The maximum likelihood optima for an economic load dispatch in presence of demand and generation variability

Abstract

A constraint to represent maximum likelihood of occurrence for coexistent variable demand and variable generation is proposed for inclusion in the conventional static load dispatch. The constraint allows direct representation of cross-correlations between variable power levels, which makes the dispatch appropriate for scenarios with high penetration of renewable power generation. At optima, the augmented dispatch converges to utility generation cost that is “most likely to occur”; and can therefore provide useful support for generation resource planning by accommodating demand and generation variability. Formulation of the maximum likelihood load dispatch, as well as Karush-Kuhn-Tucker conditions for its optima, are presented in the paper. Features of the optima are established by three-generator and twenty-generator case studies that represent both centralised and distributed variable generation scenarios.