An AI-driven electromagnetic-triboelectric self-powered and vibration-sensing system for smart transportation

Developing energy recovery technologies in the transportation sector is crucial for reducing global carbon emissions and protecting the environment. Particularly significant is the research on self-powered and sensing technologies for train transportation, especially freight trains. This work introduces an AI-driven electromagnetic-triboelectric self-powered and vibration-sensing system (ESVS) designed for energy recovery and the sensing of inter-carriage vibrations in trains. The system primarily comprises a rope-driven motion conversion mechanism, a triboelectric nanogenerator, and a coaxial reversing electromagnetic generator. In verification, the electromagnetic generator demonstrated an impressive output power improvement rate of up to 547.13 % compared to conventional generators. It achieved effective and peak output powers of 186.89 mW and 0.47 W, respectively. The system successfully powered the sensor system by charging a battery with 108 mAh within 10 min. Additionally, a train status diagnostic system, which leverages sensing from a triboelectric nanogenerator and employs Artificial Intelligence (deep learning algorithms), was proposed, boasting a recognition accuracy of 99.78 %. Vibration tests conducted for a train scenario confirmed the feasibility of the system. The proposed ESVS offers the advantages of versatility and host-friendliness, providing an effective and practical solution for self-powered and sensing applications in smart transportation. •The proposed system serves the monitoring of smart transportation.•A rope-driven motion conversion mechanism is proposed.•Coaxial reversing mode achieves 0.47 W output power with a 547.13 % improvement.•An AI-driven status diagnostic system achieves high recognition accuracy.