A New Class Of Operations To Accelerate Particle Swarm Optimization Algorithm And A Novel Hybrid Approach For Unit Commitment Problem.
An efficient technique in handling a large number of constraints is through the evolutionary computation (EC) method such as a Genetic Algorithms(GAs0, Evolutionary Programming(EP), Evolutionary Strategies(ES) etc. Particle Swarm Optimization is a population based optimization technique under EC cat...
Saved in:
| 主要作者: | |
|---|---|
| 格式: | Thesis |
| 出版: |
2004
|
| 主题: | |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | An efficient technique in handling a large number of constraints is through the evolutionary computation (EC) method such as a Genetic Algorithms(GAs0, Evolutionary Programming(EP), Evolutionary Strategies(ES) etc. Particle Swarm Optimization is a population based optimization technique under EC category. In this study, the performance of the Particle Swarm Optimization (PSO) algorithm is improved considerably by introducing a new class of operators to manipulate particles in each generation. These operators are chosen from an empirical study and testing of a large number of opeators. From this empirical study, it is found that the performance of the operators differs from each other and varies with the benchmark problems. Among these operators, the best operators are chosen and divided into three categories namely mutation, crossover and variant. |
|---|