Mr Jialun Chen¹, Dr Wenhua Zhao¹, Dr Ian  Milne¹, Prof Paul H. Taylor¹, Dr David Gunawan²

¹The University of Western Australia, Perth, Australia, ²The University of Wollongong, Wollongong, Australia

Wave induced vessel motions are of great concern in offshore operations, such as side-by-side offloading of LNG, side-by-side offloading of pipes, active control of wave energy converters, floating wind turbines, vessels parked in a port, etc. Forward forecasting of the vessel motions, even if a few minutes in advance, is of value for decision making, e.g. emergency disconnection of loading arms and mooring lines to avoid damage. An Auto-Regressive model is proposed to predict surface waves and vessel motions based on the measured time sequences at a particular location. Techniques have been explored to maximise the prediction horizon of the wave surface elevations and the vessel motions. It is found that band-pass filtering to measured data can improve the forecasting accuracy and capability. Adopting this, we have shown that the forecasting horizon can be significantly improved, compared to what could be achieved based on theoretical analysis, showing the power of data-driven model. Wave basin test data have been adopted to further support the observations.

Biography:

Jialun has completed a Master of Professional Engineering (Civil) in 2020 at the University of Western Australia. Currently, he is in the second year of PhD study in ocean engineering. Jialun’s research focuses on surface wave and wave-induced vessel motion prediction, with application to the oil & gas industry and renewable energy sector. In particular, he is interested in data-driven approaches, i.e. machine learning to capture the complex physics in offshore engineering.