模糊蜂群演算法

Fuzzy Artificial Bee Colony Algorithm

卓子菱(Tzu-Ling Cho)

指導教授 ： 蔡賢亮

義守大學/電機資訊學院/資訊管理學系/碩士(2017年)

摘要

本研究針對2005年學者Karaboga所提出之人工蜂群演算法(Artificial Bee Colony algorithm,ABC)進行探討，該演算具有收斂速度穩定、求解能力佳及少量參數控制之優點，但仍存在晚期收斂速度緩慢及陷入區域解問題等情形。因此本研究參考了相關人工蜂群改良之文獻，而提出了模糊蜂群演算法，其運用模糊理論及差分演算法之突變方式，幫助蜂群降低陷入區域解之情形且能有效地持續開發。此外將模糊蜂群演算法與目前人工蜂群改良較優越的演算法(Gbest-guided Artificial Bee Colony algorithm, GABC）相結合，期望可以提升蜂群求解能力。
從本研究以測試函數進行實驗評估，從實驗結果得知，本研究提出之模糊蜂群演算法及與GABC相結合之模糊蜂群演算法，確實改善陷入區域解之情形，使其在後期可以持續地有效開發。

關鍵字

人工蜂群演算法 ； 模糊理論 ； 差分演算法

並列摘要

This study discussed Artificial Bee Colony Algorithm(ABC) was proposed by Karaboga in 2005. This optimization algorithm has a stable convergence rate, solving capabilities and an amount of control parameters. Though ABC has many advantages, there are still has some problems, such as the regional situation and the slow convergence. Therefore, this study refers to the research of Improved ABC, presented Fuzzy Artificial Bee Colony Algorithm(FABC). This research utilized fuzzy theory and Differential Evolution to improve the searching capability and falling into local optimal solution. In addition, the FABC is combined with GABC(FGABC), which is expected to improve the accuracy of the solution.
From the experimental results, it is show that FABC and FGABC have improved the situation of falling into the regional solution, so that it can be continuous and effective development.

並列關鍵字

Artificial Bee Colony algorithm ； Fuzzy Theory ； Differential Evolution

參考文獻

3.李維平、簡璟蔚、蔡宛庭(2011)，改良突變權重的差分進化演算法，進工程學刊，vol.6。

4.李維平、江長育、蔡宛庭(2011)，搭配擾動策略之差分演化演算法，資訊科技國際期刊(IJAIT)，vol.5。

2.李維平、李元傑、謝明勳(2014)，以群中心策略改良人工蜂群演算法資訊管理學報，21(1)，25-44。

4.Karaboga, D. & Basturk, B. (2007). “A powerful and Efficient Algorithm for Numerical Function Optimization: Artificial Bee Colony (ABC) Algorithm. Journal of Global Optimization”, Vol. 39:1, pp. 459-171.

5.Karaboga, D. & Basturk, B. (2008). “On the performance of artificial bee colony (ABC) algorithm”, Applied Soft Computing, Vol. 8, No. 1, pp. 687–697.