Sensitivity analysis of offshore wind farm availability and operations & maintenance costs subject to uncertain input factors

"This thesis is jointly awarded by the University of Edinburgh, the University of Exeter and the University of Strathclyde".The deployment of offshore wind farms (OWFs) has increased in response to the threat of diminishing fossil fuel resources, climate change and the need for security of supply. The cost of offshore wind generation has not reached parity with established forms of electricity production. Operators need to simultaneously decrease the total project costs and increase energy yield to achieve a levelised cost of energy of £100/MWh. However, aspects of the Operations and Maintenance (O&M) remain uncertain, either through stochastic processes or through inexperience in the field. One way to handle uncertainty is to define how much the variance in these aspects affect the cost and availability. The thesis in hand introduces an O&M model and seeks to quantify the effects of uncertain inputs using complex sensitivity analysis methods.The sensitivity analysis is applied to an O&M computer simulation model for offshore wind that was developed prior to this project. Case study OWFs are identified to assess if the important factors are different when projects are comprised of a large number of wind turbine generators (WTGs) and are further offshore from the O&M hub port. The set of cases for the global sensitivity analysis comprises of three projects, to provide information applicable to the industry and demonstrate pertinence of sensitivity analysis on a case by case basis. A screening analysis, using the Morris method, is conducted to identify the most important factors on project cost and availability. This resulted in a list of twenty factors, relating to failure rates; duration of operations and information relating to vessels costs. An in-depth uncertainty analysis is conducted with the important factors to establish their distributions where possible. A global, variance-based sensitivity analysis, using the Sobol' method, is performed to quantify the effect on the variance of the two outputs.No single factor dominated the effect on O&M cost and availability for all cases. For each case, one to five factors contributed most to output variances. As an example, for a case of 30 WTGs located 20km offshore from the O&M hub port, the output variances are mainly a result of the change of number of crew transfer vessels and heavy lift vessel mobilisation time for nacelle component replacement. For an OWF with more WTGs, further from shore; the availability variance is dominated by more routine repair operations. Moreover, costs are largely dominated by WTG reliability. This work has confirmed that O&M costs are affected by the cost of deploying heavy-lift vessels even though only a small proportion of repairs require them. Significant factors are inconsistent across all the scenarios, supporting the conclusion that sensitivity analysis of each case is a necessary part of O&M costs and availability simulation. Using the most up-to-date information on current O&M practices, the analysis provides an indication of where to focus efforts for O&M cost reduction and improved availability.The deployment of offshore wind farms (OWFs) has increased in response to the threat of diminishing fossil fuel resources, climate change and the need for security of supply. The cost of offshore wind generation has not reached parity with established forms of electricity production. Operators need to simultaneously decrease the total project costs and increase energy yield to achieve a levelised cost of energy of £100/MWh. However, aspects of the Operations and Maintenance (O&M) remain uncertain, either through stochastic processes or through inexperience in the field. One way to handle uncertainty is to define how much the variance in these aspects affect the cost and availability. The thesis in hand introduces an O&M model and seeks to quantify the effects of uncertain inputs using complex sensitivity analysis methods.The sensitivity analysis is applied to an O&M computer simulation model for offshore wind that was developed prior to this project. Case study OWFs are identified to assess if the important factors are different when projects are comprised of a large number of wind turbine generators (WTGs) and are further offshore from the O&M hub port. The set of cases for the global sensitivity analysis comprises of three projects, to provide information applicable to the industry and demonstrate pertinence of sensitivity analysis on a case by case basis. A screening analysis, using the Morris method, is conducted to identify the most important factors on project cost and availability. This resulted in a list of twenty factors, relating to failure rates; duration of operations and information relating to vessels costs. An in-depth uncertainty analysis is conducted with the important factors to establish their distributions where possible. A global, variance-based sensitivity analysis, using the Sobol' method, is performed to quantify the effect on the variance of the two outputs.No single factor dominated the effect on O&M cost and availability for all cases. For each case, one to five factors contributed most to output variances. As an example, for a case of 30 WTGs located 20km offshore from the O&M hub port, the output variances are mainly a result of the change of number of crew transfer vessels and heavy lift vessel mobilisation time for nacelle component replacement. For an OWF with more WTGs, further from shore; the availability variance is dominated by more routine repair operations. Moreover, costs are largely dominated by WTG reliability. This work has confirmed that O&M costs are affected by the cost of deploying heavy-lift vessels even though only a small proportion of repairs require them. Significant factors are inconsistent across all the scenarios, supporting the conclusion that sensitivity analysis of each case is a necessary part of O&M costs and availability simulation. Using the most up-to-date information on current O&M practices, the analysis provides an indication of where to focus efforts for O&M cost reduction and improved availability