Design of additive models using hybrid soft computing approaches

An indispensable ability for intelligent control is to comprehend and learn about plants, disturbances, environment, and operating conditions. In this paper, a modified probabilistic incremental program evolution (MPIPE) algorithm and a random search algorithm are used as a promising tool for such purposes. In order to identify and evolve the structure and parameters of the additive models simultaneously, a hybrid method is proposed, in which the MPIPE is used for the identification of structure of the additive models, and the parameters used in additive models are optimized by a random search algorithm. Simulation results for the identification of linear/nonlinear systems show the feasibility and effectiveness of the proposed method