The memetic ant colony optimization with directional derivatives simplex algorithm for time delays identification

Abstract

The identification of time delay in the linear plant is important tasks. Most of the conventional identification techniques, such as those based on least mean-squares, are essentially gradient-guided local search techniques and they require a smooth search space or a differentiable performance index. New possibility in this field is opened by an application of the hybrid Ant Colony Optimization (ACO) with local optimization algorithm. The Directional Derivatives Simplex (DDS) as a local optimization algorithm is proposed in the paper and used in the memetic ACODDS method. The ACODDS algorithm is compared with ACO and a classical methods: Global Separable Nonlinear Least Squares (GSNLS). The obtained results suggest that the proposed method performs well in estimating the model parameters. © 2011 Springer-Verlag Berlin Heidelberg.