Method for measuring fly ash carbon content based on PSO (Particle Swarm Optimization) and epsilon-SVR

Abstract

The invention discloses a method for measuring a fly ash carbon content based on PSO (Particle Swarm Optimization) and epsilon-SVR. The method specifically comprises the following steps: (1) by takingfly ash samples of different carbon contents of an actual power plant as calibration samples, acquiring laser induced plasma emitting spectrums of the calibration samples by using LIBS (Laser-inducedBreakdown Spectroscopy) detection equipment; (2) extracting characteristic peak spectrum data related to fly ash carbon content measurement from the acquired LIBS spectrum, and carrying out normalization processing; (3) optimizing spectrum data after normalization processing by using a PSO (Particle Swarm Optimization) algorithm on the basis of support vector machine related parameters of fly ashcarbon content measurement; (4) establishing an epsilon-SVR calibration model according to the confirmed optimal parameters and an SVM (Support Vector Machine); and (5) processing the fly ash samplesto be detected so as to obtain spectrum data, inputting the spectrum data into the calibration model, thereby obtaining carbon contents of the fly ash samples to be detected. With the combination ofthe method and an artificial intelligence algorithm, LIBS is applied to measurement of fly ash carbon contents, the calculation time is short, the measurement accuracy and precision can be improved, and the application range is wide.

Description

technical field [0001] The invention relates to the technical field of industrial measurement, in particular to a method for measuring the carbon content of fly ash based on PSO and ε-SVR. Background technique [0002] The carbon content of fly ash is an important indicator of the combustion efficiency of coal-fired boilers in thermal power plants. When the carbon content of fly ash is high, it means that the coal consumption and power generation cost are too high. Real-time and accurate detection of carbon content in fly ash is beneficial to adjust the air-to-coal ratio; control the carbon content in fly ash within the optimal range and improve the level of boiler combustion control to ensure the economical, safe and stable operation of the unit, but due to The carbon content of fly ash is affected by various factors such as coal type, boiler structure, and operation level, so it is difficult to achieve online measurement. [0003] Many off-line measurement methods, such a...

AI Technical Summary

Problems solved by technology

[0004] At present, most of the analysis methods for fly ash unburned carbon detection based on LIBS use univariate or multivariate linear regression analysis methods. Although the accuracy can be improved by means of multivariate regression, the effect still cannot meet the needs of industrial sites.

Images