Accurate forecasting of cloud cover will help solar generators work in unison with retailers and networks to ensure a more stable grid, writes Solcast chief technology officer Dr Nick Engerer (pictured).


If you’re a proud solar owner (and 25% of Australians are), you may have taken an interest in tracking the performance of your system’s energy generation over time. Whether you use some fancy monitoring equipment, track your performance through some third-party software or just simply walk outside and check the diagnostics screen on your inverter, it’s getting progressively easier to pay attention to the energy you’re capturing from the sun.

As a solar data nerd (unite and be proud!), you’ll inevitably notice how much the energy generation from your PV systems varies from day to day, or even across the year. In the solar research community we refer to this as “seasonality” and “intermittency”.

Seasonality refers to the change in the sun’s position in the sky across the year, lower on the horizon in the winter, higher in the summer, while intermittency is a bit more complicated. The latter term refers to two things: 1) the changing intensity of the sun throughout the day as well as its unavailability at night, and; 2) the interruption of sunshine by clouds.

While seasonality and intermittency do not impact the solar owner directly, the fluctuations in power output from cloud cover do have an effect on local electricity networks. Clearing conditions tend to push local voltages upward; cloudy conditions push local network voltages lower.

Most of the time these effects are not large or noticeable because they are limited by a nifty science-y thing called “geographic smoothing”. This refers to the averaging effect of distributed solar power generation. A cloud which impacts your PV power output will not necessarily do the same for your neighbour’s PV system. This means that overall, as one big solar team, many rooftop solar PV systems working together can actually provide reliable energy generation. Well, most of the time anyway…

Collective ramp events

Despite the benefits of placing rooftop solar across a regional area and smoothing out the impacts of individual clouds, there are still times where fast-moving, fast-changing cloud cover conditions mean that geographic smoothing no longer provides a solution. In my academic role at the Australian National University, I’ve published some research work on this where I’ve titled such occurrences “collective ramp events” – periods of time where the power output of all of the solar PV systems in a given region changes from low to high (positive ramp) or high to low (negative ramp) due to changes in cloud cover.

When you consider the hundreds of megawatts of solar rooftop PV installed in the urban areas of Australia, you can begin to understand why these types of events are important to understand (there is nearly 1,000MW of solar installed in rooftops in southeast Queensland alone).

When a few hundreds of megawatts of solar generation undergoes a negative ramp event, the energy markets see a sudden jump in the regional demand and local distribution networks will see changes in network loads/voltages in their control rooms. This leads to challenges managing our electricity markets and networks in areas where large amounts of solar are installed.

Without managing these challenges appropriately, limitations will inevitably be imposed on the allowable amount of solar capacity that can be installed. This is referred to as the “maximum penetration level” of solar PV. When the maximum penetration level is reached, it means that an electrical distributor may cap the total size of new solar systems, or even deny new ones altogether.

Looking ahead

Thankfully, I have been working hard for the past several years on technologies which can solve these problems, and am dedicated to making sure the limiting of solar installations doesn’t happen. That’s why I’ve co-founded solar modelling and forecasting company Solcast, which is dedicated to smoothing the transition of the energy sector to high penetrations of solar energy generation and raising the total allowable levels of solar installed in our electricity networks. We’re big fans of solar power and we want to see it continue to be installed without restrictions.

Accomplishing these goals requires directly confronting the challenges of solar intermittency with new forecasting technologies. Solcast predicts changes in cloud cover across all of Australia (and also North America and Central America) through the use of satellite and weather model technologies. Over the near-term horizon (out to four hours ahead) we use the new, high-resolution Himawari 8 satellite to track and forecast the thickness and motion of cloud cover. The one-square-kilometre resolution of this satellite and its 10-minute update cycles enable Solcast to forecast the power output of individual PV systems and then aggregate those forecasts across the regions where large amounts of solar is installed.

We’re accomplishing this through work with local utilities to map our solar forecasting technologies back to the electrical grid, so that the future generation of solar in each neighbourhood and community is precisely known.

Keep up with the clouds

Our company strategy is to pair advanced knowledge of those fast-moving, fast-changing cloud cover conditions to offset the big changes in rooftop solar power generation by partnering with other companies who supply energy storage, smart inverters and demand management solutions.

It’s all part of a big plan in teaming up for the solar powered future, where we use foreknowledge of solar intermittency to coordinate the behaviour of the future grid. And as a solar (and/or storage) owner, householders get to be a part of that effort by simply collecting the glorious sunshine on their rooftops.

You can sign up for our solar forecasting API service for free at solcast.com.au. See you there!


Dr Nick Engerer is the co-founder of Solcast, a solar forecasting and modelling company based in Canberra, Australia, and an academic at Australian National University, where directs university research toward industry-relevant outcomes.