Low-Cost Solutions for Maintenance with a Raspberry Pi

Abstract

The paper describes how to develop an inexpensive automatic condition monitoring system that can reliably monitor the conveyor bearings. For this purpose, a Raspberry Pi 3 and a low-cost MEMS accelerometer have been used comparing the results to a more expensive data acquisition system. The project utilizes the open-source Mimosa data model that is installed to the Raspberry Pi to store and transmit data for analytics to diagnose the fault and determine the Remaining Useful Life (RUL). The required signal analysis is programmed with VTT Python O&M Analytics, which provides the ability to conveniently perform signal analysis, offering a comprehensive set of algorithms that can detect a bearing failure and calculate the RUL. The amplitudes of the bearing fault frequencies can be reliably seen using envelope analysis, and the magnitude of the amplitudes can be used to determine whether the bearing is defective. Furthermore, this article presents some possible low-cost data acquisition systems to monitor components in industrial use cases reliably. In conclusion, Raspberry Pi 3 is suitable for use in some industrial systems, for example, as a low-cost single-board computer for bearing maintenance. The goal of the project is to get a reliable and inexpensive data acquisition system for bearing maintenance and replace the more expensive one.