largest heat loss next to the heat loss of flue gas. For a boiler, the heat loss of mechanical incomplete combustion can be reduced, so that the heat efficiency of the boiler can be effectively improved, and for a power plant, the carbon content of fly ash in the mechanical incomplete combustion is an important parameter index for the operation of a conductive plant. However, the value of the carbon content of the fly ash is difficult to realize on-line real-time measurement and display, and cannot provide effective guidance for power plant operators, so that the operation and control level of the thermal power generating unit is restricted to a certain extent.
The existing measuring method for the carbon content of the fly ash of the coal-fired boiler mainly comprises an off-line method and an on-line method, wherein the off-line method is to burn a certain amount of fly ash samples in a high-temperature state after proper sampling, and the carbon content of the fly ash is obtained through the weight difference before and after burning, so that the measuring result is high in accuracy and has strong coal adaptability, but the defect is that the obtained result has inevitable hysteresis and cannot reflect a real-time burning state. The online method is divided into a direct measurement method and a soft measurement method, and the direct measurement method has certain defects, namely, the sample tube is easy to block dust, so that the measurement accuracy is influenced; secondly, the cost of the modified equipment is higher, and the additional equipment is complex. The soft measurement uses the relationship between auxiliary variables which are easy to measure and the carbon content of the fly ash, and the carbon content of the fly ash is estimated through mathematical means analysis, but the defects are that the performance of the boiler is changed when the boiler runs for a long time, if sample data is not updated for a long time, the nonlinear relationship between each auxiliary parameter and the carbon content of the fly ash is also changed, and the output result of the soft measurement is inaccurate.
Disclosure of Invention
The invention aims to provide a method and a device for soft measurement of carbon content in fly ash based on ignition calibration, which overcome and overcome the defects of the existing online measurement system of carbon content in fly ash and provide a system for soft measurement of carbon content in fly ash based on ignition calibration. The method continuously and periodically obtains the sample data of the carbon content of the fly ash through an on-line ignition measurement system; training a soft measurement model by using periodically updated sample data and adopting a correlation algorithm, and improving the accuracy of an output result of soft measurement; the soft measurement method is adopted, so that the hysteresis and discontinuity of a burning method measurement system are compensated, and the online measurement of the carbon content of the fly ash is realized; the soft measurement method which adopts the firing method to output data as a sample also has stronger adaptability to variable working conditions by combining the firing method with strong adaptability to coal types.
The purpose of the invention can be realized by the following technical scheme:
a soft measurement method for fly ash carbon content based on ignition method calibration comprises the following steps:
periodically sampling fly ash and soft measurement parameters corresponding to sampling time;
measuring the carbon content of the fly ash by adopting a burning method according to the fly ash obtained by sampling, and taking the carbon content and corresponding soft measurement parameters as a group of training data;
training and updating the soft measurement prediction model by using the last N groups of training data;
and collecting soft measurement parameters, and inputting the soft measurement parameters into a soft measurement prediction model to obtain the carbon content of the fly ash.
The soft measurement parameters comprise actual load of the boiler, output of the coal feeder, received base volatile components and low-level heating value.
And the last N groups of training data are used for training and updating N values in the soft measurement prediction model to be 40-60.
And the last N groups of training data are used for training and updating the N value in the soft measurement prediction model to be 50.
The soft measurement prediction model is a support vector machine model.
A soft measuring device for fly ash carbon content based on ignition method calibration comprises a processor, a memory and a program, wherein the processor implements the method when executing the program.
Compared with the prior art, the invention has the following beneficial effects:
1) the accuracy of the fly ash carbon content measured by the burning method is high, the output data of the burning method is used as basic sample data of soft measurement, a soft measurement prediction model is obtained by training, and the accuracy of the calculation result is high.
2) The burning measuring device periodically operates and updates basic sample data of soft measurement, so that the nonlinear relation between the auxiliary variable searched by the soft measurement model and the carbon content of the fly ash has real-time performance, and the burning state of the boiler can be better reflected.
3) The online measurement of the carbon content of the fly ash is realized, and the output result of the carbon content of the fly ash has real-time performance and continuity.
4) The final output result of the measuring system has stronger coal type changing adaptability.
