Degradation processes modelled with Dynamic Bayesian Networks

In this paper a generic degradation model based on Dynamic Bayesian Networks (DBN) which predicts the condition of technical systems is presented. Besides handling bi-directional reasoning, a major benefit of using DBNs is its capability to adequately model stochastic processes. We assume that the behavior of the degradation can be represented as a P-F-curve (also called degradation or life curve). The model developed is able to combine information from condition monitoring systems, expert knowledge and any kind of observations like sensor data or notifications by the machine operator. Thus it is possible to even take the environment and stress into account under which the component or system is operating. Thus it is possible to detect potential failures at an early stage and initiate appropriate remedy and repair strategies.