Monash University’s Grid Innovation Hub is collaborating with Palisade Energy and Worley, an engineering servicing company, to predict wind and solar power generation with machine learning technology, PowerPredict.
The PowerPredict project, launched in 2018, is funded by the Australian Renewable Energy Agency (ARENA). It aims to provide wind and solar power generators across Australia with more accurate and reliable self-forecasting tools, securely integrated into the national electricity grid.
The PowerPredict solution developed by Monash and Worley will provide a five-minute ahead forecast and reduce the frequency of poor dispatch. This will support a higher share of renewables in the market without compromising on overall grid stability.
The machine learning forecasting tool was developed by the Monash Department of Data Science and AI’s Dr Christoph Bergmeir. This was in collaboration with the Monash Business School’s Department of Econometrics and Business statistics.
“Predicting short-term renewable energy generation is not an easy task,” Bergmeir said.
“Renewable energy cannot be produced on demand, as it is bound to natural resources such as the wind and sun. Therefore, in order to achieve a stable network and enough power generation, we need a reliable short-term prediction method.
“By introducing machine learning methodologies to this short-term forecasting process we’re able to apply algorithms that are trained on historical time series data, resulting in the accurate forecasting of wind and solar energy,” he said.
The PowerPredict technology will increase renewable energy penetration in the grid, as a result of improved dispatchability and reduced Frequency Control Ancillary Services (FCAS) payments.
“Our forecasting solution provides immediate value to our existing renewables customers as they target lower FCAS charges,” Worley Data Science Customer Solutions global vice president Denis Marshment said.
“And with PowerPredict officially launched renewable generators, in Australia and internationally, can benefit from our power forecasting technology.”
The research and development around these models are anticipated to advance knowledge on applying machine learning and AI technologies to wind and solar forecasting.
“Natural variations in weather makes it difficult for renewable generators to accurately forecast their short-term power generation levels and this impacts grid stability,” Marshment said.
“In 2020 alone, inaccurate power predictions cost Australian generators $210 million, so using machine learning algorithms to see five minutes into the future is incredibly valuable.
“Our forecasting algorithms achieved a 45 per cent improvement in our customers’ power output predictions.”
The successful PowerPredict project demonstrated that further enhancing best practice machine learning techniques will lead to the forecasting model’s use in energy farms Australia-wide.
With the potential to lower energy prices, the technology can also potentially open avenues for hydro and other forms of clean energy, according to Bergmeir.
“If renewable generators can lower their causer pays factors they can produce electricity cheaper, and eventually that saving could be passed on to the customers,” he said.
“It would also make renewables more competitive, which is also a desirable outcome.”
“This is an exciting and timely application of one of the Monash Energy Institute’s and Grid Innovation Hub’s star computer science and AI teams,” Monash Energy Institute director Professor Ariel Liebman said.
“This project shows how industry, represented by our visionary partners Worley, and academia can create real impact together – both commercially and in contributing to the global effort to stop climate change.”
Find out more on how PowerPredict works here.