Hybrid Intelligent Technique for Solving Non-convex Economic Load Dispatch Problem

摘要

This paper, attempts to demonstrate the solution of non-convex economic load dispatch (ELD) problems by taking into consideration of consumer load pattern variations through a hybrid intelligent technique, which is a combination of the use of artificial neural networks (ANN) and improved fast evolutionary programming (IFEP) technique. In the proposed approach, an evolutionary based search algorithm is used to achieve global optimum ELD solution and an auto configuring ANN approach is adopted to obtain fast solution. This hybrid technique is tested on the IEEE 6-bus system with several heuristic load patterns. It has been observed that the trained network successfully gives the results almost instantaneously with sufficient accuracy for the unknown load patterns. This powerful and versatile feature of the proposed method may be quite befitting for real time implementation of the ELD in an energy management system without any restrictions on the nature of the cost curves.