Title :
Soft-Sensing of Oxygen-content in Flue Gases of Power Plant Based on LS-SVM and Simplex Algorithm
Author :
Liu, Changliang ; Li, Shuna
Author_Institution :
Sch. of Control Theor. & Control Eng., North China Electr. Power Univ., Baoding, China
Abstract :
Oxygen-content in flue gases is an important factor for the economical burning in the power plant. Because of influence from many different factors, it has some difficulties in testing of oxygen-content in flue gases. In this paper the author chooses indirect variables and then the soft sensor model based on least square support vector machine for oxygen-content in flue gases of power plant is put forward. Simplex algorithm is applied on searching for the two necessary parameters of LS-SVM, and practical data is handed together to test the model. Simulation result shows that the method has more evident advantages than both the traditional oxygen-content instrument and the radial basis function-based soft sensor. In this paper, we take RBF as the acronym of radial basis function. It also has a better capability index and of important significance for economical burning in the power plant.
Keywords :
flue gases; least squares approximations; power engineering computing; radial basis function networks; support vector machines; thermal power stations; LS-SVM algorithm; flue gases; least squares support vector machines; oxygen-content testing; power plant; radial basis function; soft-sensing technology; Artificial neural networks; Boilers; Combustion; Flue gases; Power generation; Power generation economics; Software engineering; State estimation; Support vector machines; Testing; Least squares support vector machines (LS-SVM); oxygen-content in flue gases; simplex algorithm; soft-sensing;
Conference_Titel :
Software Engineering, 2009. WCSE '09. WRI World Congress on
Conference_Location :
Xiamen
Print_ISBN :
978-0-7695-3570-8
DOI :
10.1109/WCSE.2009.51
