Liwei Lin (Advisor)

This work presents a new scheme to measure both fluidic viscosity and density simultaneously using PMUTs (piezoelectric micromachined ultrasonic transducers) with the assistance of machine learning. Advancements as compared to the state-of-art works include: (1) using PMUT pulse-echo signals to extract multiple fluid properties simultaneously with the assistance of machine learning; (2) differentiating viscosity and density successfully; and (3) sensing results with an error of 1.6 ± 1.57 % for viscosity and 0.20 ± 0.17 % for density for testing fluids in the viscosity range of 0.9 to 1.9...

This paper reports drastically improved accuracy of real-time gas detections by a scalable machine learning fusion system which utilizes encoder-decoder structures and a decision fusion model. Advancements as compared to the state-of-the-art works include: 1) increased gas sensing accuracy of a carbon dioxide sensor with a Mean Absolute Percentage Error (MAPE) of 2.97% in real-time tests; 2) drastically reduced sensor response/recovery time from ~9 mins to 2 mins; and 3) good proof-of-concept demonstrations for both generalization and robustness across different gas sensors. As such, this...

This work reports a platform based on ultrasound for mid-air particle manipulations using a 2×2 piezoelectric micromachined ultrasonic transducer (pMUT) array. Three achievements have been demonstrated as compared to the state-of-art: (1) high SPL (sound pressure level) of 120 dB at a distance 12 mm away by an individual lithium-niobate pMUT; (2) a numerically simulated and experimentally demonstrated 2D focal point control scheme by adjusting the phase-delay of individual pMUTs; and (3) the experimental demonstration of moving a 0.7 mg foam plastic particle of 12 mm away in the mid-air by...