[AI for the Sustainable Future] Energy Optimization
Optimizing energy is not only about efficiency anymore, also about minimizing environmental impacts. And AI can help.
Energy optimization — the process of maximizing the effectiveness of energy resources — has long been a vital concern at various scales, from businesses to national and now global levels. As energy usage is one of the critical human factors contributing to climate change, minimizing environmental impacts has become just as crucial as achieving efficiency. In the sustainability arena, applying AI to optimize energy emerges as one of its most practical uses.
While a wide array of energy optimization practices can benefit from AI, integrating and managing renewable energy into the power grid is one of the most crucial tasks.
The smart grid enhanced by AI
What is the smart grid? Wait, first of all, what is ‘the grid’? The grid, or the electric grid, is ‘a network of transmission lines, substations, transformers and more that deliver electricity from the power plant to your home or business’ by the definition from SmartGrid.gov.
And what makes a grid smart? When this electrical grid system is enhanced with digital technology for two-way communication between the utility and its end users, it becomes, so to speak, the smart grid. This modernized infrastructure aims to improve the efficiency and sustainability of electricity distribution. The smart grid plays a key role in enhancing national security and environmental health by optimizing the use of electricity. The challenges the smart grid is now facing include integrating renewable energy, managing peak demand, and modernizing aging infrastructure.
The Smart Grid | US Department of Energy's Advanced Office of Electricity
That’s why we need to make the smart grid even smarter. Integrating renewable energy can be particularly challenging as a huge number of solar panels and wind turbines are now generating electricity. This is a completely different game compared to the past when electricity was primarily supplied by a relatively small number of power plants. And this is where AI can help.
Four ways AI is making the power grid faster and more resilient | MIT Technology Review
The above article from MIT Technology Review discusses how AI is enhancing the smart grid in four aspects: 1) faster and better decision-making 2) tailored approach for every home 3) making EVs work with the grid 4) spotting disasters before they hit. The following is a summary of the article, foucusing on the four areas mentioned above:
Similar to the cases of climate modeling, covered in my previous post, a machine-learning model can help a grid operator dramatically improve data processing capabilities and complex mathematical calculations, leading to faster and more precise predictions of electricity supply (also, outage forecasting) and expedite disaster response in a cost-effective way.
At the consumer level, a noteworthy case involves Lunar Energy, a startup that has developed AI software to help optimize energy efficiency and reduce costs for its customers. The company’s Gridshare software collects energy usage data from numerous homes and combines it with weather data to feed a model that generates personalized predictions for individual households' energy needs.
More interestingly, a San Francisco-based company named WeaveGrid uses AI to relieve the overwhelming local grid burden caused by EV concentration around certain cities. By studying charging patterns, the company identifies optimal charging times and makes recommendations to customers via text message or app about when to charge their vehicles.
Meanwhile, PG&E has machine-learning models analyze photographs captured by drones and helicopters and identify areas requiring tree trimming (as overgrown trees are a leading cause of blackouts.) or pinpoint defective equipment that needs repairs. Also, climate models are used to predict the probability of grid failures resulting from extreme weather events, such as snowstorms or wildfires.
Let me also share another article below, which echoes similar points, to diversify the source.
Why AI and energy are the new power couple | International Energy Agency (IEA)
Simply put, the Internet of Things (IoT) technologies have been the backbone of the smart grid for over a decade, and now AI is taking it to the next level. (It’s interesting to note that IoT isn’t a buzzword anymore, not because the technology has disappeared but because it’s become ubiquitous, almost like the air we breathe! - Speaking of ‘ubiquitous’, that was another buzzword quite some time ago.- Will AI become just as seamlessly integrated into our lives? Remains to be seen … )
The supergrid: a new global challenge
Renewable energy is a vital component of sustainable energy optimization, but it presents unique challenges. Renewable sources are significantly impacted by climate change, (What an irony!) making power generation from solar panels and wind turbines less stable compared to traditional power plants. AI becomes instrumental in assessing and managing the variables to maintain stability.
And an intriguing Bloomberg video discusses a different type of challenges, arguing, “If a green pivot is to happen, power grids must become ‘supergrids’, continent-spanning networks that can move green energy thousands of miles. The technology is here, but politics may stand in the way.”
This mini documentary features Mongolia as a case study — a country with enormous renewable energy potential that could benefit the entire Northeast Asia region. However, geopolitical complexities (Mongolia being situated between China and Russia) add significant challenges. This situation raises questions beyond the realm of technology: Europe's energy crisis after Ukraine war broke out, driven by dependence on Russian energy, highlights the influence of political power on energy supply.
The power grid is hugely impacted by another type of power, particularly with the rise of nationalist, isolationist trends. So, how to foster a global endeavor towards the sustainable, clean energy future in this increasingly anti-globalist sentiment becomes a pressing issue.
Remember. AI technology is prepared to support this transition. Those not ready are we humans.