An improved self-adaptive harmony search algorithm for distribution system planning

Abstract

This paper proposes an improved self-adaptive harmony search algorithm (ISHSA) for distribution system planning (DSP). Harmony search algorithm (HSA) is an intelligent swarm algorithm based on a mimic of musical improvisation process inspired by the musicians try to find a better harmony through repeatedly exercise, which has been applied in various electrical engineering problems. However, it suffers from searching efficacy tardiness when applied in large-scale nonlinear mixed integer programming (NMIP) problem, such as DSP. In this work, a novel encoding methodology is proposed to increase search efficiency as well as handle non-radial structure problem particularly in DSP. In addition, a self-adaptive expedition strategy is also developed so as to make the algorithm less sensitive to initial parameters setting. A 10kV electrical distribution system feeding an urban area with 25-node is presented for a real case application. The results demonstrate the superiority of the proposed ISHSA in search efficiency on comparison with the standard HAS, as well as simulated annealing algorithm.