An Application of Neural Network to Power Plant Expansion Strategies as Constrained by Ambient Air Quality

碩士 === 逢甲大學 === 環境工程與科學所 === 91 === Fossil fuel-fired power plants, particularly coal-fired plants, are instrumental to the economic development in Taiwan in recent decades. The importance of fossil fuel-fired power plants in Taiwan is evidenced by the large generating capacity and annual electrici...

Full description

Bibliographic Details
Main Authors: Shiao-Juan Tu, 涂曉娟
Other Authors: none
Format: Others
Language:zh-TW
Published: 2003
Online Access:http://ndltd.ncl.edu.tw/handle/92563029212572671288
Description
Summary:碩士 === 逢甲大學 === 環境工程與科學所 === 91 === Fossil fuel-fired power plants, particularly coal-fired plants, are instrumental to the economic development in Taiwan in recent decades. The importance of fossil fuel-fired power plants in Taiwan is evidenced by the large generating capacity and annual electricity generation─both exceeding 60%. Privatization in power generation is the current trend; competition of electricity supply can alleviate the supply shortage and can lower the cost of power generation through competition. The objective of this research is to identify the optimal strategies in capacity expansions of electricity generation using the following tools, lBack propagation network (BPN) of the artificial neural networks lGray linear programming of the gray linear system theory The research uses the capacity expansions of Taichung Coal-Fired Power Plant and Mai-Liao Co-Generation Plant while meeting the ambient air quality standards of sulfur dioxide (SO2) to demonstrate the feasibility of the methodology. The analysis results suggest that the total expansions for both Taichung and Mai-Liao are between 5,200 MW and 18,600 MW using the BPN method and 5,200 MW using the gray linear programming method. The result of the gray linear programming method is consistent with the lower limit of the BPN method.