Artificial Hummingbird Algorithm based Dynamic Generation Expansion Planning considering Renewable Energy Sources

Abstract

Generation expansion planning (GEP) is a primary and rigorous exercise in shaping the long-term decisions in terms of capacity expansion, location and technology of the power plants, to be committed for next 25–30 years, based on forecasted electrical demand. It is a non-linear, mixed-integer, stochastic, dynamic and discrete optimization problem. The metaheuristics are deemed the best optimization techniques to answer this multi-dimensional optimization problem with a large number of complicated constraints. In this work, least cost GEP problem is solved using a new optimization technique named as Artificial Hummingbird Algorithm considering the future horizon of 14 years encapsulating power generation additions required to cater for the forecasted peak demand with significant reliability and reduced emissions. A new efficient radix-5 mapping method for the representation of population search agents and power plants selectivity method based on priority enlisting is embedded in AHA framework. AHA has been implemented on standard emission constrained test cases considered in the literature. The proposed GEP framework provides promising results in terms of least cost and computational time with enhanced reliability and reduced emissions in contrast to the approaches presented in the literature.