A method is proposed for model order reduction for a linear multivariable system by using the combined advantages of dominant pole reduction method and Particle Swarm Optimization (PSO). The PSO reduction algorithm is based on minimization of integral square error (ISE) pertaining to a unit step input. Unlike the conventional method, ISE is circumvented by equality constraints after expressing it in frequency domain using Parseval's theorem. In addition to this, many existing methods for model order reduction are also considered. The proposed method is applied to the transfer function matrix of a 10 th order two-input two-out put linear time invariant model of a power system. The performance of the algorithm is tested by comparing it with the other soft computing technique called Genetic Algorithm and also with the other existing techniques.

This paper presents a new unified power-quality conditioning system (MC-UPQC), capable of simultaneous compensation for voltage and current in multibus/multifeeder systems. In this configuration, one shunt voltage-source converter (shunt VSC) and two or more series VSCs exist. The system can be applied to adjacent feeders to compensate for supply-voltage and load current imperfections on the main feeder and full compensation of supply voltage imperfections on the other feeders. In the proposed configuration, all converters are connected back to back on the dc side and share a common dc-link capacitor. Therefore, power can be transferred from one feeder to adjacent feeders to compensate for sag/swell and interruption. The performance of the MC-UPQC as well as the adopted control algorithm is illustrated by simulation. The present work study the compensation principle and different control strategies used here are based on PI & ANN Controller of the MC-UPQC in detail. The results obtained in MATLAB/PSCAD on a two-bus/two-feeder system show the effectiveness of the proposed configuration.Index Terms-power quality (PQ) unified power-quality conditioner (UPQC), voltage-source converter (VSC),Aritifical neural network-ANN.