ABSTRACT
Multi-layer lightweight composite structures are widely used in the field of aviation and aerospace during the processes of manufacturing and use, and, as such, they inevitably produce defects, damage, and other quality problems, creating the need for timely non-destructive testing procedures and the convenient repair or replacement of quality problems related to the material. When using terahertz non-destructive testing technology to detect defects in multi-layer lightweight composite materials, due to the complexity of their structure and defect types, there are many signal characteristics of terahertz waves propagating in the structures, and there is no obvious rule behind them, resulting in a large gap between the recognition results and the actual ones. In this study, we introduced a U-Net-BiLSTM network that combines the strengths of the U-Net and BiLSTM networks. The U-Net network extracts the spatial features of THz signals, while the BiLSTM network captures their temporal features. By optimizing the network structure and various parameters, we obtained a model tailored to THz spectroscopy data. This model was subsequently employed for the identification and quantitative analysis of defects in multi-layer lightweight composite structures using THz non-destructive testing. The proposed U-Net-BiLSTM network achieved an accuracy of 99.45% in typical defect identification, with a comprehensive F1 score of 99.43%, outperforming the CNN, ResNet, U-Net, and BiLSTM networks. By leveraging defect classification and thickness recognition, this study successfully reconstructed three-dimensional THz defect images, thereby realizing quantitative defect detection.
ABSTRACT
The challenge to mitigate real-world emissions from vehicles calls for powerful in-use compliance supervision. The remote on-board diagnostic (OBD) approach, with wireless data communications, is one of the promising next-generation monitoring methods. We collected second-by-second profiles of carbon dioxide (CO2) and nitrogen oxides (NOX) emissions, driving conditions and engine performance for three conventional diesel and three hybrid diesel buses participating in a remote OBD pilot program in Nanjing, China. Our results showed that the average CO2 emissions for conventional diesel and hybrid diesel buses were 816 ± 83 g km-1 and 627 ± 54 g km-1, respectively, under a typical driving pattern. An operating mode binning analysis indicated that CO2 emissions reduction by series-parallel hybrid technology was largely because of the significant benefits of the technology under the modes of low speed and low power demand. However, significantly higher CO2 emissions were observed for conventional diesel buses during rush hours, higher than 1200 g km-1. The OBD data suggested no improvement in NOX emission reduction for hybrid buses compared with conventional buses; both were approximately 12 g km-1 because of poor performance of the selective catalyst reduction (SCR) systems in the real world. Speed-dependent functions for real-world CO2 and NOX emissions were also constructed. The CO2 emissions of hybrid buses were much less sensitive to the average speed than conventional buses. If the average speed decreased from 20 km h-1 to 10 km h-1, the estimated CO2 emission factor for conventional buses would be increased by 34%. Such a change in speed would increase NOX emissions for conventional and hybrid buses by 38% and 56%, respectively. This paper demonstrates the useful features of the remote OBD system and can inform policy makers how to take advantage of these features in monitoring in-use vehicles.
