Adapting to an uncontrollable energy resource: Research to improve the reliability of wind energy through accurate forecasting, efficient power collection, effective maintenance, and sustainable disposal
Wind energy is more sustainable than fossil fuel energy, but the reliability of wind energy must increase before a high penetration of wind into the energy portfolio can be achieved. In addition, the sustainability of wind energy can be improved and is important if wind turbines are to become more prevalent. Our research focuses on improving reliability in three important aspects: forecasting, power collection, and turbine operations and management (O&M). The uncertainty of wind forecasts is addressed by improving the numerical weather model depiction and evolution of the atmospheric boundary layer. Providing more accurate wind forecasts will allow electricity grid operators to better address the intermittency of wind power. For further improvement in wind energy conversion reliability, the use of a direct current (DC) collection system of series-connected turbines is modeled to determine the benefits in voltage and thermal stress reduction and ensuing improvement in component life. A reduction in the number of power electronic devices is possible with the DC system, eliminating converter stages and increasing system reliability. In order to improve the reliability of turbine operations and reduce turbine downtime, an auto image-based method is developed to monitor, compute, and define the type and severity of wind turbine blade surface flaws and recommend maintenance decisions. The method will eliminate human risk caused by on-tower inspection and predict future structural damage. Finally, we use life cycle assessment methodology to compare recycling methods for the spent turbine blades in order to reduce the environmental impacts in the end-of-life stage.
Ananth Iyer
Faculty: Project Co-PI
Can wind turbine blades be refurbished or re-manufactured instead of recycled ? Does appropriate maintenance increase the odds of re-manufacturing of blades ?
Ashley Mui
This is a great question. Turbine blades last around 20 years, but after this period, water damage, structural damage, and surface flaws cause the blades to be less aerodynamically effective and reliable. Appropriate maintenance can help extend the lifetime, but once a blade is damaged enough through operation, refurbishing will not produce a viable product.
The aerodynamic efficiency cannot be restored, and in an industry that relies heavily on maximum productivity for revenue, the use of refurbished blades is not economical. Furthermore, the structural damage may be so great that it is dangerous to reattach refurbished blades. Internal structural damage is difficult to repair, and the composite material, once weakened, cannot be re-strengthened. Blade strength is especially important because blade failure can be calamitous, as seen by prior incidents of blade detachment.
Huiyi Zhang
Graduate Student
Prof. Iyer, I think your question can be a good research project: re-manufacture end-of-life blades vs. recycle. It was considered in Europe and the key is which solution is more profitable. I have seen a section of an 18 year old rotor blade at Fraunhofer Institute and the color of the reinforced fiberglass near the leading edge became dark brown. I think it was caused by the leading edge erosion, which can be avoided with early inspection and repair. If rotor blades kept good condition after 20 years operation by carefully maintenance, an experts’ report is required to continue operating.
Carefully maintenance can prevent structural damages caused by early surface flaws. In this way, it “extends” the life time of rotor blades. Current inspection methods are expensive (rope man can cost up to $5000/person/day) and the uncertainty is high (hairline thickness cracks are easily missed with human eyes). The computer-based image processing method we addressed can detect hairline thickness cracks accurately and easily. It also can detect stress cracks and crazing (common early surface defects in the gel coat layer) promisingly. The method does not require specific setup angles of the field camera.
Paulette Clancy
Faculty: Project Co-PI
Of the four methods for waste blade disposal that you mention, which do you recommend overall based on the metrics you mention (toxicity, water use, etc) and how did you come to that decision?
Ashley Mui
From my analysis, it seems that either pyrolysis or cement co-incineration (incineration in the cement kiln and incorporating the ash into the cement) are the most sustainable methods of blade recycling. I recommend these because the environmental benefits provided through using the recyclate instead of manufacturing virgin materials are greater than the environmental impacts necessary to carry out these methods. There are avoided respiratory effects, avoided greenhouse gas emissions, avoided toxic emissions, and energy savings through these two methods. These are the only two methods that I considered which had benefits in all of the impact categories that I looked at.
