Distributed Control of Power Systems with Demand Participation
In this dissertation work, we present our research on decentralized and centralized control strategies in power systems using convex optimization methods. We have analyzed nonlinear models of power systems. These models include an exact representation of frequency and terminal voltage through which we are able to cope with the uncertain terms and unknown design parameters of a power system stabilizer. Decentralized control lies in two aspects. First, the control approach to the problem is aimed at solving the voltage regulation problem of the power system. We suggest a method to solve the stabilization problem, which includes uncertain time-varying parameters using a state feedback controller. The main objective of the decentralized control scheme is to control the terminal voltage of synchronous generators using a decentralized voltage controller. Our model has been expanded with a stabilizer to obtain the output feedback controller. We have optimized the design parameters, which depend on the uncertainty intervals, within the feasibility region of uncertain parameters. We have implemented a bisection procedure to determine the value of design parameters. Second, we used an observer-based control for a decentralized stabilization of a multimachine power system. We have verified the contraction region for a multimachine power system. We have also performed numerous simulations of power system models to prove the stability properties of the extended Kalman filter based on contraction theory. Finally, this work analyzes the optimal power flow problem by integrating renewable sources with demand participation in electric grids. The demand participation has been achieved by demand-side resources with renewables to curtail the actual loads. With penetrations of renewable energy in a power system, the problem has been solved by a semidefinite programming method. In addition, this method is presented for contingency scenarios, such as generation unit failures and transmission lines failures. These scenarios have been used to determine the effect of shedding load, dropping or tripping generation, or tripping transmission lines in the power system.