Enabling Predictive Maintenance on Electric Motors Through a Self-sustainable Wireless Sensor Node

Periodic maintenance and unpredictable equipment failure of industrial machinery are expensive elements in a company’s balance and potentially hazardous for human operators. Periodic inspections at predefined intervals are commonly applied to limit unplanned production downtime and safety concerns. The latest advancements in smart sensor technology enables online equipment monitoring that can directly anticipate the deterioration and incoming breakages on operating machines, reducing maintenance costs. This paper presents a deploy and forget sensor node for predictive maintenance on industrial electric motors, which targets three-phase asynchronous motors, supporting the data collection from multiple sensors, such as vibrations, environmental noise, temperature, and the external magnetic field. The sensor node features ultra-low-power design, achieving self-sustainability by exploiting a 4\(\,\times \,\)4 cm thermal electric generator with a \(\varDelta T\) of \(20\,^\circ \text {C}\) for at least 72 s. Moreover, it features short-long wireless data transfer over WiFi and the NB-IoT protocol. Results report the energy harvesting efficiency and the circuit power consumption from a real-world tests.