Forecasting High-Frequency Spatio-Temporal Wind Power with Dimensionally Reduced Echo State Networks

Event Start
Event End
Location
https://kaust.zoom.us/j/99005716923

Abstract

Fast and accurate hourly forecasts of wind speed and power are crucial in quantifying and planning the energy budget in the electric grid. Modeling wind at a high resolution brings forth considerable challenges given its turbulent and highly nonlinear dynamics. In developing countries where wind farms over a large domain are currently under construction or consideration, this is even more challenging given the necessity of modeling wind over space as well. In this talk, we introduce the proposed machine learning approach to model the nonlinear hourly wind dynamics in Saudi Arabia with a domain-specific choice of knots to reduce spatial dimensionality. Our results show that for locations highlighted as wind abundant by a previous work, our approach results in a 11% improvement in the two-hours-ahead forecasted power against operational standards in the wind energy sector, yielding a saving of nearly one million US dollars over a year under current market prices in Saudi Arabia.

Brief Biography

Huang Huang is a research scientist at the Spatio-Temporal Statistics and Data Science (STSDS) research group, King Abdullah University of Science and Technology (KAUST), Saudi Arabia. Before working at KAUST, Huang was a postdoctoral fellow at the National Center for Atmospheric Research (NCAR), the Statistical and Applied Mathematical Sciences Institute (SAMSI), and Duke University, focusing on spatio-temporal statistical inference for large climate data sets. He received his Ph.D. in Statistics in 2017 from KAUST and M.S. and B.S. in Mathematics in 2014 and 2011 from Fudan University, China. His research interests include spatio-temporal statistics, functional data analysis, Bayesian modeling, machine learning, and High-Performance Computing (HPC) for large climate data.