Clouds all have one thing in common: they affect the amount of energy solar panels produce. So, what is solar forecasting and why is it so important?


Solar forecasting will help us understand the impact of clouds on solar panels.

We’re working to produce a solar forecast from multiple different models, an ‘ensemble’ of forecasting models.

Are you a cloud watcher? Our scientists have been watching the clouds too.

Cumulus, morning glory or mammatus – all clouds have one thing in common. They affect the amount of energy that solar panels produce.

Whether it’s a big solar farm or your rooftop solar, the amount of energy generated by photovoltaic (PV) panels can drop quickly when clouds shade the sun.

More clouds mean less solar energy. It sounds simple, right? But it gets tricky quickly. For solar farms, cloud cover can cause fast changes in the energy they send to the grid. Solar farm operators are now responsible for providing the Australian Energy Market Operator (AEMO) with forecasts of their output into the grid and the central dispatch system that controls how much energy each generator puts out every five minutes.

AEMO needs these forecasts to balance energy supply and energy use. If these don’t balance, then the grid voltage and frequency will fluctuate.

Solar forecasting with blue sky thinking

Our scientists are working on a better way to predict cloud cover.

When will clouds sweep across the horizon and cover the panels on a solar farm? We’re answering this question by combining several forecast models. It’s called the Solar Power Ensemble Forecaster.

The system includes data from the Sky Imaging Camera. The camera takes photos of the entire sky every 10 seconds. Using the captured images, clouds can be tracked and their path predicted.

It’s not magic, it’s science

We’ve set the system up at five solar farms in New South Wales, Queensland and Victoria. It’s already providing five-minute forecasts that significantly outperform the older forecasting system.

Chris Knight is the group leader of our engineering team working on grids and energy efficiency systems.

“This project will help move people away from the thought that renewable energy is unreliable,” Chris said.

“Renewable energy is not unreliable, but it can be unpredictable. Models like this help us improve predictions of how much renewable energy production. Basically, this technology allows us to predict the future.”

Things are looking up

Renewable energy is providing increasingly more energy to the grid in Australia. But grid operators must carefully manage the balance between the generation and consumption of energy. By managing this energy variation, they can make the best use of abundant renewable energy.

For a solar farm, this provides greater grid stability, higher revenue and better use of what sun is available at any one time.

And for the rest of Australia, better short-term solar energy forecasts mean lower-emissions, cheaper energy and a more stable electricity grid. Win, win, win!

The Solar Power Ensemble Forecaster is an ARENA-funded project. Newcastle company Industrial Monitoring and Control is leading the project with our team, University of New South Wales and University of South Australia.

For the full version of this article head on over to our ECOS blog.


  1. Good article, but could you please tell us cloud enthusiasts what classification the cloud formation is in the photo.

    1. Hi Sarah, thanks for your comment. These are mammatus clouds.

      Team CSIRO

  2. Good to see the technology moving fast enough to support demand/supply balancing in small enough increments to put the concerns about fluctuating capacity to rest. Not mentioned in the article but presumably this would tie into using pumping to storage reservoirs, large scale battery arrays (Vanadium redox maybe?) or just brute force capacitors as Don Hampshire mentioned. The key to replacing fossil fuel generation with renewables is going to be the buffering capabilities of the delivery & storage system, and predicting supply makes managing whatever buffering tech is in place a whole lot easier. As consumers we should not be able to detect any fluctuations in electricity supply, no matter how it is being generated at any moment.

  3. Rather than play dramatic word meanings,’un’ this ‘un’ that, I prefer ‘solar fluctuates’.
    One thing I cannot recall hearing of with solar is capacitors. Its outside my knowledge except I recall the SEC importing an enormous capacitor for Yallourn which was brown coal steam electricity. It did not fluctuate but demand did.

Commenting on this post has been disabled.