iMachine wireless: online predictive maintenance system in electrical machines through the use of fuzzy logic with analysis of vibration and temperature parameters of an autonomous sensor network based on energy harvesting technology

Abstract

The iMachine Wireless project aims to develop a network of wireless sensors (WSN) based on autonomous energy harvesting technology. The Wireless sensor networks (WSNs) offers an attractive solution for many environmental, safety and process monitoring aplications. However, your life remains very limited by the capacity of the battery. Through the use of energy-harvesting techniques, ambient vibration can be captured and converted into usable electricity to create a self-powered WSN that is not limited by finite battery power. This research investigates in an analytical and experimental way the performance of an RSSF powered by an Energy Harvesting System, considering the energy consumption of a wireless sensor node in a star topology network. This is an incremental innovation of the iMachine system of predictive analysis developed by iTech Soluções, already validated and deployed in several clients in the Brazilian market. iMachine is a low cost system for monitoring and monitoring industrial equipment for predictive maintenance. The developed system monitors the vibration of the element being studied, and registers it against a vibration standard considered nominal, that is, a satisfactory operating condition of the machine. When a variation in the vibration of the monitored device is identified, its behavior must be observed, not only in amplitude but also in the frequency spectrum, since generally the incidence of faults or anomalies present vibrations at frequencies different from the working nominal of a device. The Fourier Transform of the signal and the frequent readings records allow continuous monitoring of the monitored equipment. In addition to the mechanical vibration monitoring, the electric current of the element drive motor is monitored to observe possible overload, phase unbalance and analysis of the electric current signal frequency spectrum, which allows us to evaluate the changes in the power supply of the same, indicating some anomaly of electrical nature to the engine under study. Finally, the temperature of the elements under study is monitored, since the useful life of these elements depends on the temperatures to which they are subjected in an operating regime that have a direct impact on insulation of coils and lubricants of the mechanical parts. This system allows an evaluation of a machine, without human intervention for vibration and temperature measurements, allied with the historical data acquired, it becomes a powerful tool for the implementation of a predictive maintenance program. The development of this technology significantly broadens the market for online predictive analytics solution as a wireless sensor network (WSN) offers an attractive monitoring solution but has the disadvantage of limited battery life. The development of a self-powered wireless sensor network is a high-impact solution, allowing wireless sensors to be used without the need for battery replacement. (AU)