Model order diminution of MIMO systems using the delta transform method with new firefly-based hybrid algorithms

In this work, a novel methodology for the order diminution of multi-input multi-output (MIMO) systems has been explored by applying some firefly-based hybrid methods. Constrained optimization techniques have been formulated to address this problem. A new combined domain, with the fusion of continuous and discrete-time systems at higher sampling frequencies, has been taken up to formulate lower-order models corresponding to the original system. A real-time power system model has been chosen appropriately to validate the efficacy of the suggested method. An exhaustive comparison has been performed with new heuristic methods as well as some classical approaches to demonstrate the superiority of the advocated techniques. About 75% or more betterment is produced by the suggested techniques in comparison to the parent and other existing techniques in terms of the error function, which is quite a drastic improvement. The computation time and memory usage of the algorithms have been assessed to manifest the efficacy of the novel integrated methods. The convergence plot between the normalized error function and the number of iterations also support adequate justification with respect to convergence speed and accuracy. Several nonparametric tests also prove the significance of the results thus obtained.