Using neural network based on the shaft orbit feature for online rotating machinery fault diagnosis

This study proposes a rotating machine fault diagnosis strategy based on shaft orbit technology and uses the fractal theory to extract features of the shaft orbit. The method not only effectively reduces the amount of data but also keeps a few important characteristic parameters, thus, becoming a fault diagnosis for back-propagation network algorithms of primary input sources. The experimental results confirm that the back-propagation network algorithm has excellent fault identification capabilities and computing performance and verify shaft orbit technology for mechanical fault identification. In the future one could adopt motor current signature analysis techniques with this algorithm, thereby reinforcing it due to abnormal electrical parameters caused by electrical fault identification capabilities.