A wearable frequency up-conversion-based piezoelectric-electromagnetic energy harvester for ultra-low and low frequency vibrations

Abstract

To address the challenge of discontinuous operating bandwidth in energy harvesters subjected to ultra-low-frequency human motion and low-frequency vibration environments, this study proposes a wearable frequency up-conversion piezoelectric–electromagnetic hybrid energy harvester (FUCPEEH). In both ultra-low and low-frequency ranges, bidirectional coupling between the piezoelectric and electromagnetic modules enables efficient harvesting of human motion energy as well as low-frequency ambient vibration energy. A theoretical model is established and validated experimentally, demonstrating good agreement between simulation and measurement. Experimental results show that the harvester operates effectively over a broad 1.6–10 Hz frequency range. At a running speed of 9 km h−1, the FUCPEEH delivers an output power of 95.8 mW and a power density of 177.4 μW cm−3. By processing the piezoelectric signals generated by the FUCPEEH under ultra-low-frequency excitation using deep learning, accurate motion-state recognition and user identity recognition are achieved, with accuracies of 99.79% and 99.53%, respectively. A series of electrical demonstrations, including driving an LED array and charging commercial wireless earbuds, further verifies its capability to power wearable devices. These results indicate that the proposed FUCPEEH has substantial potential for applications in smart-city infrastructure and green intelligent living, such as health monitoring, and human–machine interaction.