Open-Loop-Data-Based Integer-and Non-integer-Order Model Identification Using Genetic Algorithm (GA)

摘要

Modeling of the process is very important aspect of engineering which helps us to understand the process behavior under different conditions. Also from control point of view, a good process model always proves to be vital in designing a good controller. Based on the order of the model, the process can be modeled into two categories, i.e., integer- and non-integer-order models. As non-integer modeling provides improved precision of the process model by offering more flexibility in model identification, a good number of researchers are utilizing this concept to obtain better results. Therefore, in the present work an attempt has been made to identify models for some processes based on its open-loop data. Therefore, for open-loop-data-based model identification, both integer- and non-integer-order models are estimated by minimizing the integral error criteria using genetic algorithm (GA). Comparative analysis ratifies that the non-integer model is able to capture process dynamics more accurately as compared to integer-order model.