Lead Session Chair:
Stephan Barth, Managing Director, ForWind - Center for Wind Energy Research, Germany
Peter Jamieson (1) F P Panagiotis Chaviaropoulos (2) Spyros Voutsinas (3)
(1) University of Strathclyde, Glasgow, United Kingdom (2) Centre for Renewable Energy Sources, Athens, Greece (3) NTUA, Athens, Greece
Printer friendly version: Print
Download poster(0.86 MB)
Presenter's biographyBiographies are supplied directly by presenters at OFFSHORE 2015 and are published here unedited
Peter Jamieson has been a wind energy professional since 1980, responsible for wind turbine development and much involved in the design of the Howden wind turbines for a 26 MW wind farm erected in California in 1985. As senior principal engineer in Garrad Hassan from 1991 – 2013 he founded their Special Projects Department and since October 2009 has been employed as senior technology adviser in the Wind Energy Centre for Doctoral Training of Strathclyde University. His book “Innovation in Wind Turbine Design” (Wiley, 2010) reflects a career involvement in wind technology evaluation in development.
Large scale offshore wind energy systems - the multi-rotor solution
Work was recently completed on a 20 MW, multi-rotor system (MRS) within the European project, Innwind.EU. Continuing up-scaling of the conventional single turbine concept since the start of offshore deployment, is driven by the consideration of having fewer maintenance sites, fewer foundations and reduced extent of electrical interconnections per installed megawatt of wind farm capacity. This offsets the disadvantage in up-scaling to the turbine component costs associated with the square-cube law. The MRS avoids the penalties of the square–cube law and would appear to have major advantages in very many areas that affect cost of energy.
The initial approach was to direct specific research at potential show stoppers. The possibility of adverse aerodynamic interactions in a large array (45 turbines of 444 KW rating totalling 20 MW capacity) of closely spaced turbines was addressed by extensive simulation studies of NTUA using vortex and CFD modelling. The risk that the extensive space frame structure could be too heavy and expensive was explored with structure optimisation software developed by CRES. Such component design evaluation depends on adequate load prediction and special software of DNV GL Energy was used to evaluate turbulent wind loading on the 45 rotor array.
Main body of abstract
The MRS is to be fully assembled in dry dock and floated to site with a travelling crane attached to the structure for maintenance. Comparing power of n rotors with n x power of one, total power increased with number of rotors from 3% (7 rotors) to 8% (all 45 rotors). Relative to large single rotors, the MRS gained 2% to 5% energy in typical turbulent wind conditions.
Large reductions in structure loads of the MRS were associated with the independent operation of the small rotors. Although turbulence increases direct fatigue loading from wind on rotors or on structural members, paradoxically it reduces the loads passed from rotors into structure through averaging of inputs at variable frequency and phase. Large unbalanced loads on the structure, as may occur with a single blade of a large turbine stuck in pitch, are avoided.
The system comprises a closed-cell lattice tubular structure mounted on a barge type floater. CRES optimised selection of tubular members to meet design loads with minimum mass. The structure was found to be about 14% lighter than the tower of an equivalent single turbine. The preferred floater was a concrete barge design with estimated mass in the range 14-20000 tonne.
Levelised Cost of Energy (LCOE): Extensive studies were conducted using an LCOE model established in another task of Innwind.EU. Sensitivities to wake loss, aerodynamic loss, availability, O&M, turbine cost factors, MRS structure cost (modelled as jacket cost) and MRS power rating in relation to total swept area were explored.
A 30% reduction (Innwind.EU reference value 107€/MWh) in LCOE of the MRS was predicted (neither best nor worse scenarios). Sensitivity studies also indicated a 40 MW unit should be cost effective. Turbine technology is de-risked and market implementation can be faster. Consider by comparison development time, risks and impacts of a serial fault in turbines of 10 MW rating. With unit production of the rotors being in greater volume (factor of 10 to 20) unit reliability should increase. Single turbine faults compromise only a few % of capacity and can be ignored avoiding unscheduled maintenance in challenging sea states.
More advanced modelling capabilities have been developed to deal with the MRS design. It is hoped that awareness of a new promising solution for more cost effective wind energy production may be promoted.