A roller-bearing-based triboelectric nanosensor for freight train synergistic maintenance in smart transportation

Rail freight is a vital part of global economic development. The hunting instability of freight trains seriously affects safety. The lack of electrical pipelines between freight trains makes the energy of onboard monitoring equipment an urgent problem to be solved. In this paper, an energy self-consistent system (ESCS) based on a roller-bearing-based triboelectric nanosensor is proposed to detect the running status of freight trains while providing electric energy. The proposed ESCS consists of two modules: a roller-bearing-based triboelectric nanosensor (RB-TENS) module and a detection module. The RB-TENS module innovatively combines the advantages of low damping and wear resistance of roller bearings. Moreover, it incorporates energy harvesting and running status sensing for freight trains. While converting the train’s vibrational kinetic energy into electrical energy and storing it, the electrical signal containing the characteristics of the train’s running status is collected. The detection module includes a pre-processing module and a deep learning model based on LSTM. The characteristic electrical signal collected by RB-TENS is enhanced and extracted by the pre-processing module to generate training and testing sets. Using data set training and testing deep learning model, the freight train running status was recognized. On this basis, a dynamic model is established to study the effects of different axle loads, speeds, and pre-processing parameters on the vibration response, electrical performance, and deep learning model of the ESCS. Experiments show that the peak output power and energy density of RB-TENS reach 1.9 μW and 72 mW/m3. The results of the testing set show that the detection accuracy of ESCS reaches 96.6%, indicating that it can effectively detect the hunting instability of freight trains. This ESCS enables TENS to present a breakthrough in smart transportation and practical applications of the zero-energy Internet of Things.