Order Reduction of the High Order Systems using the Optimized RLS and MV Methods

Nowadays, as it is particularly important to save money and energy, so it is better for the systems to reach less complexity. This complexity can exists in any field, that the dynamic aspects of systems and control systems are not excluded. So if the complexity of a system dynamical model is reduced, The implementation of the system in practice and also the design of an optimal controller will be done easier, economical and in a shorter time. It should be kept in mind that after reducing the order of systems, dynamic properties have not much changes. In this paper first, the optimized RLS (Recursive Least Squares) and MV (Minimum Variance) model reduction methods will be described briefly, and then a practical example is given which is related to the dynamic model of a hospital building of order 48. After that the adjustment of the sampling rate of the RLS method by using Fuzzy and reforming the reduced model in the mentioned case will be discussed. Finally, the effective parameters in the MV method will be set by using NSGA-II algorithm so that there is less error in reducing the order of the mentioned building model.

Keywords:

Fuzzy , Genetic Algorithm , Minimum Variance , Model Order Reduction , Recursive Least Squares