Numerical Study on the Thermal and Optical Performances of an Aerogel Glazing System with the Multivariable Optimization Using an Advanced Machine Learning Algorithm

The implementation of advanced materials in high-efficient glazing system is important for green buildings. In this study, aerogel granules are implemented in the glazing system to form a translucent window with super-insulating performance. An experimentally validated numerical modeling integrating both heat transfer model and optical model is developed to characterize the sophisticated heat transfer and solar radiation transmission mechanisms. Sensitivity analysis is presented with quantifiable contribution ratio of each parameter to the total heat gain. Instead of returning back to numerical modeling repeatedly, an advanced optimization engine implemented with a generic optimization methodology with competitive computational efficiency and accuracy is proposed by implementing the supervised machine learning and advanced optimization algorithms. The research results show that the developed artificial neural network modeling is more accurate and computational-efficient than the traditional lsqcurvefit fitting methodology. In addition, the optimal case through the teaching-learning-based optimization is more robust and competitive than the optimal case through the particle swarm optimization in terms of the total heat gain. This study presents an in-depth understanding of heat transfer and solar radiation transmission of nanoporous aerogel granules together with a robust optimal design, which is important for the promotion of green buildings with high-energy performance.