An expert system for alleviation of network overloads

This paper presents a prototype of an expert system for alleviation of network overloads in the day-to-day operation of a power system. The expert system provides a solution in real time for network overload alleviation using phase shifting transformers, and if overload still persists by rescheduling generation and/or load curtailment. The knowledge base of the expert system is obtained by carrying out extensive off-line analysis. An operational load flow (OLF) algorithm is used in the off-line analysis which incorporates a more realistic representation of the generator model and load characteristics. The off-line analyses are carried out using sensitivity factors (phase shifter distribution factors, generation shift sensitivity factors) and a linear programming (LP) optimization technique. The performance of the expert system is compared with the conventional LP optimization method of network overload alleviation and concurrent results have been obtained. The proposed expert system has been tested with the simulated conditions of a few practical Indian power systems. Results obtained for two Indian power networks of 24 nodes and 82 nodes are presented for illustration purpose. The expert system is found suitable for application in Energy Control Center as the solution is obtained very fast with a minimum of numerical computations.