Online ANN Memory Model-Based Method for Unified OPF and Voltage Stability Margin Maximization

摘要

Online implementations of unified Optimal Power Flow (OPF) algorithms pose several challenges. One such challenge is to maximize the voltage stability margin (VSM), which arises because of the associated computational complexity. In this paper a new artificial neural network (ANN) memory model-based algorithm for unified OPF is proposed. The proposed algorithm maximizes VSM while minimizing two other system-level objectives of generation cost and transmission loss. The proposed algorithm uses a slower conventional algorithm like that found in [17] to schedule several load patterns and obtains their associated optimal schedules. The ANN memory model stores these load patterns and their associated optimal schedules. Whenever the ANN memory model is given a load pattern, it finds out the closest stored load pattern and its associated optimal schedule. Thereafter power flow equations are solved using the present load pattern with the recalled optimal setting vector. Load bus voltage violations, if any, are removed using a fuzzy expert system corrective control algorithm (FECCA). The proposed algorithm was implemented and tested on the IEEE 30-bus system. Its execution time was one-sixteenth of that of the conventional algorithm like that in [17].