This study conducted by NREL engineers, investigates the effect of tip-speed constraints on system levelized cost-of-energy (LCOE). The results indicated that a change in maximum tip-speed from 80 to 100 m/s could produce a 32% decrease in gearbox weight (a 33% reduction in cost), which would result in an overall reduction of 1 to 9% in system LCOE depending on the design approach. Three 100-m/s design cases were considered: a low tip-speed ratio/high-solidity rotor design, a high tip-speed ratio/low-solidity rotor design, and a flexible blade design in which a high tip-speed ratio was used along with removing the tip-deflection constraint on the rotor design. In all three cases, the significant reduction in gearbox weight caused by the higher tip-speed and lower overall gear ratio was counterbalanced by increased weights for the rotor and/or other drivetrain components and tower.
As a result, the increased costs of either the rotor or drivetrain components offset the overall reduction in turbine costs from downsizing the gearbox. Other system costs were not significantly affected, whereas energy production was slightly reduced in the 100 m/s, high-solidity case and increased in the low-solidity case. This situation resulted in system cost-of-energy reductions moving from the 80-m/s design to the 100-m/s designs of 1.5% for the high tip-speed ratio and 5.5% for the final flexible case (the latter result is optimistic because the impact of deflection of the flexible blade on power production was not modeled). The low tip-speed ratio case actually resulted in a cost of energy increase of 2.1%. Overall, the results demonstrated that there is a trade-off in system design between the maximum tip-speed and the overall wind plant cost-of-energy but also that there are several design trade-offs and design constraints that can limit the benefits of higher tip-speed designs.
Scope
Land-based wind project development has historically limited turbine designs to operate at blade-tip-velocities in the range of 75 to 80-m/s. The constraint arises from blade-tip aero-acoustic noise generation. The turbine system sound power levels are usually dominated by blade-tip noise when appropriate measures have been taken to mitigate sound emissions and audible tones from the tower head machinery within the nacelle and the power electronic converters often located within the tower base.
The study looks at five overall turbine configurations:
- A baseline 5-MW reference turbine Jonkman et al. (Feb 2009) with maximum tip-speed of 80-m/s
- An optimized version with the same 80-m/s tip-speed design constraint
- An optimized design at 100-m/s maximum tip-speed with a high-solidity rotor
- An optimized design at 100-m/s maximum tip-speed with a low-solidity rotor
- An optimized design at 100-m/s maximum tip-speed with a low-solidity rotor where the tip-deflection constraint has been removed (as a proxy for a machine that would operate downwind).
In each of the design cases, a sequential optimization was performed to design the turbine. The rotor was optimized first, then the hub and drivetrain components, and finally the tower. Each turbine design was then used in turbine aero-acoustic noise and overall system cost analyses.
Key results
The trade-off between maximum allowable tip-speed constraints on turbine design and overall system cost-of-energy was in the range of a few to several percent. The reductions in cost-of-energy would have been more, but the difficulty in packaging sufficient structural strength and stiffness into a lower-solidity rotor, as a result of the higher tip-speed, limited the reductions that could be obtained. Considering only aerodynamics, the increased tip-speed would result in a lower-solidity rotor operating at a higher tip-speed ratio with identical power performance to the reduced tip-speed case. The non-torque loads would be the same, whereas the torque would be reduced. Thus, the system would have a lighter blade and gearbox, and therefore a lighter tower. However, the strength and stiffness constraints for the rotor require that a lower-solidity blade must have increased structural weight. Or alternatively, moving to a higher-solidity design to reduce blade weight, which is accompanied by a reduced tip-speed ratio and power production and increased non-torque loads on the drivetrain. The different approaches to rotor design demonstrated the trade-offs in system design between rotor cost and the rest of the turbine costs as well as power production.
Gearbox sizing for different speed ratio requirements driven by different maximum allowable tip speed constraints.
In the case of the 100-m/s-high-solidity design, the increased rotor thrust loads and blade root bending moments demand a heavier hub, larger low-speed shaft, bigger main bearings, and overall heavier nacelle bedplate. The higher aerodynamic loads and heavier rotor-nacelle assembly (RNA) results in a heavier tower. This configuration did provide lower torque input to the gearbox that reduced overall gearbox size, but the increased loads almost entirely offset the reduced weight of the gearbox so that the overall nacelle weight actually increased slightly from the 80 m/s optimized design. The cost of the nacelle, on the other hand, was reduced by 6%. Overall turbine costs were reduced only 3% from the optimized 80-m/s case. Because of the decrease in energy production moving from the 80 m/s optimized design to the 100 m/s high-solidity design, the cost-of-energy actually increased by 1.3%. This case illustrated the need for a more integrated approach to system design considering overall cost-of-energy as the objective.
