From Recombination of Genes to the Estimation of Distributions I. Binary Parameters
PPSN IV Proceedings of the 4th International Conference on Parallel Problem Solving from Nature
Population-Based Incremental Learning: A Method for Integrating Genetic Search Based Function Optimization and Competitive Learning
Proceedings of the 9th annual conference on Genetic and evolutionary computation
Optimizing continuous problems using estimation of distribution algorithm based on histogram model
SEAL'06 Proceedings of the 6th international conference on Simulated Evolution And Learning
An evolutionary algorithm with guided mutation for the maximum clique problem
IEEE Transactions on Evolutionary Computation
Protein Folding in Simplified Models With Estimation of Distribution Algorithms
IEEE Transactions on Evolutionary Computation
| Hi-index | 0.00 |
The automated synthesis and optimization of power electronic circuits (PECs) is a significant and challenging task in the field of power electronics. Traditional methods such as the gradient-based methods, the hill-climbing techniques and the genetic algorithms (GA), are either prone to local optima or not efficient enough to find highly accurate solutions for this problem. To better optimize the design of PECs, this paper presents an extended histogram-based estimation of distribution algorithm with an adaptive refinement process (EDA/a-r). In the EDA/a-r, the histogram-based estimation of distribution algorithm is used to roughly locate the global optimum, while the adaptive refinement process is used to improve the accuracy of solutions. The adaptive refinement process, with its search radius adjusted adaptively during the evolution, is executed to search the surrounding region of the best-so-far solution in every generation. To maintain the diversity, a historic learning strategy is used in constructing the probabilistic model and a mutation strategy is hybridized in the sampling operation. The proposed EDA/a-r has been successfully used to optimize the design of a buck regulator. Experimental results show that compared with the GA and the particle swarm optimization (PSO), the EDA/a-r can obtain much better mean solution quality and is less likely to be trapped into local optima.