Enhancing the recognition accuracy of TENG-based vibration sensors through dielectric enhancement strategy

Current triboelectric nanogenerators (TENGs) reliably monitor vibration frequency, intensity, and environmental parameters, but their structural and operational constraints prevent accurate recognition of vibration waveforms. To overcome this limitation, we engineered a PDMS-encapsulated honeycomb nickel framework, in which the incorporation of nanoscale BaTiO 3 particles into the PDMS layer amplifies triboelectric output and improves waveform fidelity. Specifically, by enhancing the dielectric constant of the triboelectric layer, we achieved simultaneous improvement of TENG output performance and enrichment of the signal waveform, thereby increasing monitoring accuracy. Under compressive loading, interfacial interactions generate distinct contact and separation peaks in the triboelectric signal. By contrast, vibrational excitation produces a series of progressively attenuated peaks. The encapsulated structure is conducive to the device output stable waveform, highlighting its suitability for robust sensing in natural environments. Frequency monitoring exhibits high precision, with relative errors of 0.2–1.3%, 0–1%, 0.5–1% and 0.5–1% at 30, 50, 80 and 100 Hz, respectively, confirming accurate operation across diverse vibration regimes. Additionally, the encapsulated architecture ensures waveform stability under harsh conditions, underscoring its potential for reliable sensing in natural environments. Furthermore, the integration of a deep learning algorithm enables high-resolution classification of vibration states with an accuracy of 100% approximately, achieving effective monitoring of the operating status of the jaw crusher and vibrating screen. This work not only verifies the feasibility of designing a self-powered vibration sensor but also establishes a solid foundation for intelligent sensing in the era of AIoT.