Date of Original Version
Abstract or Description
Bayesian networks, which may be compiled to arithmetic circuits in the interest of speed and predictability, provide a probabilistic method for system fault diagnosis. Currently, there is a limitation in arithmetic circuits in that they can only represent discrete random variables, while important fault types such as drift and offset faults are continuous and induce continuous sensor data. In this paper, we investigate how to handle continuous behavior by using discrete random variables with a small number of states, without using soft evidence, which is a traditional technique for handling continuous sensor data. We do so by integrating a method from statistical quality control, known as cumulative sum (CUSUM), with probabilistic reasoning using static arithmetic circuits compiled from static Bayesian networks. We demonstrate that our ProDiagnose approach can diagnose faults that are small in magnitude (offset faults) or which drift linearly from a nominal state (drift faults). Experimentally, an arithmetic circuit model of the ADAPT Electrical Power System (EPS), a real-world EPS located at the NASA Ames Research Center, is considered. We report on the validation of this approach using ProDiagnose, which had the best performance in three of four of the industrial track competitions in 2009 and 2010 (DXC-09 and DXC-10).
Annual Conference of the Prognostics and Health Management Society 2011 (PHM-11), .