A hybrid intelligent approach for optimal control of seed cleaner

Abstract

This study presents a hybrid intelligent approach to model and optimize a seed cleaning/separating process, which separates seeds varying with different types, varieties and lots in a seed cleaner system (5XZW-1.5). In this controlling scheme, the nonlinear system is modeled by Artificial Neural Network, and optimized by Genetic Algorithms. The uniform design is used to sample datasets. Control results suggest that the intelligent approach is useful for the optimization of such a complex process. © 2006 International Federation for Information Processing.