Fault detection and isolation of multi-variate time series data using spectral weighted graph auto-encoders.
Umang GoswamiJyoti RaniHariprasad KodamanaSandeep KumarPrakash Kumar TamboliPublished in: J. Frankl. Inst. (2023)
Keyphrases
- multi variate
- weighted graph
- fault detection and isolation
- industrial processes
- fault diagnosis
- normalized cut
- control system
- closed loop
- fault detection
- graph partitioning
- gaussian distribution
- fault detection and diagnosis
- shortest path
- minimum weight
- graph structure
- data objects
- high dimensional
- adjacency matrix
- graph connectivity
- real time
- spectral analysis
- high dimensional spaces
- edge weights
- spanning tree
- expert systems
- data analysis
- high dimensional datasets
- decision making
- neural network