Loading [MathJax]/extensions/MathZoom.js
A Dual-Scale Transformer-Based Remaining Useful Life Prediction Model in Industrial Internet of Things | IEEE Journals & Magazine | IEEE Xplore

A Dual-Scale Transformer-Based Remaining Useful Life Prediction Model in Industrial Internet of Things


Abstract:

With recent advances of Industrial Internet of Things (IIoT), the connectivity and data collection capabilities of industrial equipment have be significantly enhanced, ye...Show More

Abstract:

With recent advances of Industrial Internet of Things (IIoT), the connectivity and data collection capabilities of industrial equipment have be significantly enhanced, yet bringing new challenges for the remaining useful life (RUL) prediction. To fulfill the RUL predicting demand in multivariate time series, this work proposes an encoder–decoder model termed as dual-scale transformer model (DSFormer), built upon the Transformer architecture. First, in the encoder part, a dual-attention module is designed for the weight feature extraction from both dimensions of the sensor and time series, aiming to compensate for the diverse impacts of different sensors on the prediction. Next, a temporal convolutional network (TCN) module is introduced to capture sequence features and alleviate the loss of positional information incurred by stacking blocks. Then, the feature decomposition module is integrated into the decoder for trend feature extraction from sequences, providing the model with additional sequence information. Finally, compared to existing models, the proposed method can obtain the superior performance in terms of the root mean square error (RMSE) and Score metrics on the FD001, FD002 and FD003 subsets of the C-MAPSS data set, with an average improvement of 3.2% and 2.5%, respectively. In particular, the ablation experiment further validates the effectiveness of proposed modules in handling multivariate time series and extracting features.
Published in: IEEE Internet of Things Journal ( Volume: 11, Issue: 16, 15 August 2024)
Page(s): 26656 - 26667
Date of Publication: 18 March 2024

ISSN Information:

Funding Agency:


I. Introduction

Industrial Internet of Things (IIoT) and digital twins (DTs) have emerged as two significant driving forces leading the industrial revolution [1], [2]. IIoT enable the intelligent manufacturing, intellectual management, and production optimization in the factory, by connecting sensors, devices, and production lines to the Internet, and thereby facilitating the real-time data collection and exchange [3], [4], [5], [6]. Besides, the DT, as the crucial component of the IIoT, utilizes digital technologies and data models to connect physical entities with their virtual counterparts, promoting the instantaneous monitoring, simulation, and prediction on physical entities. Thus, the production efficiency, cost reduction and product quality are further improved. In particular, DTs have been extensively applied in various fields [7], [8], [9], [10]. In the urban planning, by establishing a digital model of the city, DTs can simulate urban traffic flow, energy consumption and environmental pollution, contributing to the optimization of urban infrastructure layout and resource allocation [11]. Further, in the agriculture, crop growth environments and requirements can be simulated, rendering wiser decisions [12]. Besides, by leveraging DTs in the industrial manufacturing sector, one can promote the monitoring and optimization of production line operations, the prediction of maintenance needs, the improvement of equipment utilization, and the reduction of failure rates [13], [14].

References

References is not available for this document.