The PID Controller Optimization Design based on the Modified Water Flow-Like Algorithm

• Main Authors: Lan, Hsiao-Yun, 藍孝雲
• Other Authors: Yu, Ker-Wei
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/49255312863624895359

碩士 === 國立高雄海洋科技大學 === 輪機工程研究所 === 99 === In this study, the PID controller parameter optimization method based on modified water flow-like algorithm (MWFA) is investigated. The original water flow-like algorithm (WFA) was simulated the physical characteristics of water flow which the water flow will stop at the lowest position on the ground surface. The position will be map to optima solution within solution space of the optimization problem. The basic methodologies of WFA are flow merging, flow splitting, evaporation and precipitation. The ability of avoid local minimum problem is depend on precipitation. However, the performance does not quit good of the original WFA. A new mechanism, namely, wind brow is proposed to increase the ability of avoid the local minimum problem. There are five benchmark problems employed to exam the effective of the proposed method. The proposed MWFA is also applied to adjust the PID controller’s parameters to achieve optima goal.