Asset maintenance simulation: The case-study of an offshore wind farm

Offshore Wind Farms are very expensive assets, not only to design and build, but also to operate and maintain. Their remote location, coupled with weather´s constraints and harsh marine environment, and the lack of adapted access equipment, make their maintenance complicated and expensive due to high repair costs, high transportations costs and painful production losses due to high downtime. More generally, the offshore wind technology is in a stage of development where its economic worthiness has still to be fully demonstrated, and numerous projects are largely supported by governmental subsidies. This lack of maturity is evident, for example, through the numerous reliability problems of wind turbines components. Both the inherent aspects of the offshore wind industry and the reliability problems it is facing contribute to the high Operation and Maintenance (O&M) costs and eventually hinder its competitiveness compared to other sources of energy. It is commonly assumed that up to 25 % of the cost of energy produced by offshore wind turbines is due to O&M activities, which is twice as expensive as onshore installations. This paper describes a methodology to optimize the maintenance strategy of offshore wind farms in order to minimize life cycle costs and maximize availability and performance. The ApmOptimizer, developed by BQR Reliability Engineering is the software tool used for modelling, reliability analysis, cost and performance estimations and maintenance optimization. For the purpose of the demonstration, a case study of a typical wind farm is defined. A detailed model is built that accounts for numerous operational, cost and maintenance parameters. The ApmOptimizer optimization modules are then implemented for optimizing the preventive maintenance, periodical inspections, condition-based maintenance, and spares levels. Starting from a baseline scenario, several optimization steps are investigated, following a typical RAM/LCC analysis assessment pr- cedure: Operational Availability, Operational Downtime, Lifecycle O&M Costs and produced Energy are used as the main metrics. The optimal maintenance strategy is obtained from an exact optimization process and not from Monte Carlo simulations that can provide only an approximate assessment for a user-inputted strategy. The study shows that by following several optimization steps, the wind farm performances can be significantly increased, to a level that makes the cost of produced energy more competitive. The influence of modelling parameters like failure distribution and degradation process is also investigated.