WindPACT Turbine Design Scaling Studies Technical Area 1-Composite Blades for 80- to 120-Meter Rotor [report]

D A Griffin
2001 unpublished
The primary objectives of the Blade-Scaling Study are to assess the scaling of current materials and manufacturing technologies for blades of 40 to 60 meters in length, and to develop scaling curves of estimated cost and mass for rotor blades in that size range. Approach We investigated the scaling of current materials and manufacturing technologies for wind turbine blades of 40 to 60 meters in length. Direct design calculations were used to construct a computational blade-scaling model, which
more » ... aling model, which was then used to calculate structural properties for a wide range of aerodynamic designs and rotor sizes. Industry manufacturing experience was used to develop cost estimates based on blade mass, surface area, and the duration of the assumed production run. The structural design model was also used to perform a series of parametric analyses. The results quantify the mass and cost savings possible for specific modifications to the baseline blade design, demonstrate the aerodynamic and structural trade-offs involved, and identify the constraints and practical limits to each modification. Conclusions and Results The scaling-model results were compared with mass data for current commercial blades. For a given blade design, the scaling model indicates that blade mass and costs scale as a near-cubic of rotor diameter. In contrast, commercial blade designs have maintained a scaling exponent closer to 2.4 for lengths ranging between 20 and 40 meters. Results from the scaling study indicate that: • To realize this lower scaling exponent on cost and mass has required significant evolution of the aerodynamic and structural designs. • Commercial blades at the upper end of the current size range are already pushing the limits of what can be achieved using conventional manufacturing methods and materials. • For even larger blades, avoiding a near-cubic mass increase will require basic changes in: − Materials, such as carbon or glass/carbon hybrids. − Manufacturing processes that can yield better mean properties and/or reduced property scatter through improvements in fiber alignment, compaction, and void reduction. The extent to which such improvements would result in lower blade masses may be constrained by blade stiffness requirements. − Load-mitigating rotor designs. For the scaling results presented in this report, the basic material and manufacturing process remained unchanged. As such, a reduction in mass will correspond to a reduction of production iii blade costs in the same proportion. However, this will not hold true for mass savings realized through changes in materials, process, and rotor design. In evaluating each such change, the implications on both mass and cost must be considered. As part of the cost analysis, it was shown that the "learning curve" required to achieve a mature production process has a meaningful effect on blade costs for the range of rotor sizes considered. A production rate of 200 megawatts (MW) per year implies 800 blades at 750 kilowatts (kW), but only 120 blades at 5 MW. Therefore, the cost penalty for initial production cycles has an increasing impact on the first-year production costs as rotor sizes increase, and a complete cost assessment depends on both annual production rates and the extent (number of years) of sustained production. The results of the scaling analysis are shown in the table below. Because of the assumed increase in hub height with rotor size and associated wind shear, the energy capture scales more rapidly than the rotor swept area (exponent of 2.22). Because the scaling was performed for a fixed-blade design, the mass exponent is slightly less than cubic. However, due to the impact of the learning-curve costs on the five-year production scenario, the blade cost exponent is slightly greater than cubic. The blade cost per installed kilowatt is shown to increase in a near-linear fashion with diameter, and the cost per energy capture scales as the diameter to an exponent of 0.82. The cost estimates presented above are appropriate for the assumptions and design modeled. It should be noted that the same general trend would result for any fixed-blade design that is scaled over the same range of ratings. To realize lower exponents on blade cost and mass requires evolution of the design and/or manufacturing process as the rotors become larger. iv
doi:10.2172/783406 fatcat:oqywzkiyrverxdk3bagc2ufxyq