Investigation of energy harvester powered wireless sensor node in structural health monitoring application

To ensure the integrity of important public structures such as bridges, structuralhealth monitoring (SHM) must be performed regularly. Wireless sensor nodes (WSN) aredeployed to constantly collect the structure’s health data. Due to the hard-to-reach deploymentlocations of these wireless sensor nodes, they must harvest energy from the surroundingenvironment. For bridges, the energy harvested can come from the oscillation of thebridge as vehicles travel over the bridge. As the harvested energy is randomly fluctuatingdepending on the environment while the wireless sensor nodes try to constantly monitor andwirelessly report the bridge’s health, there is a need to manage when the node should sleep orshut down to save energy consumption. If the upcoming harvested energy is high, sleepingis more energy efficient than shutting down. On the other hand, if the upcoming harvestedenergy is low, shutting down is better than sleeping.This thesis proposes a new algorithm, which is a hybrid of two existing algorithmsalready proposed in the literature. These algorithms are Turn-Off-Only and Sleep-Only algorithms.They are beneficial in different extreme scenarios. However, for more generalscenarios, a mix of the two algorithms is better. In this work, the energy consumption ofa real wireless sensor node for bridge health monitoring is measured. Then a simulation isperformed in order to compare the proposed method and the Turn-Off-Only and Sleep-Onlymethods. The results show that, when the energy storage element is sufficiently large but nottoo large, our proposed algorithm performs better than the Turn-Off-Only and Sleep-Onlymethods.