Remaining useful life estimation via Cascaded Self-Attention and ResNet models

Abstract

Prognostics and Health Management occupies an important place in modern industrial maintenance to increase the reliability of systems. Maintenance of critical parts in the system is vital for successful prognostics and health management. For this reason, the determining remaining useful life of the parts should be accurate. This study proposes a data-based remaining useful life prediction method with a network consisting of a cascade-connected Self-Attention and Residual Network layer. The network is fed by multiple sensor signals to monitor the aero-engines. The proposed model contains four main parts Gaussian Noise layer, Self-Attention layer, Residual Network layer, and remaining useful life estimation. The Gaussian Noise layer deals with the noisy input data for a more robust predictor. The Self-Attention layer focuses on the crucial points through time. The Residual Network layer uses feature extraction and makes the model more profound help of the skip connection. Finally, the remaining useful life estimation is made with the highly correlated features obtained from the fully connected layer and the output layer. In addition, a new loss function has been offered in accordance with the evaluation metrics in the literature. With the proposed model and loss function, 11,017 and 12,629 in root mean square error, 157.19 and 218.6 in score function were obtained in the FD001 and FD003, respectively. The superior performance of these results on the C-MAPSS dataset is demonstrated by comparing the other state-of-the-art methods in the literature.