Making smart energy even smarter with data science

smart energy

The future of sustainable electricity usage is smart energy, and data scientists are matching supply and demand in smarter ways than ever.

You throw your clothes into the washing machine at 6 a.m. It beeps a warning: If you wash your clothes now, it will have to draw electricity from the grid, which will be expensive. It suggests you wait an hour until the sun comes up – using smart energy sourced from your solar panels.

While renewable energy is clean, often cheap and in infinite supply, a big drawback is that the flow can be intermittent – it changes depending on the amount of sunshine, wind or water. To make the most of it, we need to utilise smart energy – improve storage, enable the grid to both supply and absorb power, and equip houses and buildings to make the most of energy from renewable sources.

Throughout Victoria, smart meters are installed as standard. These digital devices measure in real time the amount of electricity used and send this data to energy distributors every 30 minutes. They are connected to the internet, making it possible to compile all the usage data and find patterns in it.

University of Melbourne electrical engineer and computer scientist Lachlan Andrew is an expert in electricity network data. Lachlan is using data from these smart energy meters to determine how much electricity households are using. He is looking at more than 17,000 meter readings per year for each household – hundreds of millions of electricity readings overall.

By analysing this vast dataset, he can predict what time of day households use the most energy. For example, on a 40°C afternoon, thousands of people may turn on air con at the same time. This would be a huge draw on electricity and could force rolling blackouts across suburbs because there isn’t enough power to go around.

“Even if there weren’t climate change issues, in a few hundred years we’ll run out of coal and, in just a few decades, oil will become scarce. So we need to move quickly towards renewable energy,” says Lachlan.

One smart energy solution could be improving storage at the source rather than connecting solar power to the grid. Another might be home automation, with appliances such as washing machines that self-monitor electricity prices based on popular usage times and turn on when electricity is cheapest.

Monash University PhD student Dora He is studying ways to process electricity usage data in real time. Her research relates to smart energy grids – computer technology added to traditional electricity networks to make production more efficient, reliable, economic and sustainable.

“One idea is to adjust electricity prices when the grid network is nearing capacity, to encourage people to reduce their power consumption,” says Dora.

Dora’s research is applicable to heaps of energy optimisation projects. She’s taken part in an internship project through APR.Intern which managed energy storage with battery systems connected to solar power systems.

Lachlan says it’s important to tease out problems with renewable energy now, so they can be resolved for a future where green power is a necessity.

“It’s urgent not because we need to have everything renewable tomorrow, but because the transition is going to take decades,” he says. “A change of this magnitude will only become harder if we put it off.”

Michelle Wheeler

“One idea is to adjust electricity prices when the grid network is nearing capacity.”

Monash PhD student Dora He

STEM Contributor

Author: STEM Contributor

This article was written by a STEM Contributor for Careers with STEM. To learn more, please visit our contact page.