@inproceedings{Zhou:2017:10.1109/ECC.2016.7810375,
author = {Zhou, Y and Boem, F and Fischione, C and Parisini, T},
doi = {10.1109/ECC.2016.7810375},
publisher = {IEEE},
title = {Distributed Fault Detection with Sensor Networks using Pareto-Optimal Dynamic Estimation Method},
url = {http://dx.doi.org/10.1109/ECC.2016.7810375},
year = {2017}
}

Download

TY  - CPAPER
AB  - In this paper, a distributed method for faultdetection using sensor networks is proposed. Each sensorcommunicates only with neighboring nodes to compute locallyan estimate of the state of the system to monitor. A residualis defined and suitable stochastic thresholds are designed,allowing to set the parameters so to guarantee a maximumfalse alarms probability. The main novelty and challenge ofthe proposed approach consists in addressing the individualcorrelations between the state, the measurements, and thenoise components, thus significantly generalising the estimationmethodology compared to previous results. No assumptions onthe probability distribution family are needed for the noisevariables. Simulation results show the effectiveness of theproposed method, including an extensive sensitivity analysiswith respect to fault magnitude and measurement noise.
AU  - Zhou,Y
AU  - Boem,F
AU  - Fischione,C
AU  - Parisini,T
DO  - 10.1109/ECC.2016.7810375
PB  - IEEE
PY  - 2017///
TI  - Distributed Fault Detection with Sensor Networks using Pareto-Optimal Dynamic Estimation Method
UR  - http://dx.doi.org/10.1109/ECC.2016.7810375
UR  - http://hdl.handle.net/10044/1/39085
ER  - 

Download