Improved Pole Clustering Based Simplification of Complex Systems using Big Bang-Big Crunch Optimization

A mixed approach for the approximation of complex linear time invariant (LTI) systems via model reduction is presented in this article. The proposed approach utilizes the combined benefits of the improved pole clustering (IPC) and big bang-big crunch (BBBC) algorithm. The IPC technique is employed to compute the denominator polynomial while BBBC optimization algorithm determines the numerator polynomial of reduced order model (ROM) by minimization of error. Further, uniqueness of proposed approach is that, it always offers a stable approximants for stable higher order systems (HOS). Two multiple inputs-multiple outputs (MIMO) HOS are considered to show the proficiency of the proposed technique over prevalent schemes. Finally, the superiority of the suggested method over prevailing approaches is validated through error indices and their response plots.