National Key Laboratory of Science and Technology for National Defence on Advanced Composites in Special Environments, Harbin Institute of Technology, Harbin 150001, China
Dr. Qiang Yang
National Key Laboratory of Science and Technology for National Defence on Advanced Composites in Special Environments, Harbin Institute of Technology, Harbin 150001, China
Integrated thermal protection systems with both thermal protection and load-bearing capabilities are regarded as one of the most promising thermal protection concepts for future hypersonic vehicles. Yet improper design could cause the structure to be overweight and even fail prematurely during service. In this paper, the thermal–mechanical analysis model of integrated thermal protection systems is first established, and the thermal short and mismatch effects of typical corrugated web concept are studied. Three improved structural concepts are proposed. Thermal insulation capability and analyses of structural strength, stability, and stiffness are carried out on both the original and improved concepts. The results show that the web is a key component and that structural performance can be enhanced through web redesign. Then, a performance evaluation method is presented using the fuzzy decision-making model. Besides the aforementioned performance factors, two more factors with respect to structural complexity are considered. A comparison made between different concepts reveals that a structure with lightening holes is a promising concept that combines both structural efficiency and structural simplicity. The fuzzy decision-making model provides a comprehensive evaluation method by combining quantitative and qualitative information.
Intelligent composite structures with self-aware functions are preferable for future aircrafts. The real-time location of damaged areas of composites is a key step. In this study, deep transfer learning was used to achieve the real-time location of damaged areas. The sensor network obtained acoustic emission signals from different damaged areas of the aluminum alloy plate. The acoustic emission time-domain signal is transformed into the input image by continuous wavelet transform. The convolutional neural network-based model automatically localized the damaged area by extracting features from the input image. A small amount of composite acoustic emission data was used to fine-tune some network parameters of the basic model through transfer learning. This enabled the model to classify the damaged area of composites. The accuracy of the transfer learning model trained with 900 samples is 96.38%, which is comparable to the accuracy of the model trained directly with 1800 samples; the training time of the former is only 17.68% of that of the latter. The proposed method can be easily adapted to new composite structures using transfer learning and a small dataset, providing a new idea for structural health monitoring.