Title: An energy IoT platform for real-time production and
delivery of wind power generation forecasts
Date: June 28, 2017
Bio: Pierre Huyn has over 30 years of research and advanced
development experience in data management, big data analytics, and software engineering. His current interest
is in big data architectures for IoT and deep learning for time series data in the domain of renewable energy.
Abstract: Power generation using renewable energy resources such as wind turbines has
grown increasingly popular. Because the underlying meteorological processes are highly
unpredictable, it has become important to be able to provide accurate power forecasts in real-time. In this talk we will describe an end-to-end IoT platform that enables SCADA sensor data
to be collected in real-time directly from a remote wind farm, securely and reliably transmitted
to cloud servers where data is analyzed to create forecasting models. These models are then
applied to the turbine sensor data stream to generate day-ahead power generation forecasts.
We will also describe the machine learning techniques used as the basis for the forecasting
models and our strategies to make the solution scalable for other big data applications.