04/07/2021
By Sokny Long

The Francis College of Engineer, Department of Mechanical Engineering, invites you to attend a doctoral dissertation defense by Stephen B. Johnson on "A High-Fidelity Design-Driven Techno-Economic Model for Wind Blade Manufacturing."

Doctoral Candidate: Stephen B. Johnson
Date: Friday, April 16, 2021
Time: 10 to 11 a.m. EST
Location: This will be a virtual defense via Zoom. Those interested in attending should contact the student at Stephen_johnson@student.uml.edu at least 24 hours prior to the defense to request access to the meeting.

Committee Chair (Advisor): James Sherwood, Dean of Engineering, Francis College of Engineering, UMass Lowell

Committee Members:

  • Joseph Hartman, Provost, UMass Lowell
  • Sammy Shina, Professor, Department of Mechanical Engineering, UMass Lowell
  • Stephen Nolet, Principal Engineer & Senior Director of Innovation & Technology, TPI Composites
  • Shridhar Nath, Wind Blade Technology Leader, GE Global Research (Retired)

Brief Abstract:

Since the early 2000s the steady increase in wind blade lengths has led to dramatic increases in wind turbine power and capacity factors. This increase in lengths along with other advances in wind turbine technology, e.g. loads management, gearbox reliability, and power conversion technology, have resulted in wind-generated electricity to be one of the lowest-cost forms of newly installed power generation. However, advances in wind blade design have not been matched with major changes in blade manufacturing techniques. This absence of major changes in manufacturing can be attributed in part to the lack of robust techno-economic modeling tools which can accurately capture the full product lifecycle costs of wind blade manufacturing and generate confidence in the return on investments (ROI) for capital expenditures on new technologies. Without such confidence, the culture of the wind blade manufacturing industry will continue to be one where only small incremental changes will be considered for implementation.

In this dissertation, existing costing methods are reviewed and a new techno-economic model for estimating wind blade costs is proposed. Data collected from three active wind blade makers is used to generate a complete suite of parametric process models to predict manufacturing labor hours. From surveys and dialog with industry experts, material costs, scrap rates, equipment costs, salaried staffing, annual reports, and other sources, a model is generated for the “mature” production cost. This mature cost is adjusted by analyzing cash flows over multiple years, and includes the impact of a learning curve ramp up from zero to the full production rate. A software tool is developed to interface seamlessly with existing detailed blade design and modeling tools such as NuMAD and CATIA™. To be user friendly, Excel worksheets are used for input and output data; however, all calculations are done in VBA which enables complex scenarios and alternatives to be examined. The tool incorporates a graphical user interface to allow user selection of alternative costing scenarios and data analysis algorithms.

The proposed techno-economic model is validated for its ability to predict the final weight of the blade, the number of man-hours required to make the blade and the overall cost of the blade and is used to investigate the ROI for new ideas for manufacturing. Validation of the process models in the proposed techno-economic model is achieved by comparison against alternative blades of various lengths that the team participants had experience fabricating and against models of publicly available designs for both process hours and total cost. The tool is subsequently used to evaluate four process-improvement ideas, and one of these ideas, “One-Step Close”, is selected for a full process trial. Detailed engineering and tooling fabrication is completed on shear web locating fixtures which subsequently enabled trials to confirm the validity of the model predictions and the value of the tool. The proposed techno-economic cost modeling tool is shown to deliver accurate and realistic estimates over prior models.

All interested students and faculty members are invited to attend the online defense via remote access.