Drawings
FIG. 1 is a schematic structural view of the present invention;
FIG. 2 is a schematic structural diagram of an embodiment of the present invention;
FIG. 3 is a flow chart of the method for soft measurement of carbon content in fly ash according to the present invention;
fig. 4 is a flow chart of the soft measurement system arrangement operation of the present invention.
Wherein: 1. the device comprises a boiler flue, 2, a fly ash sampling device, 3, a burning measuring device, 4, a fly ash discharging device, 5, a signal processing and controlling device, 6, a soft measurement calculating unit, 7, a database, 8, a DCS system, 21, a fly ash sampling pipe, 22, a compressed air pipe, 23, a flue gas separating unit, 24, a descending pipe, 25, a weighing container, 26 and a mechanical device.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Referring to fig. 1 and 2, the measurement system of the present invention is divided into two parts, one part is used for obtaining sample data by a burning method, and the other part is used for soft measurement of carbon content in fly ash.
In the process of obtaining sample data by a burning method, ash-containing flue gas is obtained from a boiler flue 1 through a fly ash sampling tube 21, fly ash is separated from the flue gas in a flue gas separation unit 23, the fly ash uniformly falls into a weighing container 25 in a descending tube 24, the weighing container 25 is conveyed into a burning measurement device 3 through a certain mechanical device 26, residual fly ash after measurement is discharged through a fly ash discharge device 4, the whole process is controlled by a signal processing and controller 5, so that the burning measurement system periodically operates, and measured parameters are sent into a database 7.
In the soft measurement process of the carbon content of the fly ash, firstly, the DCS raw data is appropriately preprocessed, and then the number of soft measurement variables is determined by using a relevant calculation method, such as a variable selection method of a multiple linear regression, a partial least square method, a compression coefficient method and the like. And modeling the existing data sample by using a related intelligent algorithm, optimizing key parameters in the modeling by using the related intelligent algorithm to finally obtain the nonlinear relation between the main auxiliary variable and the carbon content of the fly ash, and calculating to obtain the current carbon content of the fly ash according to the data of the main auxiliary variable acquired by the current DCS.
When the system integrally operates, the number of soft measurement training samples, the measurement period of soft measurement and the measurement period of burning measurement are given firstly. The burning measuring device 3 works first, the carbon content of the fly ash is recorded every time the carbon content of the fly ash is measured, and the carbon content of the fly ash is synchronized with the auxiliary variable data collected by the DCS at the current moment and is completely recorded into the database until the number of samples in the database meets the number of training samples for soft measurement. The soft measurement module starts to operate and records the data to the database, judges whether the burning measurement device obtains the latest data, replaces the old data with the longest retention time, ensures the stability and the real-time performance of the number of soft measurement training samples, and can continuously and stably output the value of the carbon content of the fly ash according to the soft measurement.
Specifically, as shown in fig. 3, a soft measurement method for fly ash carbon content based on calibration by a burning method includes:
periodically sampling fly ash and soft measurement parameters corresponding to sampling time;
measuring the carbon content of the fly ash by adopting a burning method according to the fly ash obtained by sampling, and taking the carbon content and corresponding soft measurement parameters as a group of training data;
training and updating the soft measurement prediction model by using the last N groups of training data;
and collecting soft measurement parameters, and inputting the soft measurement parameters into a soft measurement prediction model to obtain the carbon content of the fly ash.
In some embodiments, the soft measurement parameters include boiler actual load, feeder output, received base volatiles, and low calorific value.
In some embodiments, the last N sets of training data are used for training to update the value of N in the soft measurement prediction model to 40-60. In one embodiment, the last N sets of training data are used to train and update the soft-measurement prediction model to have an N value of 50.
The soft measurement prediction model is a support vector machine model or a prediction algorithm such as a neural network, and is used for explaining the basic operation principle of the technical idea of the invention by taking a partial least square method, a support vector machine and a genetic algorithm as examples, and part of data adopted in calculation is shown in table 1.
TABLE 1
When the system starts to operate, the burning method periodically generates the data of measuring the carbon content of the fly ash until the data amount reaches a certain value, the set value of the data amount is assumed to be 50, after the 50 th group of data is generated, 1-50 groups of data are jointly used as model samples of soft measurement of the carbon content of the fly ash, and modeling is carried out by utilizing the 1-50 groups of data. Data were screened using pre-processing and partial least squares methods. The search range of the genetic algorithm for the penalty factor c of the support vector machine parameter is 0.01-100, the search range for the insensitivity epsilon is 0.001-10, and the search range for the radial basis kernel parameter sigma is 0.01-100.
