Johannesburg has a very characteristic skyline. With the tallest buildings in the continent, the horizon of the South African city is dominated by the Carlton Center tower, the Marble Towers, Trust Bank Building and the Ponte City Apartments, among others.
Of course, its sunsets and sunrises are famous and very acclaimed. Therefore, when they became rarer to see, Tapiwa Chiwewe, an Artificial Intelligence researcher, started to concern about this.
Chiwewe was concerned about the problems in gazing at the city’s skyline, not because it was rainy and cloudy… but because it was polluted.
The researcher created a solution based on Machine Learning, which could predict the levels of polluting substances in the atmosphere, helping administrations and citizens handle their movements in a more sustainable way.
In fact, consumption habits (the one that Chiwewe handles in his solution) is just one of the many ways in which Artificial Intelligence (AI) can help us protect the environment, and to become more energy efficient.
“There are only three limitations to building successful AI systems — computing power, availability of data and imagination.” Tadas Jucikas, PhD at the University of Cambridge
As long as we can count on data, imagination and computing power, Artificial Intelligence can have impacts at many different levels in the energy efficiency sector, but we can determine 3 main levels where its impact is noteworthy: production of energy, energy grid balancing and consumption habits.
As stated in the PwC report Fourth Industrial Revolution for the Earth “Machine Learning is being used to match energy generation and demand in real-time, releasing more fully the potential of “smart grids”, decreasing unpredictability and increasing efficiency, power balancing, use and storage of renewable energy.”
Renewable energy forecasting is one of the areas where Artificial Intelligence has the biggest impact. Its applications in energy production are so broad and notorious that it has already helped some countries change their energy production from fossil to renewable.
China is an example. In spite of its high pollution levels (or maybe because of them), China is leading the change in renewable energy production methods thanks to Artificial Intelligence. Germany is another country making the most out of Artificial Intelligence to predict the energy production and consumption. Using a Machine Learning-based early warning system, they obtain “real-time data from wind turbines and solar panels around the country to predict the energy that will be generated over the next two days”, as explained by digital specialist Ashiss K. Dash.
Regarding energy production forecasting, Terminus 7 has shown impressive figures in the wind sector: 95% accuracy in 1-hour predictions, 89% in 2-hour predictions, 78% in 4-hour predictions. Particularly, the 2-hour predictions have the biggest impact on business profitability.
Also at the production level, the impact of AI on the maintenance of machinery and equipment is also remarkable, making them much cheaper and affordable, what results in a positive impact on the energy price. For instance, Terminus7 has demonstrated that it can predict single component failure, helping extend component life beyond its service life or anticipating breakdowns before they happen.
Given the intermittent nature of renewable energy, and the fact that this is the main barrier for finally moving to clean energies, the expectations on Artificial Intelligence are big and the results have already been proven.
Energy grid balancing
In the energy grid, Artificial Intelligence can enhance the predictability of demand and supply for renewables, improve energy storage and management, facilitate the integration and consistency of renewables and provide a very competitive pricing.
For instance, “Stanford University is using Artificial Intelligence applications to help utilities better integrate their solar resources and make more informed planning decisions for enhancing grid reliability, resiliency and security”, according to the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy.
But this strategy of implementing AI in the energy grid is not exclusive of the United States, and in Europe it is also having a great success, as it has been already stated by the consultancy firm PwC, among others. Norway is one of the European countries leading the move to artificially intelligence and renewable energy.
Smart grids also provide a constant stream of data associated with how customers are consuming energy and at what time. For example, companies can set up a specialized plan for every customer. But the information related to how data is used also goes in the other direction, and customers now can make better decisions. In a broader context, knowing our usage patterns better (geographic and demographic characteristics) will help us develop more effective policies and laws (this is the case of the solution that Chiwawa, our engineer from South Africa, developed. Thanks to it, public administrations and citizens can now handle shifting consumption in a less pollutant way by themselves).
Real-time AI-optimized energy efficiency can have an immediate and substantial impact on energy consumption. Google, for example, cut power use in its data centers by 40 per cent by using DeepMind’s reinforcement learning algorithms to optimize cooling.
Therefore, we will be able to know how much energy we will consume, and how much renewable energy we will produce. With this estimation, we can decide if we need additional energy to be produced (with fossil fuels, if needed) or not, saving a lot of time, money… and polluting much less.
This is just the beginning, but Artificial Intelligence is already contributing to making skylines all over the world cleaner and Tapiwa Chiwewe will be able to enjoy the Johannesburg landscape again.
If you want to apply Artificial Intelligence to your business and check the amazing results (as in the renewable energy sector at every level), simply let us know. We will be delighted to guide you.
Would you like to know more?
This is just an example of what Artificial Intelligence can do for you. If you want to know more, you can download our free guide of AI basic concepts.
Do you want integrate AI to your business?
You are using Artificial Intelligence (AI) solutions right now, even if you don’t know it. Image recognition is one of its main uses. Science, business or education are fields benefiting from this application, but medicine is probably one of the most interesting.read more
How can your coding skills help others? This is your chance to play your part building predictive Machine Learning & Deep Learning models to help detect the risk of diabetes and breast tumor!read more
Artificial Intelligence allows businesses to learn more about consumer preferences and help them deal with these huge audiences, predicting and identifying their preferences and needs.read more
The organizations capable of analyzing and exploiting information with a business vision will stand out from the rest, saving time and money while delivering the right products to their customers.read more
“Big Data is not about the data”. The crucial issue when talking about this is, how do we unravel the strands of big data and pick out the relevant parts?read more