In the past decade, inaccurate short-term weather forecasts for wind and solar power generation in Australia have cost the industry close to AUD 5 million ($3.58 million). The developers could do better if they have accurate forecasting of the cloud cover for instance. To help with the topic, a research team headed by the Professor of Environmental Mathematics of the University of South Australia, John Boland is working on a $1.2 million project to design a ''world-first, short-term statistical forecasting model'. It is expected to accurately predict weather conditions from 5 minutes up to 1 to 2 hours.
The research team will also have colleagues from Commonwealth Scientific and Industrial Research Organization (CSIRO), the University of New South Wales, and Genex Power Ltd to work on the project for the next 18 months.
"Clouds can move and form very quickly, creating complex atmospheric layers which often move in different directions. The existing forecasting systems for wind and solar are designed for longer-term timeframes and have led to multiple issues over the years. This highlights the need for reliable short-term forecasts to provide confidence to both renewable generators and the entire industry," explained Prof Boland.
The team will implement a short-term solar forecasting system on 5 operational solar power projects from Queensland to Victoria taking into consideration sites that experience changeable weather conditions. They plan to use skyward-facing cameras with machine vision algorithms to track and predict cloud motion, satellite imaging, statistical models and power conversion models to achieve the accuracy.
If successful this Australian Renewable Energy Agency (ARENA) funded project could help the industry reduce both grid instability and cost, and would save renewable energy generators from being 'unfairly penalised and the benefits will be passed on to consumers', according to Prof Boland.