Residual Generation Using Physically-Based Grey-Box Recurrent Neural Networks For Engine Fault Diagnosis
This work addresses fault diagnosis for industrial systems like engines, offering a hybrid approach that improves classification and identifies unknown faults, though it appears incremental as it builds on existing grey-box methods.
The paper tackled the problem of fault diagnosis in engines when fault classes are unknown and training data is limited, by developing an automated residual design using grey-box recurrent neural networks that incorporate physical insights, and demonstrated its potential in a real industrial case study with an internal combustion engine.
Data-driven fault diagnosis is complicated by unknown fault classes and limited training data from different fault realizations. In these situations, conventional multi-class classification approaches are not suitable for fault diagnosis. One solution is the use of anomaly classifiers that are trained using only nominal data. Anomaly classifiers can be used to detect when a fault occurs but give little information about its root cause. Hybrid fault diagnosis methods combining physically-based models and available training data have shown promising results to improve fault classification performance and identify unknown fault classes. Residual generation using grey-box recurrent neural networks can be used for anomaly classification where physical insights about the monitored system are incorporated into the design of the machine learning algorithm. In this work, an automated residual design is developed using a bipartite graph representation of the system model to design grey-box recurrent neural networks and evaluated using a real industrial case study. Data from an internal combustion engine test bench is used to illustrate the potentials of combining machine learning and model-based fault diagnosis techniques.