A Spatial Information-Based Self-Adaptive Differential Evolution for Distribution Substations Location and Sizing

The location and sizing of distribution substations integrated with urban spatial information is a optimization problem which primary characteristics are the existence of many suboptimal cost peaks and the solution space is on a discrete floating-point variables rage with some unfeasible area constraints. To solve this problem, a spatial information based parameter self-adaptive schema of DE (ISADE) is proposed. The populations are formulated by spatial coordinates of candidate substation locations which chosen randomly and attempt to cover the entire planning area. A randomized self-adaptive scheme for DE mutation weighting factor F is used to increase the ability of searching out the global solution. The effectiveness of the proposed method is then tested on two practical substation planning areas of Beijing Electric Power Corporation and the results are compared with PSO and standard DE.