Study on HS-RNN in Vibration Prediction of Mechanical Spindle
Jinxiang Pian, MasumaRifah Tamanna, Alkasim Usman Abdulmajid · 2021
Mechanical spindle in the high-speed operation due to an uneven distribution of the rotor mass, causing the spindle to vibrate thus affecting its accuracy. Therefore, the model of mechanical spindle vibration prediction is of great significance. The mechanism of mechanical spindle vibration prediction is complex, and the amplitude of vibration varies nonlinearly with the speed. It is difficult to establish accurate model of mechanical spindle vibration prediction. In this paper, RNN (Recurrent Neural Network) recurrent neural network is used to establish the model of mechanical spindle vibration prediction. The vibration amplitude of the built-in balance block at different positions and spindle speeds is predicted. Harmony Search(HS)Through self-learning update, thereby improving the accuracy of the mechanical spindle vibration prediction model. The experimental results show that the HS- RNN -based mechanical spindle vibration prediction method proposed in this paper can automatically determine the network structure and accurately predict the vibration amplitude of the mechanical spindle.