Structural capacity and the 20 MW wind turbine

• Published in: Proceedings of the Institution of Mechanical Engineers. Part A, Journal of power and energy Journal of power and energy, 2010-12, Vol.224 (8), p.1083-1115
• Main Author: Garvey, S D
• Format: Article
• Language: English
• Subjects: Design optimization; Mechanical engineering; Turbines; Wind power
• ISSN: 0957-6509; 2041-2967
• DOI: 10.1243/09576509JPE973

Abstract This article addresses a generic and fundamental question: given the specification for a structure, how can one assess or even express its capacity without first designing it? The question is motivated by the consideration of how wind turbines (offshore wind turbines in particular) may scale up further. An underlying premise is that the wind-turbine designs that have evolved from 50 kW to 5 MW may warrant re-examination to check whether scaling effects demand new approaches at larger scales. Since several in the industry are already setting sights on 8 MW and even 10 MW machines as extrapolations of current designs, this article considers the development of a 20 MW (232 m diameter) machine. The conventional three-bladed arrangement on a slender tower is examined for notional 5 and 20 MW machines for similar rated wind conditions. Initially, the article considers the structural capacity requirements for conventional designs at both 5 and 20 MW and how these can be reduced substantially – using a floating framework in place of a tower and by deploying cable bracing on the rotor. When tower bending and the non-productive blade bending moments have thus been addressed, the turbine torque itself emerges as a major cause of cost. The turbine torque directly influences the costs of blades, pitch and yaw bearings, direct-drive generators (if present), gearbox (if present), and main shaft, and it is confirmed that in all cases, the total amounts of material required are at least proportional to the turbine torque for conventional designs. An alternative wind-turbine configuration is examined. In this, gravity is co-opted as an ally rather than an enemy, and power conversion takes place entirely within the rotor by allowing gravity to move bodies within the rotor. The concept is illustrated for a 20 MW machine and shown to be entirely workable in principle. A strong feature of this proposed new machine configuration is that the costs are almost exclusively structural and can be quantified in terms of the uniform currency of structural capacity. The scalability of this concept becomes clear by observing that the mean velocities of the moveable masses relative to the main rotor structure are virtually independent of machine scale.