Engineers from Purdue University have developed a new sensing module that takes advantage of machine learning to monitor electric currents and garner information, including energy usage, system problems, and the best approaches to manufacturing. The team states that the sensor module is the first of its kind that is noninvasive, safe, and offers a level of precision than others currently on the market.
Most platforms that monitor and diagnose system issues via an electric current rely on resistors or Hall effect sensors, which are not capable of measuring small currents, thus limiting their applications. What makes Purdue’s sensor different from the others is that they paired their sensor with a machine learning algorithm to boost its monitoring and diagnostic capabilities.
According to professor of electrical and computer engineering Byunghoo Jung, “Our technology enables someone to discover through current. This sensor could be used with machine learning to train manufacturing robots, provide precise tips for homeowners on cutting down their energy usage or help diagnose issues with electric vehicles and scooters.”
The small sensor is wrapped around a central wire to monitor currents, which makes for easy installation and maintenance in tight spaces, like those found in electric vehicles. It’s also able to transmit current data to computer systems through Bluetooth, USB, or other methods. It can utilize machine learning to detect something such as a microwave oven being used at a certain time, and whether or not that time is optimal for energy consumption.
The engineers have already worked with the Purdue Research Foundation Office of Technology Commercialization to patent their design, and are currently searching for partners to license the technology.