IoT-Integrated WCIP-Based Modeling and Machine Learning Optimization of a THz Hydrogen Sensor for Green Hydrogen Monitoring

This article presents a high-sensitivity terahertz (THz) hydrogen smart sensor based on three laterally cascaded spiral resonators functionalized with graphene, black phosphorus (BP), and vanadium dioxide (VO2). The electromagnetic behavior is rigorously modeled using an extended wave concept iterative process (WCIP), enabling efficient full-wave characterization of tunable 2-D materials without volumetric meshing. Hydrogen detection is achieved through resonance frequency shifts induced by refractive-index perturbations over concentrations spanning trace ppm levels to sub-lower-explosive-limit (LEL) regimes. The optimized architecture achieves sensitivities ranging from 8.4 to 9.03 THz/RIU while preserving high quality factors and sharp spectral features. To enable rapid parametric exploration, a gradient boosting regressor (GBR) surrogate model is developed to capture the nonlinear coupling between material conductivities and sensing performance, demonstrating strong predictive accuracy (R² > 0.996). The proposed smart sensor is designed for seamless integration with Internet-of-Things (IoT) frameworks, enabling real-time data acquisition, intelligent processing, and scalable deployment in advanced industrial environments. This smart sensing platform supports real-time hydrogen monitoring, hydrogen storage safety, and intelligent inspection of hydrogen infrastructure. Furthermore, integration within a 6G-enabled IoT ecosystem facilitates predictive analytics, automated monitoring, and efficient infrastructure management. The proposed system contributes to sustainable infrastructure development by enabling reliable hydrogen monitoring, predictive maintenance, and enhanced industrial safety. Overall, the proposed platform establishes a scalable pathway toward next-generation THz smart sensor technologies by synergistically combining physics-based electromagnetic modeling and data-driven optimization for industrial innovation and smart infrastructure applications.