A deep learning framework for adaptive compressive sensing of high‐speed train vibration responses

Abstract

Onboard monitoring plays an important role in real-time condition assessment of rail systems. However, the data amount is typically tremendous due to the high sampling rate needed and long traveling distance, especially for vibration data collected from high-speed trains (HSTs). As for fault diagnosis of mechanical systems, compressive sensing (CS) has been increasingly adopted to reduce the data amount. In comparison to rotary bearings and bolted joints in machinery that operate in relatively steady working environments, HSTs run in an open and varying environment throughout the traveling mileage, and the data amount is normally much larger, making it tricky to directly apply the classical CS methods. This study aims to bridge the gap by investigating the sparsity of HST vibration signals and CS approaches. Considering the lack of sparsity and long reconstruction time, we propose an efficient adaptive CS approach for dynamic responses of HSTs. More specifically, we unroll the iterative soft thresholding algorithm (ISTA) in a deep learning (DL) framework and configure it into a data reconstruction machine. Compared to the conventional CS methods, our approach exhibits two advantages: (i) The dictionary learning and signal reconstruction are integrated into one neural network and can be conducted in an end-to-end manner; (ii) the process is highly efficient since encapsulating ISTA in a DL framework can naturally leverage the capability of GPU. The proposed approach is validated using data collected from an in-service HST, and results show that our approach achieves superior reconstruction performance over fixed bases and redundant dictionaries.adaptive compressive sensing, deep learning, high-speed train, onboard monitoring, sparse coding | INTRODUCTIONCondition monitoring is vital for the safety and punctuality of railway systems, and various onboard monitoring systems have been implemented on high-speed trains (HSTs) collecting real-time data of key components. Specifically,

Distributed Compressive Sensing for Wireless Signal Transmission in Structural Health Monitoring: An Adaptive Hierarchical Bayesian Model-Based Approach

“…This indicates that CS can not only maintain robustness to data loss but also facilitate data compression for energy efficiency. In SHM, related work has been carried out on vibration signals with high sparsity in the frequency domain [ 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 ]. Bao et al [ 22 ], O’Connor et al [ 23 ], Klis et al [ 24 ], Jayawardhana et al [ 12 ], and Wan et al [ 25 ] implemented CS-based data compression and reconstruction on different types of vibration signals.…”

“…Huang et al [ 27 , 28 ] innovatively applied Bayesian compressive sensing (BCS) to SHM. Recently, an adaptive CS method incorporating deep learning has been explored for vibration data transmission in high-speed railroads with ideal results [ 29 ].…”

Core Needle Biopsy Guidance Based on Tissue Morphology Assessment with AI-OCT Imaging

“…Therefore, we investigated using user-assisted deep learning (DL), a subset of artificial intelligence (AI) based on deep neural networks, for rapid image analysis. AI has made remarkable breakthroughs in medical imaging, especially for image classification and pattern recognition [ 29 , 30 , 31 , 32 , 33 , 34 , 35 ]. Studies showed that OCT image evaluation by DL algorithms has achieved good performance for disease detection, prognosis prediction, and image quality control, suggesting that the use of DL technology could potentially enhance the efficiency of the clinical workflow [ 36 , 37 ].…”

Core Needle Biopsy Guidance Based on Tissue Morphology Assessment With AI-OCT Imaging

“…Therefore, we investigated using user-assisted deep learning (DL), a subset of artificial intelligence (AI) based on deep neural networks, for rapid image analysis. AI has made remarkable breakthroughs in medical imaging, especially for image classification and pattern recognition [29][30][31][32][33][34][35]. Studies showed that OCT image evaluation by DL algorithms has achieved good performance for disease detection, prognosis prediction, and image quality control, suggesting that the use of DL technology could potentially enhance the efficiency of clinical workflow [36,37].…”