Integrating Embedded Electrical Sensing, Edge AI Inference, and Digital-Twin Power Management in a Translucent Polymer-Composite Smart Facade
Abstract
functions within a single element; however, no validated platform currently achieves this without performance degradation. This study
presents a 30 mm-thick glass-fiber-reinforced polymer mesh–enhanced translucent polymer-concrete (GFRP-TRPC) panel. Embedding a fiber
Bragg grating (FBG) sensor network, 4, 096-channel PWM LED system, and edge AI unit evaluates coexistence of sensing, power management, and structural integrity. Vacuum-assisted resin infusion (VARI) fabricated at 95 kPa and 60°C using epoxyvinyl ester concrete with 30
vol% recycled glass aggregate and 1 mm PMMA fibers yielded <1.2% void fraction, 65 MPa compressive strength, and 22% optical transmittance. TinyML CNN on an NVIDIA Jetson AGX Orin processed FBGs with 9 ms latency, detecting 0.1 mm cracks with 96.7% accuracy and
triggering safety overrides within 200 ms. Additionally, a Vision Transformer–GAN converted skeleton streams into 30 fps light patterns,
increasing visitor dwell time by 340%, consuming 32% less power than static illumination. A calibrated digital twin predicted 7.8% bendingmodulus loss and 4.6% transmittance loss after 30 years of Shanghai climate loading, predicting maintenance at year 26.4 with 4.8% error
against validation data. This demonstrates how co-designed sensing, AI, and power management sustain all four functions across a 30-year
horizon without compromise.
Keywords
Full Text:
PDFReferences
[1] Chiadighikaobi, P. C., Adegoke, M. A., Kharun, M., Paul, V. J., Abu Mahadi, M. I., & Finbarrs-Ezema, B. (2023). A review of the structural properties of translucent concrete as sustainable material. The Open Construction & Building Technology Journal, 17(1).
[2] Chandra, S., & Ohama, Y. (2020). Polymers in concrete. CRC Press.
[3] Zhang, W., & Wang, D. (2020). Damage identification using deep learning and long-gauge fiber Bragg grating sensors. Applied Optics,
59(33), 10532–10540.
[4] Heywood, H. (2019). 101 rules of thumb for low energy architecture. RIBA Publishing.
[5] Tao, F., Qi, Q., Liu, A., & Kusiak, A. (2018). Data-driven smart manufacturing. Journal of Manufacturing Systems, 48, 157–169.
[6] Boje, C., Guerriero, A., Kubicki, S., & Rezgui, Y. (2020). Towards a semantic construction digital twin: Directions for future research.
Automation in Construction, 114, 103179.
[7] Rasheed, A., San, O., & Kvamsdal, T. (2020). Digital twin: Values, challenges and enablers. IEEE Access, 8, 21980–22012.
[8] Ali, M., Ali, M., Hussain, M., & Koundal, D. (2025). Generative adversarial networks (GANs) for medical image processing: Recent
advancements. Archives of Computational Methods in Engineering, 32(2), 1185–1198.
[9] Hai, N. M., Quang, P. D., & Nam, H. P. (2025). Evaluation of lighting energy savings and embodied carbon of light-transmitting concrete incorporating polymethyl methacrylate for building envelopes. Journal of Building Engineering, 113807.
[10] Amasyali, K., & El-Gohary, N. M. (2018). A review of data-driven building energy consumption prediction studies. Renewable and Sustainable Energy Reviews, 81, 1192–1205.
[11] Wei, Y., Zhang, X., Shi, Y., Xia, L., Pan, S., Wu, J., & Zhao, X. (2018). A review of data-driven approaches for prediction and classification of building energy consumption. Renewable and Sustainable Energy Reviews, 82, 1027–1047.
[12] Dirar, S., Sogut, K., Caro, M., Rahman, R., Theofanous, M., & Faramarzi, A. (2025). Effect of shear span-to-effective depth ratio
and FRP material type on the behaviour of RC T-beams strengthened in shear with embedded FRP bars. Engineering Structures, 332,
120105.
[13] Cerny, T., Abdelfattah, A. S., Yero, J., & Taibi, D. (2024). From static code analysis to visual models of microservice architecture. Cluster Computing, 27(4), 4145–4170.
[14] Paul, R. (Ed.). (2019). High performance technical textiles. John Wiley & Sons.
[15] Wu, S., Chen, Q., Hu, C., An, Y., & Han, F. (2026). Digital-twin enabled "Living Fresco." ACS Omega, 11(5), 8011–8023.
DOI: http://dx.doi.org/10.70711/frim.v4i9.10226
Refbacks
- There are currently no refbacks.
