MALSTM-MCN Ensemble Learning-based Planetary Gearbox Fault Diagnosis method

Gearbox health monitoring is critical for minimizing downtime in industrial systems. Vibration-based methods are commonly used for gearbox diagnosis, but they face challenges in noisy environments and various operating conditions. This paper proposes a method for fault diagnosis of planetary gearboxes using multivariate attention LSTM multi-scale convolutional network ensemble learning. The proposed method uses a multi-scale convolutional network to capture the periodic characteristics of gearbox vibration signals and employs both LSTM and attention mechanism to fully exploit the small size of input data. The experimental results show that the proposed method achieves high accuracy and outperforms state-of-the-art methods, demonstrating its effectiveness in real-world applications.