It was determined that moving to a lower-solidity, higher tip-speed ratio rotor configuration might improve the system cost-of-energy by allowing for higher energy production and reducing non-torque loads on the drivetrain. Moving to a 100 m/s, low-solidity design did result in a cost-of-energy reduction from the 80 m/s optimized case of 1.5%. The reduction was primarily because of the fact that the new design had thrust loads and blade root bending moments comparable to the 80-m/s design so that the nacelle components were similar in size to the 80-m/s case. This result coupled with the large reduction in gearbox size led to a nacelle that was 2% lighter than the 80-m/s optimized design and a nacelle cost that was 7% less than the 80-m/s optimized design; however, structural modifications were needed for the lower-solidity blade to meet deflection constraints and strength requirements so that the new blade was 9.3% heavier than the 80-m/s optimized blade. The added cost of the blade meant that overall turbine cost reductions were only 2%. However, the high-solidity design had improved aerodynamic efficiency and power capture, resulting in a plant energy production that was higher than the 80-m/s optimized case and an overall reduction in cost-of-energy.
As a final analysis, the 100-m/s-low-solidity design was run in a “flexible” configuration, in which the tip-deflection constraint was removed to mimic a downwind rotor configuration. This approach resulted in lower thrust loads and an overall nacelle that was 5% lighter than the 80-m/s design case. In addition, the removal of the tip-deflection constraint enabled a much more flexible blade design with less structural reinforcement and a mass that was 9.3% lighter than the 80-m/s blade design. The result was a turbine cost that was 7% lower than the 80-m/s case. One caveat here is that the flexible downwind rotor would operate with significant deflection, which could reduce the overall energy production. This effect was not captured in the subsequent annual energy production (AEP) analysis and thus, AEP remained at the same level as the first 100 m/s, high tip-speed ratio case. Thus, the resulting overall cost-of-energy reduction of 5.5% from the 80-m/s optimized design case is likely optimistic. Future work should model the effect of blade flexibility on energy capture.
Analysis limitations and future work
The following are the main limitations of the current study:
- Sequential optimization can only capture the impacts of coupled interactions in system design if it is used in an iterative process which can be inefficient. The rotor design is selected without complete information about the downstream effects on other subsystems and overall cost-of-energy. It is possible that an integrated system optimization approach may improve a sequential process, even if the models used are of a lower fidelity. A system optimization may encounter designs that reduce LCOE by more efficiently balancing trade-offs in energy production and cost. If a sequential optimization is used in the future, it is critical to balance the objectives of the rotor design with the rest of the system design.
- Detailed design of the controller for each of the cases was outside the scope of the study. Sophisticated models were used for the rotor design but did not include a suboptimization of the controller design to mitigate against higher loads in the 100-m/s tip-speed cases.
- Higher fidelity and more detailed design-based models are needed for many of the models used. The cost models represent the lowest-fidelity/highest uncertainty models used. The turbine cost model needs updating with more data representative of current turbine technologies and costs. The plant cost models for operational expenditures are overly simplistic and the balance-of-station cost model is not sufficiently detailed to reflect a design-based/bottom-up approach. Improvements in the accuracy and detail of these models are needed to further explore the impact of tip-speed constraints on system cost.
- Only a single turbine and plant configuration with just two different maximum tip-velocities were considered in the analysis. Future work should assess the impacts of differing site conditions as well as size and class of the turbine design on the effects of tip-speed on system cost-of-energy. This assessment should also be done over a range of tip-velocities. Performing analysis over a range of turbines, sites, and maximum tip-velocities will provide insight as to of potential benefits.
In addition to the caveats above, significant changes to and innovation in the turbine design was outside the scope of this study. Potential innovations in each subsystem could take advantage of the higher tip-speed constraint.
- For the rotor, the use of novel materials in the blade design (such as carbon throughout the blade or only near the tip where deflection is the highest) may mitigate against the tip-deflection constraint impacts on the blade weight and cost. However, to explore the use of novel materials, a detailed, design-based blade cost model is necessary to account for the materials and manufacturing costs. Future work could also consider using different airfoil families, such as flat-back airfoils or even thicker airfoils, which may have decreased aerodynamic efficiency but could mitigate against the increased weight of the blades to meet deflection and strength requirements. Another potential research opportunity would be to shift from a three-bladed to a two-bladed rotor configuration
- For the drivetrain, transitioning to higher tip-velocities might create opportunities for novel drivetrain configurations (for example, removing one of the gearbox stages or moving to a medium-speed gearbox/generator or direct-drive configuration).
- For the tower, many innovations are possible through the use of new materials and configurations that would avoid current constraints on design because of logistics considerations related to transportation and installation.
Despite the study limitations, it appears that there is a cost/noise trade-off in the design of wind energy systems. Future work should be conducted to obtain additional insight into this relationship and quantify the trade-off to better understand the potential cost reductions and the need for research on aero-acoustic noise mitigation strategies.This report is available at no cost from the National Renewable Energy Laboratory at
www.nrel.gov/publications
http://www.nrel.gov/docs/fy15osti/61726.pdf
Filed Under: Gearboxes, News, Turbines