Economic considerations must be taken into account as well – from initial analyses, it seems that cement co-incineration would be more viable because pyrolysis requires the construction of an entirely new facility, while cement co-incineration can be carried out in existing cement production facilities.
Paulette Clancy
Faculty: Project Co-PI
Thank you, Ashley; that was helpful.
Ranjit Koodali
Faculty: Project Co-PI
Could you briefly comment on the typical composition of the materials used in the wind turbine blades?
Huiyi Zhang
Graduate Student
Prof. Koodali, wind turbine blades are constructed of E-glass fabrics stitched with an organic yarn, then impregnated with a polyester/ epoxy matrix. Longer blades (~50 meters and above) also use carbon fiber to reduce total weight and to enhance its mechanical properties. The core material is typically foam/basswood. Finished blades are coated with a thickness between 0.3mm where loads are light to 0.6mm where loads are high gel coat, which is commonly based on epoxy or unsaturated polyester resin. The surface paint is typically white or light gray.
The inner structure of a wind turbine blade looks like an I-beam, where the shear web is mainly constructed of foam/boss wood reinforced with several layers of fiberglass and the spar cabs are purely reinforced fiberglass. Core materials also can be found near leading edge. The fibers have different orientations such as spar cabs use unidirectional fiberglass and the skin uses +45° and -45° fiberglass. Shear web and spar cabs are bonded together with a structural adhesive.
All of these materials are very sensitive to moisture, salt, and etc. Gel coat and paint are the major protection layers. Therefore, rotor blades surface inspection can prevent structural damages and “extend” the lifetime of wind turbine blades.
Ranjit Koodali
Faculty: Project Co-PI
Thank you for the excellent response.
Ian Harrison
Faculty: Project PI
Efficiencies for AC and DC power collection designs differ by ~12% at low wind speeds but the annual energy losses for the two approaches differ by only about 2%. Is that because most energy is produced at high wind speeds or are there other reasons? Could you outline how the wind speed specific efficiencies play in to the annual energy loss calculations?
Michael Johnson
Graduate Student
You are correct that the smaller difference in the energy yields are due to the fact that higher wind speeds result in higher power outputs thus adding more importance to the efficiencies at the higher wind speeds compared to the lower.
Since the power potential is related to the cube of the wind speed, the energy produced at the rated wind speed and above dwarfs the energy produced at the lower wind speeds. This means the efficiencies in the higher wind speed range are the most influential. It can be seen in the efficiency graph that the dc system dips below the efficiency of the ac system. Consequently, the energy production of the dc system is not as improved over the ac system as one might expect from the initial impression.
Ian Harrison
Faculty: Project PI
Interesting. So, it’s an average over the flux weighted translational energy distribution of the wind that is proportional to the windmill power that gets you to a wind velocity cubed dependence I presume. Nevertheless, on average, the DC coupled system is slightly more energy efficient than the AC and your other argument is that DC is simpler and possibly less prone to breakdown than an AC coupled system. Thanks!
Matthew Yates
Faculty: Project Co-PI
This may be a naive question, but why did you not compare landfill disposal to the other alternative uses of blade materials? Landfill disposal is the base case, and I presume it has negative impacts that can be quantitatively measured in comparison to other disposal or recycle methods?
Ashley Mui
This is a question that a lot of people have and one that I should have addressed more clearly. The used blades are inert material in the landfill. Since their composition is mainly fiberglass, few emissions associated with the impact categories that I considered (global warming potential, respiratory effects, ecotoxicity, eutrophication, and acidification) would be emitted by the very slow to decompose blades. Essentially, landfilling doesn’t have many harmful environmental effects besides being a waste of energy, materials, and landfill space. There is one category in Life Cycle Analysis used to quantify this type of waste – Abiotic Resource Depletion – and initial analyses did show that all recycling methods result in the salvage of abiotic resources in comparison to landfilling.
I decided that demonstrating that certain recycling methods have net environmental benefits (through the production of useful products from recycling) was enough to make those methods more desirable than landfilling, which has no environmental benefits nor costs. Essentially, this analysis is to make sure that there are recycling methods which make sense by ensuring that the efforts for recycling are outweighed by the ultimate rewards. Please let me know if you have any additional follow up questions on this issue!