The year 2022 was very challenging for the global energy market. Russia’s invasion of Ukraine and its long-term consequences affected households, businesses and entire economies. Everyone can see and feel the rising costs of electricity generation and fuel.
The German electricity grid, where Russian gas flow has stopped, is a great example of a strongly affected area. New, more eco-friendly ways of producing energy and heat, along with the faster adoption of electric vehicles, caused the risk of power grid overload and local power cuts. The mentioned situation happened in southern Germany this month. Similar problems might soon occur in the entirety of Europe. In 2022, the south of the Netherlands experienced issues with balancing the demand in their network. In Finland, drivers have been asked to avoid heating their cars in the morning, and in the UK people have been asked to reduce their electricity consumption between 4 and 7 PM.
Can quantum technology relieve the energy issues?
Let’s start with raw power consumption. Expert predicts that high-performance computing will consume between 3% to 7% of the world’s energy by 2030. AI also is a massive contributor to the global energy consumption. Estimations show that the calculation of GPT-3 consumed 936 MWh. For the new model’s calculation, the power consumption will definitely be higher — it doubles approximately every 3.4 months.
We believe that, when it reaches the required maturity level, quantum technology will support calculations in many different areas. It should reduce the calculation time, improve the results’ quality, and allow us to work with more significant problems. I saw a lot of positive developments and significant growth in quantum machine learning last year. Still, there is a lot to do when it comes to practical cases. If this technology is successful, it will positively impact the entire energy sector and our planet as a whole, by reducing energy demand.
Are we able to use this technology now?
E-ON has done exciting work. They used quantum annealing to optimise the power grid. Taking the three standardised configurations of power grids: IEEE-14, IEEE-33 and IEEE-118, allowed them to test their approach. The experiment produced an increase in performance. In smaller grids, quantum devices managed to provide better quality results. For the biggest one, the result was comparable to the classical benchmark, with much less time needed for the calculation. This approach can open the path to take real-time decisions and planning the operations that will maximise the network performance. It is essential, especially for the decentralised networks, which combine a lot of small energy manufacturers like households with solar panels.
Quantum technology can support us in optimisation topics. Imagine that you have a wind farm with hundreds of turbines. It would be best if you created a farm layout that will optimise the equipment and setup to maximise energy production. Due to the size of the farms and the correlation between turbine locations and lifetime, even small optimisation can hugely impact the amount of produced energy.
The usage of the Azure Quantum QIO demonstrated better (between 1% and 3%) performance than the best traditional approach. With these insights, the wind farm would be able to manufacture more megawatts of energy without additional investment.
It’s crucial to broaden your perspective and remember about energy consumers. One of the first examples I spotted when I started my journey with quantum computing was the optimisation of waste collection by Groovenauts and Mitsubishi Estate. This was almost three years ago. At that time, they were using D-Wave 2000Q quantum computer to optimise the route for collecting waste. The results were impressive — the covered distance was reduced from 2300 to 1000 km, CO2 by about 57% and the number of vehicles by 59%.
The global situation is currently forcing us to find new ways of energy manufacturing, usage patterns and distribution optimisation. The price increases and unbalanced power grids likely are unsustainable and will lead to optimisation of the power consumption process and its costs. We’ll need to find better ways to optimise these processes. Quantum computing is one of the potential answers, and it’s already starting to prove it’s a viable one. The technology, however, still needs to be improved; I believe it will happen shortly. Imagine that we can increase energy production by 5%, reduce consumption by another 5%, and get the ability to make real-time decisions. It would be a massive game changer for the whole sector.
Additionally, these changes will also impact our planet. New ways of power generation and heating will decrease carbon emissions. The same can be said about reduced energy consumption that can be achieved through optimisation.