Non-linear minimum variance estimation for fault detection systems

Abstract

A novel model-based algorithm for fault detection in stochastic linear and non-linear systems is proposed. The non-linear minimum variance estimation technique is used to generate a residual signal, which is then used to detect actuator and sensor faults in the system. The main advantage of the approach is the simplicity of the non-linear estimator theory and the straightforward structure of the resulting solution. Simulation examples are presented to illustrate the design procedure and the type of results obtained. The results demonstrate that both actuator and sensor faults can be detected successfully

Performance evaluation of redundant OPC UA architecture for process control

“…Ensuring the communication reliability remains out of the scope of the OPC UA interface standard. There are many methodologies for fault detection of the control loop (Alkaya and Grimble, 2014; Frank et al, 2000; Patton et al, 1995) or fault tolerance for communication systems (Du et al, 2013; Yu et al, 2013). One of the strategies for providing the fault recovery capability is redundancy, i.e.…”

Machinery fault diagnosis via an improved multi-linear subspace and locally linear embedding

“…When the labels of the data set are unknown, LSBFE will fail. Usually, a complex data set can be better treated by nonlinear manifold learning algorithms (Alkaya and Grimble, 2015; Ding et al, 2015; He, 2013; Wang and Yin, 2014; Yin et al, 2015b). Nonlinear methods explore local structures in the original space and compute the embedding result by these local structures.…”