An improved harmony search algorithm for solving day-ahead dispatch optimization problems of integrated energy systems considering time-series constraints

Abstract

Day-ahead dispatch optimization, as an essential dynamic function required for the optimum energy management of integrated energy systems, is suffered from the curse of dimensionality and high computational cost when it faces with nonlinear, non-convex, and in particular high-dimensional situations. The critical one factor contributing to these phenomena is the dynamic constraints which are coupled in a time-series way. To this end, a new search strategy is proposed in this paper to handle the time-series constraints in harmony search (HS) based day-ahead dispatch problems, called as conductor harmony search (CHS). The CHS introduces a state matrix to record the values of time-series constraints. Moreover, when the variable values at a certain moment are altered during the improvement process, the variables at the coming moment are modulated to enable the corresponding values in state matrix remain unaffected. In contrast to existing constraint handling methods, the CHS guarantees the least impact on the dynamic boundaries of feasible regions in the remaining moments. Simulations reveal that the proposed CHS is effective and powerful to solve the day-ahead dispatch problem by comparing with some existing algorithms.