上海科技大学知识管理系统

IntelliSense fabrics that can sense transient mechanical stimuli are widely anticipated in flexible and wearable electronics. However, most IntelliSense fabrics developed so far are only sensitive to quasi-static forces, such as stretching, bending, or twisting. In this work, a sheath-core triboelectric nanogenerator (SC-TENG) yarn was developed via a rational design, electroassisted core spinning technique, that consisted of a rough nanoscale dielectric surface and mechanically strong and electrically conductive core yarns. The resulting system was used to sense and distinguish the instantaneous mechanical stimuli generated by different materials. To further improve the sensing accuracy, a machine learning model, based on a classification coding and recurrent neural network, was built to predict the type of contact materials from the peak profiles of output voltages. With these experimental and algorithmic optimizations, we finally used SC-TENG yarn to identify the type of materials in real-time. Moreover, by applying Internet of Things techniques, we investigated that SC-TENG yarn could be integrated into an IntelliSense system to recognize and control various electronic and electrical systems, demonstrating promising applications in wearable energy supply, IntelliSense fabrics, and human-machine interactions.