Powering a Wireless Sensor Network for Machine Condition Monitoring

Abstract

Failure of a machine can lead to production downtime and significant financial losses. Condition monitoring is implemented to avoid such downtime and devices can be used to collect data used for monitoring machine health. Vibration data is the most common type of data used for predicting machine failure. To reduce the need for hazardous cables, such devices are often battery-operated, but this can decrease monitoring device lifespans to less than 3 years, if non-rechargeable batteries are used. This thesis first proposes a design framework for implementing radio frequency energy harvesting (RFEH) at a network level. All of the necessary inputs and parameters to ensure the successful implementation of RFEH for a wireless sensor network are explored. A second design framework is then proposed for using RFEH as a source of energy to power devices for condition monitoring. This includes a power analysis of all device components, as well as the design details for an implementation of wireless power transfer using a wireless transmitter and receiver. A comparison of different types of energy sources for the device is given, followed by a case study, using commercially-available components. A simulation is used to analyze the trade-offs for different values of RFEH parameters, trading off the total cost of implementation with the system's lifetime, based on total energy consumed.