臺灣博碩士論文加值系統

本篇論文提出"修正遺傳演算法"以加強一般遺傳演算法的韌性,搜尋
最佳解的能力及速度,並實際應用於背壓式汽電共生系統的最佳化運轉.
在論文中首先以Sheble所提出的方法為例,來說明比較修正遺傳演算法的
特性及優點,然後應用於背壓式汽電共生系統在時間電價下的操作實例,由
於不需調整初值及任何參數,且能快速穩定的找到最佳解,修正遺傳演算法
非常適合於線上及時的運轉應用.

This thesis proposed the Modified Genetic Algorithm (MGA)
to enhance the robustness, search capability, and speed of
Genetic Algorithms(GAs), such thatthe algorithms can be applied
efficiently to the optimum operation of a back-pressure
cogeneration system[1]. First, we utilize a simple example
proposed by Sheble[2] to show thestrategies of MGA and
compare our performances to the result of Sheble, andthen we
make use of MGA to solve the optimum operation problem of the
back-pressure cogeneration plant under time-of-use rates. Need
not to adjust anyscaling factor and initial condition, MGA
can efficiently find the optimumsolution. It is proved that
MGA is a suitable method for on-line applications.