The solution is designed to eliminate “nuisance trips,” where harmless electrical spikes such as those caused by common household appliances cause an arc fault circuit interrupter (AFCI) to interrupt – or “trip” – current flow and thus power to the outlet. While designed to prevent the risk of fire by detecting an arc fault in a circuit and isolating it, such devices, say the researchers, often err on the side of being overly sensitive and shut off an outlet’s power in response to electrical signals that are actually harmless.

Their smart power outlet, they say, is able to distinguish such benign electrical spikes from dangerous arcs, such as sparking from faulty wiring. 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 design’s hardware uses a Raspberry Pi Model 3 single-board computer, an inductive current clamp, and a USB sound card to process electrical current data in real-time, while its software analyzes the data via a machine learning algorithm programmed to determine whether a signal is harmful or not by comparing a captured signal to others that the researchers previously used to train the system. The more data the network is exposed to, say the researchers, the more accurately it can learn characteristic “fingerprints” used to differentiate good from bad, or even one appliance from another.

After training the network on four devices – a fan, an iMac computer, a stovetop burner, and an ozone generator – the researchers tested their design on new data from the same four devices, and found it was able to discern between the four types of devices with 95.61% accuracy. In identifying “good from bad” signals, the system achieved 99.95% accuracy — slightly higher than existing AFCIs. The system was also able to react quickly and trip a circuit in under 250 milliseconds, matching the performance of contemporary, certified arc detectors.

The smart power outlet is an IoT device able to connect to other devices wirelessly, and is ultimately envisioned being used with a smartphone app through which users will be able to analyze and share data on their electrical usage. These data – such as what appliances are plugged in and where, and when an outlet has actually tripped and why – would be securely shared with the researchers to further refine their machine learning algorithm, making it easier to identify a machine and to distinguish a dangerous event from a benign one.

“By making IoT capable of learning, you’re able to constantly update the system, so that your vacuum cleaner may trigger the circuit breaker once or twice the first week, but it’ll get smarter over time,”says Joshua Siegel, a research scientist in MIT’s Department of Mechanical Engineering. “By the time that you have 1,000 or 10,000 users contributing to the model, very few people will experience these nuisance trips because there’s so much data aggregated from so many different houses.”

For more, see “Real-time Deep Neural Networks for internet-enabled arc-fault detection.”

Related articles:
Fraunhofer Institute develops smart wireless power outlets
Smart wallpaper acts as fire alarm
Sensor detects cable fire before it occurs
Smart home automation market to grow to $130B By 2025
Electronic DC circuit breakers for the smart grid developed


Linked Articles
eeNews Europe