The Application of Neural Network in Combustion system of Refuse Incineration Plant

• Main Authors: JAU-AN JONG, 鍾招安
• Other Authors: Lu; Shi
• Format: Others
• Language: zh-TW
• Published: 1996
• Online Access: http://ndltd.ncl.edu.tw/handle/21123484824070580718

碩士 === 國立臺灣科技大學 === 機械工程研究所 === 84 === Incineration treatment of refuse can achieve the goals of refuse volume reduction ， resourcefulness ，sanitization and stabilization.As a result,To build incineration plant is the very best way to manage with refuse for an island country.The eligible way to deal with refuse in Taiwan should major in in- cineration plus coverage infuture to follow up the requirement of increment of living level and environment. However,the makeup and characteristices of refuse are the main factors to affect the design and planning of a incineration plant.If we can''t handle the characteristics of refuse very well not only planning of plant not well done,but also the performance and life of plant is greatly shortened. The refuse depends on variant periods and places i.e. the characteristics will be not same according to difference of places and seasons.The tradition- al sampling analysis is quite complex and dependent on time- variant factors,so we propose this method which is combined with theory of neural network to derive exactly the output such as lower heating value of refuse、the amount of burnt refuse etc.) according to the input as（sources of refuse ，climate，characte- ristics of seasons）to get a set of best inputs in advance to let system operated well to increase the the efficiency of incinerat- ion.