Title:  Autonomous Multi-Stage Flexible Optimal Power Flow

Committee: 

Dr. Meliopoulos, Advisor        

Dr. Saeedifard, Chair

Dr. Sun

Abstract:

The goal of the proposed PhD research is to optimally coordinate the control of different devices through autonomously formulating and solving multi-stage flexible optimal power flow (OPF) problems of the corresponding system. The method incorporates detailed device models using a physically based object-oriented modeling approach. First, every device is modeled as a mathematical object in a standard syntax. When different devices are connected to form a system, a network formulation procedure is performed to construct the network model presented in the same standard form. Then, combining the network model with a user-defined objective function yields a multi-stage alternating current quadratic OPF problem, which is solved through a sequential linear programming (SLP) algorithm. Current conservation equations at nodes instead of conventional power flow equations are used as the equality constraints. The power flow model, together with the operating states and controls as well as the past history vector at the first stage of the OPF, are required by the SLP algorithm. The required initial system information can be provided by either simulation or dynamic state estimation. Each iteration of the algorithm goes through constraint violation check, OPF linearization, linear programming problem solving and operating point update. Finally, the optimal operating point of the system is obtained upon the convergence of the SLP algorithm. By implementing the optimal control actions in the system, the objective of the multi-stage OPF problem is achieved. The whole process of the proposed method is done autonomously, i.e. all steps are automatically performed by manipulating the objects of the component models.