A novel high-order optimization approach using block-based machine learning (BBML): An investigation of phase-change material (PCM) encapsulation in pore-scale porous Trombe walls

The simulation of flow patterns and the thermal behavior of pore-scale porous media (PSPM) require multilayer calculation when the simulation domain is large. In this study, multi-layer calculations based on the tensor-based data transfer protocol (TBDTP) were used to simulate the flow pattern and the thermal behavior of PSPM walls, including phase-change materials (PCMs) in Trombe walls. A study of the thermal behavior of PSPM walls over a 24-hour period was conducted with a steady-state flow passing through the wall. Constant-concentration PCMs help to reduce the temperature gradients between day and night. Therefore, a novel solution for high-order optimization was developed by using block-based machine learning (BBML) and numerical simulation. High-order optimization helps us to find the optimum PSPM wall porosity and PCM concentration from a wide range of options ( 10 % < ε < 90 % , 0 < C P C M ( × 10 3 n u m b e r / c m 3 ) < 9 ) . As a result, we found that the optimal combination of the PCM concentration and the wall porosity was ε = 48 % , C P C M = 7.1 × 10 3 n u m b e r / c m 3 . Thus, PCMs reduce a solar radiation–heated wall (outer side) by 5.2%. Furthermore, the temperature gradient over the course of the day is 6.34% lower than it is when PCMs are not used. For the period from 17:00 to 06:00, the average temperature of the PSPM wall with PCMs is up to 6.64% higher than it is without PCMs. • A PCM-based Trombe wall is modelled. • Machine learning and CFD simulation are used. • Using PCM in the wall can increase the wall temperature about 6.6% in night.