A distributed diagnosis strategy using bayesian network for complex wiring networks

Overview & Synopsis

In this paper, we propose a distributed diagnosis strategy by using reectometry in highly complex wiring networks. Although the problem of sensors number optimization is greatly studied in the literature, it is not well investigated in complex wiring networks diagnosis. Our proposed approach is based on two principles which are diagnosis sensors number and location optimisation using Bayesian Networks and measure uncertainty estimation. It consists in four steps: (1) sensors implementation in a deterministic case, (2) in uential parameters on diagnosis measure identification, (3) diagnosis measure modelling using Bayesian Networks, (4) sensor number and location optimization. Here, our objective is to minimize both sensors number and a wire diagnosis measure uncertainty.