Real-time identification of composite load model using multiple model approach

Undoubtedly power system is one of the most complex engineering systems which have been built by human being. Simulation studies of the power system are of great importance since experimental tests in the power systems are very expensive and may interrupt continuous power delivery. Similar to other dynamical systems, a prerequisite step toward conducting valid simulations of the power system is accurate modeling and identification of the system model and parameters. Accurate models for generators and transmission lines are available while load modeling is still a challenge. Its difficulty lies in the aggregation of different loads with different characteristic and stochastic changes. Load modeling of power systems is of great importance in power system simulation for stability analysis, impedance relay setting, and load shedding.;In this thesis, dynamic and static load modeling techniques are reviewed and load modeling importance on static and dynamic voltage stability is demonstrated. Fixed and moving structure multiple model estimation algorithms are reviewed and used to design an online load estimator. Simulation results are provided for a composite load including IEEE type-6 motor to show the effectiveness of the proposed real time estimator. This thesis provide a step in the right direction toward developing accurate real time techniques for stability analysis.;The contributions of this research include a new technique to perform improved real time load parameter estimation. Contrary to existing off line stability techniques, the real time feature allows, for the first time, the performance of online stability analysis. The illustration of real time static stability is provided in this work.