2026
Neural network-supported vibration analysis of turbomachinery for early fault prediction
A new hybrid deep learning architecture, combining one-dimensional convolutional neural networks (1D-CNN) with bidirectional long short-term memory (BiLSTM) networks to the problem of automatic detection and early forecasting of mechanical faults based on raw vibration signals is suggested.
N. Bharani
· Materials Research Proceedin... · 0 citations