A self-powered triboelectric nanosensor based on track vibration energy harvesting for smart railway

Rail transport plays a major role in the development of a nation's economy. Due to the high maintenance requirements of train tracks, traditional monitoring sensors need to be connected to the power grid. The rail surface environment is complex, and there is a lack of power supply equipment. Therefore, a track vibration energy harvester-based self-powered triboelectric nanosensor (TVH-TENS) is designed in this paper. The TVH-TENS system has five modules: motion transformation, rectification correction, dual channel power generation, energy storage and deep learning. The motion transformation module uses a bevel gear set with one-way bearings to transform the track's two-way linear vibration into one-way rotational motion, addressing both circuit rectification and motion transformation issues simultaneously. The voltage signal output of the triboelectric generator is used for deep learning to classify variables and live monitoring. Experimental results reveal that the TVH-TENS system achieves a mean power output of 6.69 W with sinusoidal input of 6 mm amplitude, 6 Hz frequency and 3 Omega external load in MTS bench experiments. The deep learning accuracy of each variable exceeds 98.3 %. The high-performance TVH-TENS can power wireless sensor networks by harvesting vibration energy while also acting as a monitoring sensor. This system provides a reference method framework for intelligent track.