Analytical Model for Availability Assessment of Large-Scale Offshore Wind Farms Including Their Collector System

With the increase of offshore wind farm size, the use of classical analytical reliability methods becomes computationally intractable. This paper proposes a holistic approach combining multi-state Markov processes and the universal generating function for the availability assessment of radial large-scale offshore wind farms. The proposed model combines multi-state wind turbine output, wind turbine reliability, and inter-array cable reliability models to assess the wind farm output at the point of common coupling. A strategy is developed to split the network into its feeders while still accounting for the wind turbine output dependence, significantly reducing the computational burden. Although the failure rates of inter-array cables are low, their inclusion is pertinent given high repair times and impact on wind farm output given the radial topology of the collection system. A case study on the Anholt wind farm indicates the necessity of accounting for the collection system, showing a significant reduction of 12 % in generation ratio availability for a generation ratio criterion of 95 %.