A Mechanism-Data Driven Fault Prediction Method Based on Cloud-Terminal Fusion

The feed axis is a key component of machine tools. Once the feed axis fails, it will have a huge impact on the overall performance of machine tool. The traditional fault prediction methods of feed axis mostly judge according to the operating mechanism, but due to the increasing complexity of components, it has become more and more difficult to establish the operational mechanism model. With the application of machine learning methods in the field of fault prediction, the data-driven method has made great progress, but the accuracy of prediction is limited. Based on the two reasons above, this paper proposes a mechanism-date driven fault prediction method of feed axis. The data-driven model requires a large amount of historical data for training, but the storage capacity and computing power of the terminal device are limited. Therefore, the cloud-terminal fusion architecture is used to implement the mentioned fault prediction model above.