DocumentCode :
2335679
Title :
New hybrid optimization model for power coal blending
Author :
Liao, Yan-Fen ; Wu, Chang-Hong ; Ma, Xiao-Qian
Author_Institution :
Electr. Power Coll., South China Univ. of Technol., Guangzhou, China
Volume :
7
fYear :
2005
fDate :
18-21 Aug. 2005
Firstpage :
4023
Abstract :
Power coal blending in power plant is an optimization problem with many complicated restriction conditions. As slagging properties of blended coal being regarded increasingly, the method that setting upper and lower boundaries for slagging properties indexes can´t satisfy the demand of the actual projects. Considering the fuzzy and non-linear characteristics of blended coal´s slagging, a slagging tendency index was brought out to take place these slagging properties in the paper, which is based on the coal properties of each blended coals and come from a fuzzification process by fuzzy neural network. The optimization model of power coal blending which aiming at the lowest price of blended coal was calculated based on genetic algorithm. Result showed that combining fuzzy neural network and genetic algorithm to solve the problem of power coal blending in power plant is very practical, as it can acquire the relative optimum results in a short time, which has a great significance on real-time blending coal and monitoring in power plant.
Keywords :
coal; fuzzy neural nets; fuzzy set theory; genetic algorithms; power engineering computing; power plants; power system measurement; thermal power stations; fuzzy characteristics; fuzzy neural network; genetic algorithm; hybrid optimization; nonlinear characteristics; power coal blending; power plant monitoring; slagging tendency index; Boilers; Combustion; Educational institutions; Fuzzy neural networks; Fuzzy sets; Genetic algorithms; Mathematics; Neural networks; Power generation; Silicon; Slagging; fuzzy neural network; genetic algorithm; power coal blending;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
Type :
conf
DOI :
10.1109/ICMLC.2005.1527641
Filename :
1527641
Link To Document :
بازگشت