Predicting Power Outages Through Big Data and Machine Learning

Thursday - March 7 2024, 19:39 UTC - 8 months ago

Dr. Mladen Kezunovic and his team at Texas A&M University are using big data and machine learning to predict power outages caused by environmental conditions. By combining historical outage data and weather-related data, the team creates models that can predict when and where outages are likely to occur. They are also working on educating consumers, particularly children, about power outages to increase awareness and reduce panic.

Unplanned power outages due to environmental conditions (wind, lightning, tree growth, etc.) leave those without power on their own, sometimes for long periods of time. Utility companies lack the ability to predict when forced outages will occur, so no mitigation measures targeting consumers are deployed ahead of time to reduce the impact of an outage.

Dr. Mladen Kezunovic, a professor in the Department of Electrical and Computer Engineering at Texas A&M University, and his team are combining historical outage data and weather-related data, often called big data, and machine learning to predict outages and change the outage mitigation paradigm from reactive to proactive. This will help consumers be prepared for potential outages and minimize their impact.

Using machine learning and a variety of data describing the causes of outages, the team can study data from the past to make predictions about the future. This includes historical outage data, weather data, and data on factors such as vegetation and animal intrusion. By analyzing this data, the team can create models that can predict when and where outages are likely to occur, helping utility companies and consumers be more prepared.

Kezunovic explains the process of testing their predictions: "If you're making predictions using data from the past, you are predicting what actually happened in the past, and then comparing it to what actually happened. If you were correct about the past, it should work in the future." By learning from the past, the team can make better predictions for the future.

One of the challenges of predicting outages is the multitude of factors that can contribute to an outage, such as wind, rain, and lightning. To address this, the team combines database models with physics-based models to create a more comprehensive understanding of the risk of an outage. They also use geographic information systems (GIS) to overlay predictions onto a map of the power grid, giving a visual representation of where outages are most likely to happen.

The team is also working on developing specific communication messages for different types of consumers. By tailoring the information to different groups, they hope to increase awareness and understanding of power outages and what to do when faced with one. They are also working with The DoSeum, a museum for kids in San Antonio, to educate young children about outages and how to respond to them.

In the future, the team hopes that their work will not only help utility companies better predict and prepare for outages, but also reduce the inconvenience and negative impacts on consumers that occur during prolonged outages caused by environmental factors.