Ensemble model aggregation using a computationally lightweight machine-learning model to forecast ocean waves
- Fearghal O'Donncha
- Yushan Zhang
- et al.
- 2019
- Journal of Marine Systems
I am a Research Scientist at IBM Research Europe - Ireland focused on applying technologies such as AI, hybrid cloud, accelerated discovery, and mathematical modelling, to develop innovative solutions in practical settings. This encompasses combining high fidelity simulations and low-cost surrogate models, integrating sensor data from a variety of IoT platforms, developing AI-based models that extract value from sensor or expert data, and optimizing these tools to a variety of HPC and cloud-based platforms. A key research focus is how sophisticated and human-centered Digital Twin technologies can be deployed to create virtual replicas of entities such as environmental systems, ports and shipping, and commercial real estate. Creating virtual representations allows us to rapidly explore many scenarios involving different management decisions and environmental conditions to explore optimal strategies. Ultimately, Digital Twin can serve to accelerate scientific discovery and identify innovative pathways that simultaneously contribute to net-zero objectives and sustainable industry developments.
Previous research interests include leading development of a decision support framework for aquaculture based on IoT sensors and petabyte-scale geospatial datasets (link), developing AI-based surrogate models for ocean processes (link), developing advection-diffusion model (with data assimilation) within an advanced 'Exascale' programming environment (http://www.allscale.eu/home), and implementing data assimilation techniques to improve ocean modelling capabilities (link).
I am a chartered engineer, hold a Ph.D, in Civil Engineering from National University of Ireland, Galway, and have published more than 40 scientific articles and 30 patents.