Dottorato in INGEGNERIA ELETTRICA, DEI MATERIALI E DELLE
NANOTECNOLOGIE

Titolo della tesi: PIEZORESISTIVE GRAPHENE-BASED INKS FOR STRAIN AND PRESSURE SENSING THROUGH ELECTRICAL RESISTANCE TOMOGRAPHY-INSPIRED METHODOLOGIES AND ARTIFICIAL INTELLIGENT-ASSISTED PATTERNS RECOGNITION TECHNIQUES

The growing demand for advanced health monitoring systems has led to significant research into flexible, high-sensitivity sensor technologies capable of continuous and unobtrusive pressure and strain detection. This thesis explores the development of graphene-based piezoresistive coatings, integrating advanced nanomaterials, scalable fabrication techniques, and computational intelligence to enhance sensing performance. Two distinct sensor platforms were investigated: polymeric foams infused with graphene nanoplatelets (GNPs) and screen-printed GNP-based inks on fabrics, both engineered to provide flexible, high-resolution, and large-area sensing capabilities. These materials were systematically characterized for their electromechanical properties, with a particular focus on conductivity, mechanical response, and pressure sensitivity under varying conditions. To improve precision, functionality, and spatial resolution, Electrical Resistance Tomography (ERT)-inspired techniques were employed, enabling high-resolution pressure mapping without the need for discrete sensor arrays. Additionally, Machine Learning (ML) algorithms were integrated to enhance pattern recognition, strain detection, and posture classification, significantly improving accuracy, robustness, and adaptability. A key component of this approach was the use of Finite Element Method (FEM) simulations, which played a dual role: first, in modeling the mechanical behavior and electrical response of the piezoresistive coatings under deformation, and second, in generating synthetic datasets to complement experimental data. These simulations provided detailed strain and conductivity distributions, improving the training and calibration of ML models for more accurate strain estimation and sensor optimization. These advancements culminated in the development of a sensorized mat, capable of detecting pressure distributions and recognizing postures with exceptional sensitivity and precision. The findings of this research demonstrate the potential of scalable, intelligent piezoresistive coatings for transformative applications in healthcare monitoring, robotics, and beyond, paving the way for the next generation of adaptive and high-performance sensing technologies.