Joint Renewable Generation Maximization and Network Reconfiguration for Power Distribution Systems

Abstract:

This talk concerns the network reconfiguration problem for radial power distribution systems, a joint optimization problem to determine the output of distributed generation and power network switch status in order to maximize the utilization of renewable energy sources. This is a classical problem for power system operations, planning and analysis with numerous variants and extensions. Traditional solution methods typically involve alternating optimization with respect to the continuous generation variables and the discrete switch status variables, without any guarantee of optimality. In this talk, we will introduce a “one-shot” optimization model for the joint problem as a mixed integer program with linear and non-convex bilinear constraints. The problem can be efficiently solved using modern optimization technology, namely Gurobi 9 or later, if an appropriate problem formulation is used. Numerical studies indicate that the proposed method outperforms well-known tools such as MATPOWER interior point optimization, second order cone relaxation and linearization in relevant situations. The effect of different formulations on computation performance will be illustrated, though in-depth analysis of the performance difference remains to be done.

Biography:https://www.nsysu.edu.tw/p/406-1000-251722,r3244.php?Lang=en