Parametric optimization of reconfigurable designs using machine learning
ARC'13 Proceedings of the 9th international conference on Reconfigurable Computing: architectures, tools, and applications
| Hi-index | 0.00 |
Differential Evolution Algorithm (DE) is excellent optimization tools for complex high-dimensional multimodal problems. However, they require a very large number of problem function evaluations. In many engineering optimization problems, like design optimization or structure parameters identification, a single fitness evaluation is very expensive or time consuming. Therefore, standard evolutionary computation methods are not practical for such applications. Applying models as a surrogate of the real fitness function is a quite popular approach to handle this restriction. However, those early methodologies do suffer from some limitations, the most serious of which being the extra tuning parameter. A novel surrogate-assisted DE evolutionary optimization framework based on Gaussian process for solving computationally expensive problem is present. The study result indicates Gaussian Process assisted Differential evolution Optimization Procedure (GPDE) clearly outperforms standard DE evolutionary strategies on benchmark functions. Result is also presented for applications to a real-world problem: displacement back analysis for identification of rock mass parameters of a tunnel.