• Composites Research
• /
• v.21 no.1
• /
• pp.30-39
• /
• 2008

In this paper, a parallel micro genetic algorithm was utilized in the optimal design of composite structures instead of a conventional genetic algorithm(SGA). Micro genetic algorithm searches the optimal design variables with only 5 individuals. The diversities from the nominal convergence and the re-initialization processes make micro genetic algorithm to find out the optimums with such a small population size. Two different composite structure optimization problems were proposed to confirm the efficiency of micro genetic algorithm compared with SGA. The results showed that micro genetic algorithm can get the solutions of the same level of SGA while reducing the calculation costs up to 70% of SGA. The composite laminated structure optimization under the load uncertainty was conducted using micro genetic algorithm. The result revealed that the design variables regarding the load uncertainty are less sensitive to load variation than that of fixed applied load. From the above-mentioned results, we confirmed micro genetic algorithm as a optimization method of composite structures is efficient.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.6
• /
• pp.44-55
• /
• 2009

Diagnosis of diseases using gene expression data obtained from microarray chip is an active research area recently. It has been done by general machine learning algorithms, because it is difficult to analyze directly. However, recent research results about the analysis based on the interaction between genes is essential for the gene expression analysis, which means the analysis using the traditional machine learning algorithms has limitations. In this paper, we classify the gene expression data using the hyper-network model that considers the higher-order correlations between the features, and then compares the classification accuracies. And also, we present the new hypo-network model that improve the disadvantage of existing model, and compare the processing performances of the existing hypo-network model based on general sequential processor and the improved hypo-network model implemented on parallel processors. In the experimental results, we show that the performance of our model shows improved and competitive classification performance than traditional machine learning methods, as well as, the existing hypo-network model. We show that the performance is maximized when the hypernetwork model is implemented on our parallel processors.