[7A1] Distinguished Overview Lecture: Deep information fusion for mechanical fault diagnosis
H R Karimi
Politecnico di Milano, Italy
Deep learning methods for fault diagnosis play a critical role in the monitoring and detecting of operating conditions of mechanical equipment. However, the developed algorithms based on single-source sensor and data features exhibit some deficiencies for the complex and harsh real-world factory environments. Therefore, this talk addresses some high-accuracy and reliable deep learning-based algorithms to identify the fault state of the rotating machinery, considering some issues of the deep learning algorithms, such as the limited training samples, feature extraction capability and interpretability, as well as multi-source information fusion for mechanical fault diagnosis.
