All solutions come with a problem. Arc fault detectors are designed to prevent dangerous electrical fires, but sometimes they can be too sensitive and shut off power unnecessarily.
At least that's how a team of engineers at the Massachusetts Institute of Technology (MIT) sees it. To avoid this inconvenience, the team has developed a “smart power outlet."
As reported by MIT News, the device can analyze electrical current usage from a single outlet or multiple outlets, and it can distinguish between benign and dangerous arcs. Benign arcs are defined as "harmless electrical spikes" such as those caused by common household appliances. Dangerous arcs are those resulting from faulty wiring, which could lead to a fire. The device can also be trained to identify what might be plugged into a particular outlet, such as a fan versus a desktop computer.
The team’s design product includes custom hardware and software. The hardware processes electrical current data in real-time, while the software analyzes the data using a set of machine learning algorithms, referred to as a "neural network," that is inspired by the workings of the human brain. The more data the network is exposed to, the more accurately it can learn characteristics or "fingerprints" used to differentiate good from bad arcs.
Joshua Siegel is a research scientist in MIT’s Department of Mechanical Engineering and a co-author of the article, which describes the results of the project in the September 2018 issue of the journal "Engineering Applications of Artificial Intelligence."
Siegel said users will be able to update their system, “so that your vacuum cleaner may trigger the circuit breaker once or twice the first week, but it’ll get smarter over time.”
The smart power outlet is also able to connect to other devices wirelessly, which will enhance its learning capabilities. Siegel explains the larger goal is to "have everything connected, all of the time, for a smarter, more interconnected world."