Improved forecasting for Australian wind farms
Wind is a fantastic source of renewable energy, but it is variable and can be hard to forecast. As renewables become dominant, we’re presented with the challenge of more accurately forecasting renewable energy production. We need to maintain a secure and efficient electricity system over the energy transition.
In March 2019, Meridian Energy Australia (MEA) and the University of Melbourne won support from the Australian Renewable Energy Agency (ARENA) to develop improved, real-time forecasts of wind farm power generation. Until recently, the Australian Energy Market Operator (AEMO) forecasted the generation from all the wind and solar farms across the National Electricity Market (NEM). Using this ARENA support, the team from MEA and the University have developed and implemented their own wind forecasting system and integrated this into AEMO’s real-time processes for managing the NEM.
More accurate forecasting of renewables helps their uptake in the NEM. It also supports AEMO to maintain system security while reducing total emissions and costs. Asset owners and operators are also well placed to understand how an asset should perform, and the NEM rewards wind farm owners financially for improved wind farm forecasting.
The team from MEA and the University have also created an interactive website for the wider community. This website is open for anyone to access, and is an educational resource for primary, secondary and tertiary students as well as the general public. Our hope is that this site builds community understanding of renewables and the prospects of a secure and decarbonised energy system.
This project has been a close collaboration between MEA and the University of Melbourne.
MEA has provided technical support and project leadership, implementing the necessary systems and interfaces for the forecast to be accepted by AEMO on an ongoing basis. The University’s efforts are truly interdisciplinary, bringing together expertise in meteorology, wind farm modelling and machine learning. Dr. Claire Vincent, Dr. Grant Skidmore and Mr. Mathieu Pichault are playing key roles in this ongoing project. Their efforts are complemented by other research fellows and students, who will contribute further research into more advanced forecasting systems.
“Meridian has been a wonderful partner of the Melbourne Energy Institute over several years.” says MEI Director, Prof. Michael Brear. “This project has already demonstrated that this partnership can produce state-of-the-art energy technology. We are now forging ahead and making these wind farm forecasts even better.”