Multi-scale modeling in thermal conductivity of Polyurethane incorporated with Phase Change Materials using Physics-Informed Neural Networks

• DOI: 10.1016/j.renene.2023.119565
• Publication Date: 2023-11-15T07:26:59Z
• Authors: Bokai Liu; Yizheng Wang; Timon Rabczuk; Thomas Olofsson; Weizhuo Lu
• Keywords: Building envelope; Phase-change material; Material Design

Polyurethane (PU) possesses excellent thermal properties, making it an ideal material for insulation. Incorporating Phase Change Materials (PCMs) capsules into has proven to be effective strategy enhancing building envelopes. This innovative design substantially enhances indoor stability and minimizes fluctuations in air temperature. To investigate the conductivity of Polyurethane-Phase foam composite, we propose a hierarchical multi-scale model utilizing Physics-Informed Neural Networks (PINNs). allows accurate prediction analysis material's at both meso-scale macro-scale. By leveraging integration physics-based knowledge data-driven learning offered by Networks, effectively tackle inverse problems address complex phenomena. Furthermore, obtained data facilitates optimization design. fully consider occupants' comfort within envelope, conduct case study evaluating performance this optimized detached house. Simultaneously, predict energy consumption associated with scenario. All outcomes demonstrate promising nature design, enabling passive significantly improving comfort. The successful development Networks-based holds immense potential advancing our understanding Material's properties. It can contribute materials various practical applications, including storage systems insulation advanced