Particle-Based Triboelectric Vibration Sensor for Speech Recognition and Remote Machine Condition Monitoring

Self-powered vibration sensors, particularly those utilising triboelectric nanogenerators, hold great promise for real-time monitoring of dynamic loads and structural health without external power sources. Based on the triboelectric effect and electrostatic induction, a triboelectric vibration sensor can convert mechanical energy into electrical energy. This technology holds significant promise for diverse applications such as voice recognition, real-time dynamic load monitoring, and structural damage or degradation detection. However, significant challenges exist in designing TENG vibration sensors that can operate over a wide frequency range and overcome electromagnetic interference and signal attenuation issues during long-distance transmission. To address these challenges, I introduce a high-bandwidth, self-powered triboelectric particle-surface vibration sensor. The sensor comprises a flat round capsule filled with dielectric particles, with an inner tribo-layer backed by electrodes. When attached to a vibrating structure, particle impacts on the inner surfaces cause opposite charges to form, generating an alternating voltage through the dynamic motion of the charged particles. The sensor exhibits a flat frequency response ranging from 500 Hz to 7000 Hz, significantly exceeding the bandwidths of recently reported triboelectric vibration sensors. Through further structural optimisation, introducing a single-particle cell has enabled the sensor to achieve similar vibration sensitivity at different angles to gravity while maintaining the flat-frequency response of the original design across a wide frequency range. Applications in voice recognition and monitoring mechanical operating conditions have demonstrated the potential of the new vibration sensor. A novel optical method is introduced to address the challenge of long-distance signal attenuation in measuring the outputs of TENG sensors using electrical cables. This approach integrates a triboelectric vibration sensor with an optical transducer. The alternating voltage generated by the triboelectric nanogenerator modulates the optical reflectivity of a thin liquid crystal layer within the optical transducer. A photodetector detects the change in the intensity of the reflected light, which is proportional to the electrical voltage of the triboelectric nanogenerator. This sensor can operate without a local power supply and enables long-distance monitoring while minimising signal loss. Since the optical signal propagates within the fibre, it is immune to electromagnetic interference, and signal attenuation over the same distance is significantly lower than that of electrical wires. Specifically, this sensor achieves a signal attenuation of 1.05 dB per kilometre, considerably lower than the 11.73 dB attenuation level of a 60-meter coaxial cable. The optical triboelectric sensor system also demonstrates a signal-to-noise ratio of 18.5 (at a distance of 1000 meters), which is markedly higher than the 3.21 dB value for 60-meter electrical wires. These results indicate its great potential for remote load and structural health monitoring, as it does not require a local power supply at the sensor location and is unaffected by electromagnetic interference.

Grantor dc:publisher

UNSW, Sydney

Year dc:date

2025

Author dc:creator

• Gao, Shanshi

• Particle-Based Triboelectric Vibration Sensor
• Speech Recognition
• Remote Machine Condition Monitoring

Statement dc:rights

• embargoed access
• CC BY 4.0

Licence

• http://purl.org/coar/access_right/c_f1cf
• https://creativecommons.org/licenses/by/4.0/

Language dc:language

en

Identifier

https://doi.org/10.26190/unsworks/30841

OAI identifier oai:identifier

oai:unsworks.library.unsw.edu.au:1959.4/104044