AI opens up a wide range of opportunities for the energy industry: It can help optimize grid management and, in the future, automate it by incorporating both forecast and status data from equipment into its decisions. At the equipment level, AI-supported systems can also independently analyze sensor data and automatically report that maintenance is necessary, for example, even before a fault occurs. In the future, AI could also plan service calls independently, thereby making better use of scarce human resources. These are just a few examples that show how important AI will be for a stable energy supply in the future. The potential for power grids is huge: The International Energy Agency (IEA) has calculated that intelligent sensors and AI-based grid management could free up 175 gigawatts (GW) of transmission capacity without the need to build new lines.

However, in order for AI to realize its potential, it first needs a large amount of data—without data, it is completely worthless. For energy suppliers to use AI effectively, comprehensive data collection from transformers, circuit breakers, and cables is therefore an important prerequisite. Roland Götz, Head of Innovation and Technology at MR, explains: "To make this possible, we at MR are working to reduce the costs of digitization." And AI is supposed to help with this: "We are currently testing algorithms that draw conclusions from the measurement data of a few sensors and calculate values for other fictitious or digital sensors." The effect: fewer sensors would be needed to digitize a transformer, allowing grid operators to save a lot of money when modernizing their fleets. 

Machine learning has played a central role at MR for around 20 years. It all started with dehydrating breather with self-learning control algorithms. Over the years, the sensor portfolio grew and other products were equipped with intelligent algorithms. These include MSENSE^®VAM, a vibroacoustic measurement method that uses self-learning algorithms to monitor the switching process in tap changers. AI enables increasingly precise analyses, particularly in the area of equipment assessment. The ETOS® automation solution, for example, is already established in the field as a system for monitoring and data acquisition on transformers. Functions such as asset intelligence and the analysis of dissolved gases in insulating oil (DGA interpretation) already provide data- and model-based insights for more efficient condition assessments—including specific recommendations for action. But once the equipment has been digitized across the board, much more will be possible in the future.