Welcome to

Lancashire Online Knowledge

Image Credit Header image: Artwork by Professor Lubaina Himid, CBE. Photo: @Denise Swanson


Comprehensive techno-economic analysis of phase change material-stabilized building-integrated photovoltaic thermal façades in residential buildings: a comparative machine learning approach for hot-arid climates toward less carbon emission

Almazam, Khaled, Nasrallah, Eman orcid iconORCID: 0000-0001-8183-1903, Bashir, Faizah Mohammed, Mahmoud, Abubakar Sadiq, Dodo, Yakubu Aminu and Andiye, Yohannes Mehari orcid iconORCID: 0009-0002-4129-9259 (2026) Comprehensive techno-economic analysis of phase change material-stabilized building-integrated photovoltaic thermal façades in residential buildings: a comparative machine learning approach for hot-arid climates toward less carbon emission. International Journal of Low-Carbon Technologies, 21 . ISSN 1748-1317

[thumbnail of VOR]
Preview
PDF (VOR) - Published Version
Available under License Creative Commons Attribution.

2MB

Official URL: https://doi.org/10.1093/ijlct%2Fctag108

Abstract

Building-integrated photovoltaic-thermal (BIPVT) façades can help hot-arid cities use limited envelope area for both electricity and domestic hot water production, but their assessment is often weakened by uncertain climate inputs and weak operational forecasting. To address these gaps, this study developed a simulation-based framework for a phase change material-stabilized BIPVT façade in Tabuk, Saudi Arabia. The framework combines satellite/reanalysis climate inputs, station-validated temperature and wind data, NASA Prediction of Worldwide Energy Resources–Copernicus Atmosphere Monitoring Service irradiance screening, clear-sky index checks, elastic net regression and Extreme Gradient Boosting (XGBoost) day-ahead forecasting, layer-resolved thermoelectrical modeling, EnergyPlus cross-validation, and life-cycle economic analysis. XGBoost gave the best irradiance forecast against the screened satellite irradiance series, with root mean squared error of 41.2 W·m−2 and daylight mean absolute percentage error of 8.6%. Increasing water flow from 0.02 to 0.10 kg·s−1 lowered photovoltaic temperature from ~52°C to 33°C and raised electrical efficiency from ~9% to nearly 16%. EnergyPlus deviations remained within ~5%, and east/west façades reached payback in 9–10 years.


Repository Staff Only: item control page