Probabilistic reasoning in intelligent systems: networks of plausible inference
Probabilistic reasoning in intelligent systems: networks of plausible inference
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Fault diagnosis in power networks with hybrid Bayesian networks and wavelets
IEA/AIE'11 Proceedings of the 24th international conference on Industrial engineering and other applications of applied intelligent systems conference on Modern approaches in applied intelligence - Volume Part I
| Hi-index | 0.00 |
In this work we propose a two-phases fault diagnosis framework for industrial processes and systems, which combines Bayesian networks with neural networks. The first phase, based just on discrete observed symptoms, generates a set of suspicious faulty process components. The second phase analyzes continuous data coming from sensors attached to components of this set and identifies the fault mode of each one. In first phase we use a discrete Bayesian network model, where probabilistic relationships among system's components are stated. In second phase, we analyze sensor measurements of suspicious faulty components with a probabilistic neural network, previously trained with the eigenvalues of collected data. We show promising results from simulations performed with a 24 nodes power network.