Peng Chen
Software-In-the-Loop method to predict the global dynamic responses of full-scale floating wind turbines by artificial neural network.
Email: p.chen8@ncl.ac.uk
Project supervisors
Project description
Floating offshore wind turbines (FOWTs) can generate electricity in water depths which aren't feasible for fixed-foundation turbines.
My research will accurately predict and validate the full-scale global dynamic responses of the FOWT system.
Methodology and objectives
I will combine DARwind and ANN to accurately simulate the experiment (SILANN).
I will conduct at least two model tests with different scale ratios. I will propose a method that can describe the scale relationship between two models by SILANN.
I will then apply the scale relationship of SILANN between model and prototype.
Results
I have conducted a literature review of the hybrid FOWT basin experiment.
I have run numerical simulation using the inhouse program DARwind, according to OC3 Hywind.
I have created the ANN model in MATLAB ANN tools to validate the numerical simulation results of the basin model test.
I have prepared a workshop paper for the 11th International Workshop on Ship and Marine Hydrodynamics.
I will build the Software-in-the-Loop concepts by extracting weight thresholds. I will transfer functions in ANN into DARwind simulation.
I will verify this method using code-to-code and code-to-experiment methods.
I will carry out a different scale basin model test to find a relationship between two models.
I will create extrapolation, which can apply the SLANN approach to full-scale FOWTs.
Interests
Floating wind, Renewable energy, Deep learning
Qualifications
MSc in Naval Architecture from Newcastle University