Reduced Order Modeling of Large Scale Systems Using Soft Computing Techniques

Abstract

The system which gives the direct output is an open-loop system and combination of newlinemachines, equipments and with the controller in a manner with complete loop is closed-loop newlineknown as a complete system. These machine systems are designed for a particular output. The newlinesystem can be represented by mathematical model. The comprehensive description of the newlinemathematical procedure leads to differential equations or partial differential equations newlineresulting to higher-order transfer function of a system. The higher-order systems (HOS) are newlinehectic and difficult to design and analysis. The computation time to analysis these systems are newlinealso high. The economies in a hardware requirement for simulation of the system are large. newlineThe procedure suggested by researchers to reduce HOS in the lower order system newline(LOS) using a systematic mathematical format and known as model order reduction (MOR). newlineThe LOS preserves the important characteristics of the original HOS with an acceptable newlineapproximation. The obtained reduced order if correct, quickly traces the essential newlinecharacteristics of the HOS. The procedural reduction of these large-scale systems is suggested newlineby the researcher and categorised as time-domain and frequency-domain. newlineThe classical reduction method (CRM) has limits of instability, non-minimum phase newlinebehavior and law accuracy in the medium and high-frequency range. The stability preservation newlinemethod (SPM) has the drawback of lack of flexibility. The Moment matching from CRM make newlinethe system unstable due to mathematical outcomes from matrix cofficeints. This unstability newlinehas been removed by pole placement technique. In making the composite methods a method newlinefrom CRM and SPM are utilized. The result from the CRM uses to form unknown numeraotr newlineparameter and SPM for obtaining the denominator parameter. The method from SPM also newlinegives the mixed method result. The composite method uses the Routh stability criteria and newlinestability equation method for obtaining the stable result and easily preserve the stability of newlinereduced order system.