MSc candidate
Project: Recycling of Wind Turbine Blades - Strength Prediction of Recycled Materials
zijian.wang [at] mail.mcgill.ca (Mail)
Supervisor: Larry Lessard
Recycling composite fibreglass from wind turbines is a complex problem that requires a sustainable solution. Any solution must be environmentally friendly and at the same time, economically viable. In this research, the fibreglass from end-of-life wind turbine blades is to be transformed into fibre-reinforced products that have consistent and predictable properties. These will take the form of recycled fibre-reinforced pellets for compression molding applications and recycled fibre-reinforced filaments for use in the 3D printing industry. Recycled fibres will be added into thermoplastic matrix (ABS or PLA) for 3D printing. In order to ensure the consistent, predictable properties, models have been developed that mimic the fibre-reinforced materials that are generated by the industrial recycling process.
Thus far, these models have been able to predict the stiffness properties of the recycled materials with sufficient accuracy using the representative volume element (RVE) method for use in finite element analysis models. However, the prediction of strength properties remains an important road block to the RVE method. Here, it is proposed that we use experimental data from strength tests performed on output recycled materials in order to build a strength prediction model. A data analytics method will be used to create predictable strength maps that can be accessed by the RVE models. Based on a sufficient number of strength tests, the resulting strength models should be able to interpolate any recycled material, resulting from various combinations of percentage and type of recycled reinforcing material and thermoplastic matrix used. The result will be essential to a model that predicts both stiffness and strength of a recycled material in order to generate predictable specifications for industries that want to know what they get when they recycle materials.