Descripció del projecte
The thesis proposal is in the area of optimization of wireless sensor network nodes for monitoring infrastructure (Structural Health Monitoring, SHM) to extend the node lifetime as much as possible. In particular, the target application is the SHM to predict failure, which is an unsolved problem under research.
The thesis proposal aims to provide hardware and software co-design guidelines for a node capable to adapt to: a) Operating and environmental condition; b) Available energy; c) State of health of the monitored structure based on a high-level knowledge model of the failure mechanisms.
Open lines of research:
· Adaptive hardware using COTS components. Published example: [1]
· Deep learning for SHM. Some references can be found in [2].
· Application of existing failure prediction models. Some examples: [4], [5], [6].
[2] R. G. Mishalani and S. M. Madanat, “Computation of Infrastructure Transition Probabilities Using Stochastic Duration Models,” J. Infrastruct. Syst., vol. 8, no. 4, pp. 139-148, Dec. 2002.
[3] R. Zhao, R. Yan, Z. Chen, K. Mao, P. Wang, and R. X. Gao, “Deep Learning and Its Applications to Machine Health Monitoring: A Survey,” arXiv:1612.07640, Dec. 2016.
[4] J. Guo, X. Xie, R. Bie, and L. Sun, “Structural health monitoring by using a sparse coding-based deep learning algorithm with wireless sensor networks,” Pers. Ubiquitous Comput., vol. 18, no. 8, pp. 1977-1987, Dec. 2014.
[5] R. G. Mishalani and S. M. Madanat, “Computation of Infrastructure Transition Probabilities Using Stochastic Duration Models,” J. Infrastruct. Syst., vol. 8, no. 4, pp. 139-148, Dec. 2002.
[6] J. Sadeghi and H. Askarinejad, “Development of improved railway track degradation models,” Struct. Infrastruct. Eng., vol. 6, no. 6, pp. 675-688, Dec. 2010.
