CorroSense: Self-powered corrosion monitoring

The main goal of this project is to research and develop a low-power reliable wireless (corrosion) sensor network (WSN) enabling continuous health monitoring of reinforced concrete structures, allowing for early detection and prediction of corrosion through sensor fusion of the collected data from corrosion, temperature, humidity, etc. A low cost WSN will eliminate the need for intensive manual labor-work and will allow for continuous measurements if needed. For distributed WSNs over a large area, a concerning issue is how to power such devices. This project, CorroSense, therefore  aims to bring smartness to reinforced concrete as well as to eliminate batteries, an environmentally friendly and green goal as it would eliminate millions of electrochemical batteries. Furthermore by miniaturization, it will be possible to retrofit the sensors into existing structures. To achieve these goals the following scientific and technical objectives are sketched as below:
1- Corrosion sensing through a piezoelectric impedance sensing technique
2- Harvesting energy from the corrosion process for sensing, simple edge-processing, and communication.
3- Interfacing the corrosion sensors using a low-power (<10 μW) integrated chip
4- Data processing on edge using artificial neural networks (ANNs), and Spiking Neural Networks (SNNs) not only to reduce the data volume for communication but also brining temporal information into event-based processing that guides better decisions even in realtime without human intervention
5- Spike-mode compressed data communication using low-power communication schemes for a reliable data transmission to the cloud server
6- The delivery method/platform for managing the sensors and accessing the data and Application Programming Interface (API) for third party solutions.