Application of Continuous Ant Colony Optimization in Economic Dispatch Problems  

Oleh:

Yin, Yueh-Chuan ; Liang, Yun-Chia

Jenis: Article from Proceeding
Dalam koleksi: The 14th Asia Pacific Industrial Engineering and Management Systems Conference (APIEMS), 3-6 December 2013 Cebu, Philippines, page 1-8.
Topik: Ant Colony Optimization (ACO) ; Continuous Ant Colony System (CACS); Probability Distribution Function (PDF); Economic Dispatch (ED).
Fulltext: 1094.pdf (302.76KB)

Isi artikel

Ant colony optimization (ACO) was first proposed by Dorigo in 1992 as a multi-agent approach to solve difficult combinatorial problems. Many variants have been presented and applied to different combinatorial problems since then. Recently, some methods took inspiration from ACO and were proposed for continuous domains, but have seemed to lose the real spirits of ACO. However, one pheromone-based ACO algorithms for continuous optimization problems that claimed to keep all major characteristics of ACO was proposed lately. The new algorithm called Continuous Ant Colony System (CACS) was introduced by Pourtakdoust and Nobahari in 2004. To deal with a continuous function, the pheromone distributions over the search space were modeled in the form of a Probability Distribution Function (PDF). The center of the PDFs is the last best global solution found, and its variance depends on the aggregation of the other promising areas around the best one. CACS employs a state transition rule with both exploration and exploitation strategies and a pheromone updating rule which concurrently simulates pheromone accumulation and pheromone evaporation. In this paper, the CACS algorithm and a Modified CACS (MCACS) algorithm are presented to solve the Economic Dispatch (ED) problem with transmission losses. It is believed to be the first application of the CACS algorithm for the power system problems. The feasibility of the two algorithms is tested on the system with 6 generators, 30 buses and 41 lines (also known as IEEE 6-generator 30-bus system) instance. The experimental results show the effectiveness and efficiency of the CACS algorithms.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali