Comparison of Several Data-driven Models for Remaining Useful Life Prediction

At present, there are many data-driven models for remaining useful life (RUL) prediction, but few literature compare the advantages and disadvantages of those models. Understanding the characteristics of different models is one of the key steps in model selection. Therefore, six typical data-driven models for remaining useful life prediction are chosen for experimental comparison. These models are linear regression (LR), support vector regression (SVR), random forest (RF), gradient boosting decision tree (GBDT), convolutional neural network (CNN) and recurrent neural network (RNN). A characteristics analysis is conducted to provide an intuitive result for model comparison. Then, the suggestion of model selection is given based on existing data and task requirement.