Harmony Search (HS) is a meta-heuristic algorithm which was first introduced in 2001 and it became a widely used optimization algorithm in various areas in engineering application as well as in water resources planning and management. However, as most meta-heuristic algorithms are, the HS shows a good performance in global search but not as good in local search. This study aims the improvement of both exploration and exploitation capability of the algorithm. The mission has been carried out by changing algorithm operators or parameters in the search process. Several types of Improved Harmony Search (IHS) have been successfully developed resulting better exploiting (local) search. Alternative way is to utilize the superior local search of other models or algorithms. The combined, so called hybrid algorithms can significantly supplement the weak local search aspect of the original HS. A newly developed hybrid algorithm, Smallest Small World Cellular Harmony Search (SSWCHS), is developed and proposed shorter characteristic path length and higher clustering coefficient, resulting good exploration and exploitation efficiency. Application to benchmark functions and design of pipe networks proves the superior performance of the newly developed hybrid algorithm.
Kim, Joong Hoon; Yoo, Do Guen; Lee, Ho Min; and Im, Sung Soo, "Improving Exploration And Exploitation Capability Of Harmony Search Algorithm" (2014). CUNY Academic Works.