Use + Remix

If the world doesn't meet its net zero targets by 2030, experts predict disaster. AI might be the key to help us succeed where currently we're failing.

AI could improve the way we harness solar energy. : Giorgio Trovato, Unsplash Unsplash licence AI could improve the way we harness solar energy. : Giorgio Trovato, Unsplash Unsplash licence

If the world doesn’t meet its net zero targets by 2030, experts predict disaster. AI might be the key to help us succeed where currently we’re failing.

The ambitious net zero targets set under the United Nations’ Paris Agreement are falling further behind schedule at an unprecedented rate.

In some parts of the world, sustainability has not been a top priority. Where it has, countries have faced the critical problem of trying to transform existing infrastructure to fit an innovative, futuristic grid.

So far, the world’s main focus in chasing those goals has been expanding the use of renewables such as solar and wind, as well as sustainable transport, such as electric vehicles.

New energy technologies allow us to imagine a clean picture of future cities, where smokeless electric vehicles are charged using renewables. But such a vision doesn’t match reality. To realise such an advanced future demands energy grids be completely revolutionised to be faster, more reliable and more intelligent.

Without the help of artificial intelligence, it seems almost impossible to build such a sophisticated society.

Australia has set an ambitious target to achieve 82 percent renewable electricity generation by 2030. As of June 2023, that figure sat around 30 percent. A huge part of closing the gap is dealing with the added demand resulting from the soaring number of electric vehicles.
Nearly 310,000 GWh of energy was consumed on Australian roads in 2021-2022, with only a negligible 140 GWh being driven by electricity.

Meeting that abundant extra energy demand through the electricity grid would require adding a massive generation capacity to the network.

The most profound application of AI in future energy grids emerges through the concept of ‘microgrids’.

Unlike conventional grids, where energy flows from fossil fuel-burning power plants to end-users in houses, microgrids operate on a smaller scale and can function independently or in conjunction with the main grid.

With distributed generators on every rooftop around the city, the centralised nature of energy and its energy dictatorship are long gone.

As a byproduct of the new decentralised grids, the concept of peer-to-peer energy sharing could finally be realised. Previously, the complexity of price bidding within conventional networks made it unfeasible.

But the use of AI can democratise the energy network, replacing the monopoly of centralised energy providers with a network of ‘prosumers’, people who consume and produce all at once. In this network, prosumers can trade their energy back and forth based on its profitability.

In the generation sector, solar energy has been the main focus for many years. However, excessive addition of solar capacity to the network leads to an overabundance of energy during daylight hours, a period with no significant demand.

However, on the demand side there is a massive consumption peak during the evening, just as solar generation plunges to zero. This eventually leads to an excess of energy during peak sunlight hours and a deficit in the evening.

Optimal energy management is necessary to match the generation and consumption patterns.

The brilliant power of AI in analysing and interpreting data could potentially be the key to overcoming the situation. AI can serve as a highly intelligent coordinator among multiple agents in the energy market, ensuring both profitable and stable operations.

Future cities are envisioned as green spaces powered by fully renewable resources.

Maintaining long-term sustainable operation remains a challenge. Apart from the poor efficiency of solar panels (less than 30 percent), the major bottleneck lies within its intermittent nature.

It would be unwise to rely on such an unpredictable source of energy to power our homes, drive our electric vehicles and sustain our industries. AI could add much more certainty, such as through recognising and predicting weather patterns and conditions, allowing upcoming solar generation to be forecast with more precision.

By integrating renewable generation, energy storage and AI, fossil fuel power plants can be phased out from the primary generation layer, serving instead as backup capacity. This strategic shift ensures both reliability and sustainability in energy provision.

With 2030 just around the corner, the transition to a whole new grid seems more crucial than ever before. Given the abundance of newly emerging energy prosumers in the grid, AI will play a crucial role in its control. However, human supervision remains a necessity, even with a fully AI-integrated grid, to ensure its smooth operation.

Professor Mehdi Seyedmahmoudian is Discipline Director of Electrical and Electronics at School of Science Computing and Engineering Technologies and Director of Siemens Swinburne Energy Transition Hub at Swinburne University of Technology, Australia. 

Saeid Ghazizadeh is a PhD candidate at Swinburne University of Technology. His research is on electric vehicles.

Professor Alex Stojcevski is the Dean of the School of Science, Computing, and Engineering Technologies, Swinburne University of Technology. He leads pioneering research projects on developing sustainable energy solutions.

Professor Saad Mekhilef is a Distinguished Professor at Swinburne University of Technology. He has an outstanding academic record in electrical engineering and renewable energies.

Originally published under Creative Commons by 360info™.

Are you a journalist? Sign up for our wire service