Rotating machinery fault diagnosis problems have been well addressed when sufficient supervised data of the tested machine are available using the latest data-driven methods. However, it is still challenging to develop effective diagnostic method with insufficient training data, which is highly demanded in real industrial scenarios since high-quality data are usually difficult and expensive to collect. Considering the underlying similarities of rotating machines, data mining on different but related equipments potentially benefit the diagnostic performance on the target machine. Therefore, a novel transfer learning method for diagnostics based on deep learning is proposed in this paper, where the diagnostic knowledge learned from sufficient supervised data of multiple rotating machines is transferred to the target equipment with domain adversarial training. The experimental results on four datasets validate the effectiveness of the proposed method, and show it is feasible and promising to explore different datasets to improve diagnostic performance.

Intelligent machinery prognostics and health management (PHM) methods have been attracting growing attention in the past years, with the rapid development of the artificial intelligence algorithms. The remaining useful life (RUL) prediction problem is critical in prognostics for optimization of the maintenance strategy. Despite the promising advances, the current algorithms basically assume the training and testing entities are operating under identical condition, which is less practical in the real industries. In the cross-domain PHM studies, domain adaptation techniques have been successfully applied for building generalized data-driven models. However, the availability of target-domain data in full life cycle is basically required by the existing methods. In most scenarios, only the target data at early degradation period can be obtained, that poses great challenges in transfer learning. This article proposes a partial domain adaptation method for RUL prediction with incomplete target-domain data. Deep neural network-based adversarial learning strategy is adopted as the main framework, and the source-domain instance-weighted degradation fusion scheme is proposed for conditional domain adaptation at similar degradation levels. The source outliers can be well filtered out in learning generalized features across domains. Experiments of machine run-to-failure tests are implemented for validation, and the results indicate the proposed methodology is well suited for practical cross-domain RUL predictions.

In the hot strip rolling (HSR) process, accurate prediction of bending force can improve the control accuracy of the strip crown and flatness, and further improve the strip shape quality. In this paper, six machine learning models, including Artificial Neural Network (ANN), Support Vector Machine (SVR), Classification and Regression Tree (CART), Bagging Regression Tree (BRT), Least Absolute Shrinkage and Selection operator (LASSO), and Gaussian Process Regression (GPR), were applied to predict the bending force in the HSR process. A comparative experiment was carried out based on a real-life dataset, and the prediction performance of the six models was analyzed from prediction accuracy, stability, and computational cost. The prediction performance of the six models was assessed using three evaluation metrics of root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R²). The results show that the GPR model is considered as the optimal model for bending force prediction with the best prediction accuracy, better stability, and acceptable computational cost. The prediction accuracy and stability of CART and ANN are slightly lower than that of GPR. Although BRT also shows a good combination of prediction accuracy and computational cost, the stability of BRT is the worst in the six models. SVM not only has poor prediction accuracy, but also has the highest computational cost while LASSO showed the worst prediction accuracy.