A Novel Multi-Objective CMAES Algorithm for Economic Emission Dispatch Incorporating Wind and Solar Power Uncertainties

To address the climate change challenges, the substantive pollutant emissions from traditional thermal generation plants need to be significantly reduced, leading to increased interests in the economic emission dispatch (EED) problems in recent years. Further, the penetration of renewable energy such as wind power into the power system is also gaining momentum. This paper proposes a covariance matrix adaption evolution strategy based on the non-dominated sorting and archive mechanism (NSA-CMAES) to solve the EED problem which takes into account both the economic cost and emissions from thermal plants, as well as uncertainties associated with renewable power generations. The NSA-CMAES extends CMAES into a multi-objective algorithm by using a non-dominated sorting mechanism, and the resulting Pareto solutions are stored in an external archive that is constantly updated until the termination conditions are met. The EED problem is formulated with the consideration of stochastic wind energy and solar energy generation, and the NSA-CMAES is applied to solve the EED problem in comparison with some popular algorithms used for the benchmark problems as well as some classical multi-objective algorithms. The simulation experiments confirms that the proposed NSA-CMAES algorithm outperform existing approaches for the EED problems.