Optimization Approach for Wind Farm Cable Layout Based on Iterative Self-Organizing Data Analysis Technique Algorithm and Swap Sequence Based Particle Swarm Algorithms

Abstract

Abstract In the fast evolving and increasingly competitive wind energy market, to minimize initial capex and maximize production profits are essential when designing a wind farm. The electrical cable accounts for around 10% of the total capex, reasonable layout of cable connections might induce significant reduction of electrical and construction cost. In this paper, the Fuzzy C-means (FCM) clustering algorithm is adopted to determine the location of the substation(s) and the ISODATA (Iterative Self-Organizing Data Analysis Technique Algorithm) is employed to divide the wind farm into finite sub-domains, effectively reducing the complexity of the NP-hard (Non-deterministic polynomial-time hard) problem of cable layout optimization. The optimal cable layout in each sub-domain is obtained through the swap sequence based particle swarm optimization (SSPSO) and practical electrical constraints, while the objective function is to minimize the total cable costs, including material cost and cable laying cost. Current study highlights both applicability of clustering algorithm and particle swarm optimization to the cable layout problem in wind farm.