Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Peter Sandborn (1) F P Xin Lei (1) Roozbeh Bakhshi (1) Amir Kashani-Pour (1)
(1) University of Maryland, College Park, United States of America
Printer friendly version: Print
Download poster(0.78 MB)
Presenter's biographyBiographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited
Dr. Sandborn is a Professor in the Department of Mechanical Engineering at the University of Maryland and member of the Center for Advanced Life Cycle Engineering (CALCE). Dr. Sandborn’s research interests include: technology obsolescence management, prognostics and health management, technology tradeoff analysis, parts selection and management, and system life-cycle and risk economics. He has done work on return on investment, design for availability, and maintenance optimization for aerospace and control systems, and wind turbines. He is a Fellow of the IEEE and ASME
Development of a maintenance option model to optimize offshore wind farm O&M
Offshore wind farms are capital intensive projects whose economic viability depends on many things including the successful long-term management of the turbines. The prediction and optimization of maintenance activities provides a significant opportunity for wind energy operation and maintenance (O&M) cost reduction. This paper introduces the concept of maintenance options applied to single wind turbines and wind farms managed via complex power purchase agreements (PPAs). A PPA is a contract between wind power seller and buyer that defines the schedule for energy delivery, contract rate, excess rate for over-delivery, and penalties for under-delivery.
Maintenance practice for turbines consists of preventive scheduled maintenance, preventive maintenance based on condition (CBM - condition based maintenance) and corrective maintenance. For a turbine, a maintenance option is created by the incorporation of CBM into subsystems such that a remaining useful life (RUL) is predicted as the subsystem’s health degrades. The price of the option is the cost of implementing and supporting the CBM hardware and software; the option is exercised when preventive maintenance is performed (based on the RUL) before subsystem or turbine failure; and the option expires if maintenance is not performed prior to failure.
Main body of abstract
Time-history paths of stochastic future cost avoidance and cumulative revenue are simulated with the inclusion of uncertainties in wind and the forecasted RUL. Using a simulation-based real options analysis (ROA), possible maintenance dates (governed by when preventive maintenance can actually be performed) after an RUL prediction are estimated and analyzed, and the optimum maintenance date that maximizes the value of the maintenance option can be determined.
The ROA has been extended to wind farms with multiple turbines. If ROA is performed on each turbine with one or more subsystems separately indicating a finite RUL, and the option values are summed for all turbines, then the analysis implicitly assumes the optimum maintenance dates for all turbines are independent. If the simulated time-history paths of cost avoidance and cumulative revenue are combined for multiple turbines, and ROA is performed on the accumulation, then the analysis implicitly assumes that all turbines (i.e., all turbines with one or more subsystems indicating a finite RUL) are maintained together at the selected optimum maintenance date. Using these accumulation schemes, the optimum maintenance dates for a wind farm can be determined.
The key to modeling maintenance options for a wind farm subject to a PPA is that the cumulative revenue and cost avoidance calculation (used in the ROA analysis) for a single turbine (or subsystem) depends on the operational state of all the other turbines and the amount of energy to be delivered by the whole farm.
The maintenance options approach demonstrates that the optimum maintenance plan for the turbines in a wind farm subject to a PPA is not the same as the optimum maintenance plan for individual turbines managed in isolation. Researchers have addressed the optimization of maintenance for wind farms; however, in previous works CBM generally assumes a constant threshold for maintenance rather than a maintenance threshold that depends on state of health of the entire wind farm and the specific requirements of the complex PPA that the farm is subject to.
The goal of the analysis described is to find the optimum maintenance schedule for wind farms that are subject to a complex PPA that may include variable rates and penalties. Uncertainties in wind and the accuracy of the RULs forecasted by the CBM are included.