Implementation of artificial neural network in a building benefits from radiant floor heating /cooling enhanced by phase change materials. (January 2023)

• Record Type: Journal Article
• Authors: Hai, Tao; Dhahad, Hayder A.; Zhou, Jincheng; Abdelrahman, Anas; Almojil, Sattam Fahad; Almohana, Abdulaziz Ibrahim; Alali, Abdulrhman Fahmi; Kh, Teeba Ismail; Sharma, Kamal; Ali, Masood Ashraf; Almoalimi, Khaled Twfiq
• Is Part Of: Engineering analysis with boundary elements. Volume 146(2023)
• Page Start: 66
• Page End: 79
• Publication Date: 2023-01
• Subjects: Building -- Radiant floor -- PCM -- ANN -- Numerical analysis; Boundary element methods -- Periodicals; Engineering mathematics -- Periodicals; Équations intégrales de frontière, Méthodes des -- Périodiques; Mathématiques de l'ingénieur -- Périodiques; Boundary element methods; Engineering mathematics; Periodicals; 620.00151
• DOI: 10.1016/j.enganabound.2022.10.012
• Languages: English
• ISSNs: 0955-7997

Abstract

Abstract: In this study, the numerical analysis of the radiant floor system was investigated for a building in the presence of PCM inside the external walls as well as the roof at a thickness of 2 cm. By injecting cold/warm fluid into the radiant tubes inside the roof, the cooling/heating requirements were met. Several PCMs with identical thermal properties (except melting point) were selected and based on numerical analysis, the energy utilization in the heating/cooling sections was evaluated by comparison with the simple building (without PCM). Four main variables were defined for the neural network, and energy consumption was trended for two climate zones, Shenyang (41.7922°N, 123.4328°E), and Zhengzhou (34.7578°N, 113.6486° E). For each region, the PCM with the best phase transition was selected and it was realized that for the first region, energy consumption was diminished by 12.6% and for the second region by 15.9%. According to the temperature conditions and radiation intensity in the environment, the ANN could forecast annual energy utilization with an error of less than 6%.