Evolutionary Programming Incorporating Neural Network for Transient Stability Constrained Optimal Power Flow

Abstract

Evolutionary Programming incorporating neural network (EPNN) is proposed to obtain the solution of Transient Stability Constrained Optimal Power Flow (TSCOPF) in this paper. Evolutionary Programming (EP) is selected as a main optimizer while the neural network is a supplementary tool to enhance the computational speed by screening out individuals, which have very high or low degrees of stability. Swing equation and limit of rotor angle deviation with respect to centre of inertia (COI) are treated as additional constraints in transient stability concern. The generator fuel cost minimization is selected as the objective function for TSCOPF. The proposed method is tested on IEEE 30-bus system with three different generator cost curves to account for the combined cycle generating unit and valve point loading effect of a thermal generating unit. A three-phase fault at a specific transmission line is considered as a single contingency. The simulation results show that the proposed method is capable of searching for the optimal or near optimal solution of TSCOPF. Moreover, the exploitation of the neural network leads to huge computational time saving due to absence of time-domain simulation for some individuals during EP search.