Smart sensing and data-driven characterization of asphalt response
My doctoral project, SN4SI — Sensors Network for Smart Self-Sustaining Interactive Infrastructures, is part of CIRCOPAV WP1 — Smart and Connected Assets. My work includes experimental measurement, sensor integration, data acquisition, signal processing, material characterization and machine-learning model evaluation.
Sensor Integration & DAQ
Development and implementation of a Raspberry Pi-based acquisition workflow for ADXL355 accelerometer measurements and comparison with laboratory reference measurements.
Signal Processing
Data cleaning, digital filtering, FFT, PSD, sinusoidal curve fitting, frequency extraction, amplitude estimation, phase-angle determination and automated processing.
Machine Learning
Machine-learning workflows are evaluated for predicting dynamic/complex modulus and phase angle, and for exploring sensor-calibration approaches.
Sensor Calibration & Validation
Comparison with reference measurements using synchronization, amplitude and phase comparison, frequency-response evaluation, repeatability and signal-quality assessment.
Controlled Experimental Testing
Cyclic laboratory testing with MTS force measurements, LVDT displacement, temperature measurements and MEMS accelerometers under different temperature and loading-frequency conditions.
Viscoelastic Characterization
Temperature- and frequency-dependent analysis of asphalt response, including dynamic/complex modulus and phase angle, alongside conventional viscoelastic modelling and machine-learning model evaluation.
Physical measurement → signal interpretation → model evaluation
Asphalt material testing is the application domain for the sensing, data-acquisition, signal-processing, validation and machine-learning work presented here.
