FINDING ALL SOLUTIONS OF MULTIMODAL OPTIMIZATION USING FLOWER POLLINATION ALGORITHM WITH NICHING METHOD AND CLUSTERING METHOD
At this time the optimization is a problem that is almost found in all areas of research. The problem is not only in the form of functions with one or many variables but also in the form of unimodal and multimodal. Most of algorithms, both Gradient and Metaheuristic based, including Flower pollinati...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Online Access: | https://digilib.itb.ac.id/gdl/view/30146 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | At this time the optimization is a problem that is almost found in all areas of research. The problem is not only in the form of functions with one or many variables but also in the form of unimodal and multimodal. Most of algorithms, both Gradient and Metaheuristic based, including Flower pollination algorithm, are generally constructed to solve global optimization problems, so they do not work well when applied to multimodal problems that have a lot of optimum both local and global minimum and maximum or a combination of them. The goal of this study is to find all optimum solutions of multimodal problems in a single run. In this thesis, we have modified two methods namely FPA with Niching method and FPA with Clustering method. Both methods are validated using a benchmark of multimodal function that is competed in the 2013 CEC session. The results of the comparison between the two methods show that the FPA with Niching method only offers an efficiency but can not find all solutions of benchmark function and the optimum number obtained will frequently change in every running time, whereas FPA with Clustering method is able to localize all optimum solutions both local and global optimum of multimodal benchmark function with optimum number always same in every running time. |
|---|