A team from Monash University’s Grid Innovation Hub, Worley, and Palisade Energy Ltd have embarked on a joint study to precisely predict wind and solar power generation using machine learning technology. The collaborators aim to securely integrate the power into the national electricity grid through the findings.

Launched in October 2018, the project is funded by the Australian Renewable Energy Agency. The initiative aims to provide wind and solar power generators with more accurate and reliable self-forecasting tools that are ahead by five minutes.

The five-minute ahead forecast mechanism is expected to reduce the frequency of poor dispatch and support a higher share of renewables in the market without compromising on the overall stability of the grid.

Dr. Christoph Bergmeir, from the Department of Data Science and AI at the Faculty of Information Technology at Monash University, led the machine-learning forecasting methodology development.

Stressing on the need for a reliable forecast technique for renewable energy, Dr. Bergmeir said, “Predicting short-term renewable energy generation is not an easy task. Renewable energy cannot be produced on demand, as it is bound to natural resources such as the wind and sun. Therefore, to achieve a stable network and enough power generation, we need a reliable short-term prediction method.”

The researchers added that introducing machine learning methodologies to this short-term forecasting process allows them to apply algorithms are based on historical time-series data, resulting in the accurate forecasting of wind and solar energy.

The innovative forecasting models are expected to apply the existing information around the application of machine learning and other AI technologies to wind and solar forecasting.

The forecasting models developed are based on machine learning algorithms drawing on internal supervisory control, and data acquisition feeds from the 130.8 MW Waterloo Wind Farm in South Australia and the 11 MW Ross River Solar Farm in Queensland.

The project had an overall budget of around $1 million.

The researchers stressed that the spending shows that forecasting accuracy could be improved for wind and solar generators by using best-practice machine learning techniques.

These forecasting models could be applied to all energy farms in Australia, noted the researchers, adding that applying the technology could bring down prices and open ways for hydro and other forms of clean energy. The experts, however, noted that more research is needed on the solar side.

Recently, Mercom had reported how distributed companies need distribution system operators to forecast renewables and manage the load.

Mercom had earlier reported why forecasting and scheduling weather patterns are imperative for stable and efficient grid management owing to the intermittent nature of wind and solar energy.