KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Siriboon, KritawanSubject: search.filters.subject.HS

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Improved Rerun Particle Swarm Optimization Algorithm with Harmony Search
    (2019-04-10)
    Phuchan, Wikrom
    ;
    Kruatrachue, Boontee
    ;
    Siriboon, Kritawan
    One of the most crucial problem of the particle swarm optimization is that it can easily trap in local optima. There are some studies tried to resolve the drawback using the reset and mutation mechanism. Although it unraveled the problem well when the optimal position of benchmark function is at zeros in all dimensions. It cannot locate optima when the exact answer randomly shifted over the possible region. This study integrates Harmony Search (HS) to the rerun and reset mechanism to PSO. It performs well with selected benchmark functions, when the optimal position is randomly shifted in most of the test functions. Since it does not favor optimal point at zeroes when the optimal position is not shifted, the number of calls of the evaluation function is more than the other algorithms in some cases.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Publication,
    Hybrid multi-swarm with Harmony Search algorithm
    (2017-11-03)
    Phuchan, Wikrom
    ;
    Kruatrachue, Boontee
    ;
    Siriboon, Kritawan
    This paper proposed a new metaheuristic algorithm, Hybrid Multi-swarm with Harmony Search algorithm which combines two famous metaheuristics, particle swarm optimization (PSO) and Harmony Search algorithm (HS). The main advantage of PSO is its convergence speed while its main drawback is trapping in local optimum problem. To improve PSO performance, this research use HS to increase PSO diversity and extend its convergence point to a better local optimum. The proposed algorithm was compared with three related algorithms on the optimization benchmark functions. The experiment results show the proposed algorithm yields better fitness value solution.