Through calculation, the optimal penalty factor c, the insensitivity epsilon and the radial basis kernel parameter sigma obtained by 1-50 groups of optimized sample data are respectively as follows: 9.544, 0.2886, and 0.3944. And performing soft measurement calculation of the carbon content of the fly ash by using the three coefficients. Basic input data are obtained from the DCS system, and the result of calculating the carbon content of the fly ash is shown in Table 2:
TABLE 2
The soft measurement system periodically outputs data until the burning method measurement system generates new data to be recorded into the database, at the moment, 1-51 groups of data exist in the database, the oldest group of data is removed, namely 2-51 groups of data are used for carrying out a new modeling. The modeling and calculation processes are the same as above, and the operation is continuously performed periodically. The soft measurement database is updated in real time by the burning method measurement, and the soft measurement system continuously and rapidly outputs the numerical value of the carbon content in the fly ash.
Claims (10)
1. A soft measurement method for fly ash carbon content based on ignition method calibration is characterized by comprising the following steps:
periodically sampling fly ash and soft measurement parameters corresponding to sampling time;
measuring the carbon content of the fly ash by adopting a burning method according to the fly ash obtained by sampling, and taking the carbon content and corresponding soft measurement parameters as a group of training data;
training and updating the soft measurement prediction model by using the last N groups of training data;
and collecting soft measurement parameters, and inputting the soft measurement parameters into a soft measurement prediction model to obtain the carbon content of the fly ash.
2. The soft measurement method for the carbon content of the fly ash calibrated based on the burning method as claimed in claim 1, wherein the soft measurement parameters comprise actual load of a boiler, output of a coal feeder, received base volatile matter and low calorific value.
3. The soft measurement method for fly ash carbon content based on ignition calibration according to claim 1, wherein the value of N in the soft measurement prediction model is updated to 40-60 by training with the last N sets of training data.
4. The soft measurement method for fly ash carbon content based on ignition calibration according to claim 3, wherein the value of N in the soft measurement prediction model is updated to 50 by training with the last N sets of training data.
5. The soft measurement method for carbon content in fly ash calibrated based on the burning method as claimed in claim 1, wherein the soft measurement prediction model is a support vector machine model.
6. A soft measuring device for fly ash carbon content based on ignition method calibration comprises a processor, a memory and a program, and is characterized in that the processor implements the following steps when executing the program:
periodically sampling fly ash and soft measurement parameters corresponding to sampling time;
measuring the carbon content of the fly ash by adopting a burning method according to the fly ash obtained by sampling, and taking the carbon content and corresponding soft measurement parameters as a group of training data;
training and updating the soft measurement prediction model by using the last N groups of training data;
and collecting soft measurement parameters, and inputting the soft measurement parameters into a soft measurement prediction model to obtain the carbon content of the fly ash.
7. The soft measurement device for the carbon content of fly ash calibrated based on the burning method as claimed in claim 6, wherein the soft measurement parameters comprise actual load of a boiler, output of a coal feeder, received base volatile matter and low calorific value.
8. The soft measurement device for fly ash carbon content calibrated based on the burning method according to claim 6, wherein the value of N in the soft measurement prediction model is updated to 40-60 by training with the last N sets of training data.
9. The soft measurement device for fly ash carbon content calibrated based on the burning method according to claim 8, wherein the value of N in the soft measurement prediction model is updated to 50 by training with the last N sets of training data.
