This work addresses a critical bottleneck in current self-powered vibration sensing for transmission lines, where existing methods rely on oversimplified mappings between vibration parameters and electrical signals, with lacking sustainable power for reliable deployment. First, a hybrid aeolian vibration harvester (HAVH) integrating a multilayer-elastic triboelectric nanogenerator (TENG) and a plug-cored electromagnetic generator (EMG) is proposed. Through structural optimization, the TENG achieves an open-circuit voltage of 498 V and a peak power of 4.4 mW in the 5–50 Hz, while the EMG supplies a short-circuit current of 23 mA and a peak power of 53 mW, enabling simultaneous energy harvesting and multimodal vibration signal acquisition. Second, nine vibration states are defined based on vibration frequency and amplitude characteristics. Triboelectric signals characterize their dynamic responses, and a mapping model is built by training a two-dimensional convolutional neural network (2D-CNN) on experimental data, which learns to extract discriminative features from triboelectric signals and establish the relationship between vibration excitations and corresponding output patterns. Additionally, an integrated power-management and signal-processing circuit is designed to provide stable power while enabling vibration signal processing. Finally, a self-powered system integrating the HAVH, circuit module, and upper computer demonstrates 99.5% recognition accuracy for all nine vibration states with graded warning, offering a viable approach for intelligent sensing of power transmission systems.


扫码关注我们
求助内容:
应助结果提醒方式:

