Structural monitoring of critical bridge structures can greatly benefit from the use of wireless sensor networks (WSNs), however energy harvesting for the operation of the network remains a challenge in this setting. While solar and wind power are possible and credible solutions to energy generation, the need for positioning sensor nodes in shaded and sheltered locations, e.g., under a bridge deck, is also often precluding their adoption in real-world deployments. In some scenarios vibration energy harvesting has been shown as an effective solution, instead.This paper presents a multihop vibration energy-harvesting WSN system for bridge applications. The system relies on an ultra-low power wireless sensor node, driven by a novel vibration based energy-harvesting technology. We use a receiver-initiated routing protocol to enable energy-efficient and reliable connectivity between nodes with different energy charging capabilities. By combining real vibration data with an experimentally validated model of the vibration energy harvester, a hardware model, and the COOJA simulator, we develop a framework to conduct realistic and repeatable experiments to evaluate the system before on-site deployment.Simulation results show that the system is able to maintain energy neutral operation, preserving energy with careful management of sleep and communication times. We also validate the system through a laboratory experiment on real hardware against real vibration data collected from a bridge. Besides providing general guidelines and considerations for the development of vibration energy-harvesting systems for bridge applications, this work highlights the limitations of the energy budget made available by traffic-induced vibrations, which clearly shrink the applicability of vibration energy-harvesting technology for WSNs to low traffic applications.
|Title of host publication||Proceedings of the 2018 International Conference on Embedded Wireless Systems and Networks|
|Subtitle of host publication||EWSN 2018, Madrid, Spain, February 14-16, 2018|
|Number of pages||12|
|Publication status||Published - 16 Feb 2018|