10. The soft measurement device for carbon content in fly ash calibrated based on the burning method as claimed in claim 6, wherein the soft measurement prediction model is a support vector machine model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110576286.8A CN113358811A (en) | 2021-05-26 | 2021-05-26 | Soft measurement method and device for fly ash carbon content based on firing method calibration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110576286.8A CN113358811A (en) | 2021-05-26 | 2021-05-26 | Soft measurement method and device for fly ash carbon content based on firing method calibration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113358811A true CN113358811A (en) | 2021-09-07 |
Family
ID=77527613
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110576286.8A Pending CN113358811A (en) | 2021-05-26 | 2021-05-26 | Soft measurement method and device for fly ash carbon content based on firing method calibration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113358811A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN107133460A (en) * | 2017-04-26 | 2017-09-05 | 中国能源建设集团广东省电力设计研究院有限公司 | A kind of online dynamic prediction method of boiler flyash carbon content |
| CN111754093A (en) * | 2020-06-11 | 2020-10-09 | 浙江浙能技术研究院有限公司 | Fly ash carbon content prediction method based on coal quality prediction and PSO-SVM |
| CN112287598A (en) * | 2020-09-28 | 2021-01-29 | 山西漳山发电有限责任公司 | Fly ash carbon content prediction method based on particle swarm parameter optimization |
-
2021
- 2021-05-26 CN CN202110576286.8A patent/CN113358811A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102778538A (en) * | 2012-07-06 | 2012-11-14 | 广东电网公司电力科学研究院 | Soft measuring method based on improved SVM (Support Vector Machine) for measuring boiler unburned carbon content in fly ash |
| CN107133460A (en) * | 2017-04-26 | 2017-09-05 | 中国能源建设集团广东省电力设计研究院有限公司 | A kind of online dynamic prediction method of boiler flyash carbon content |
| CN111754093A (en) * | 2020-06-11 | 2020-10-09 | 浙江浙能技术研究院有限公司 | Fly ash carbon content prediction method based on coal quality prediction and PSO-SVM |
| CN112287598A (en) * | 2020-09-28 | 2021-01-29 | 山西漳山发电有限责任公司 | Fly ash carbon content prediction method based on particle swarm parameter optimization |
Non-Patent Citations (1)
| Title |
|---|
| 楼锐 等: "一种基于 LSSVM 的飞灰含碳量软测量方法及装置", 《南方能源建设》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111461355B (en) | Dioxin emission concentration migration learning prediction method based on random forest | |
| CN103576655B (en) | A kind of power boiler burning subspace modeling and Multipurpose Optimal Method and system | |
| CN111754093B (en) | Fly ash carbon content prediction method based on coal quality prediction and PSO-SVM | |
| WO2021159585A1 (en) | Dioxin emission concentration prediction method | |
| CN110991756B (en) | MSWI furnace temperature prediction method based on TS fuzzy neural network | |
| CN112163350B (en) | Real-time on-line soft measurement system and method for pulverized coal fineness of double models of pulverizing system | |
| CN112131517B (en) | Method for measuring and calculating lower calorific value of garbage in garbage incineration power plant | |
| CN111931346A (en) | Real-time calculation method and application of low calorific value of coal as fired in coal-fired boiler | |
| CN111780127B (en) | Garbage incinerator combustion management system | |
| CN102226772B (en) | Device for measuring carbon content of fly ash in boiler of electric power plant | |
| CN110500596A (en) | A kind of dangerous waste incineration autocontrol method | |
| CN111462835A (en) | Soft measurement method for dioxin emission concentration based on deep forest regression algorithm | |
| CN113177352A (en) | Boiler combustion optimization system and method based on numerical simulation and artificial intelligence | |
| CN113358811A (en) | Soft measurement method and device for fly ash carbon content based on firing method calibration | |
| CN111505236B (en) | Coal quality monitoring method for acquiring element analysis in real time based on coal quality industrial analysis | |
| US6230495B1 (en) | Method for optimizing fossil-fueled power stations | |
| CN111855953A (en) | Method and system for quickly feeding back calorific value of coal as fired in coal-fired power plant | |
| CN116401948A (en) | Online prediction method and system for generating amount of power station boiler ash based on LSTM | |
| WO2022258811A1 (en) | A method and a system for monitoring and on-line determining of a calorific value of solid fuel that is currently combusted in a boiler | |
| CN113836794B (en) | Soft and hard combined fly ash carbon content online monitoring method | |
| CN100366876C (en) | Online analysis method and system for operation efficiency of combined gas-steam cycle power station | |
| CN201965099U (en) | Device for measuring fly ash carbon content of boiler in power plant | |
| CN111091251A (en) | Boiler operation optimization method and system based on big data technology | |
| CN113761454B (en) | Uncertainty-based garbage heat value testing system and optimization method | |
| CN109344561A (en) | One kind realizing flexible measurement method based on similarity principle |
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 |