Condition monitoring architecture for maintenance of dynamical systems with unknown failure modes

In this paper, a condition monitoring architecture for maintenance of dynamical systems with unknown failure modes is proposed. In many real applications, dysfunctional analysis techniques do not allow the determination of the complete list of failures that may impact a system. Our proposed architecture allows us to update this a priori analysis. The considered faults are slowly evolving gradual faults known also as drift. The architecture is based on dynamical clustering algorithm which leads to the detection and characterization of drifts amongst the normal operating mode and failure operating modes. The method is highlighted on a case study of a tank system.