CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Predication of Trombe wall thickness for best thermal performance using artificial neural network

عنوان مقاله: Predication of Trombe wall thickness for best thermal performance using artificial neural network
شناسه ملی مقاله: ICRSIE05_250
منتشر شده در پنجمین کنفرانس بین المللی پژوهش در علوم و مهندسی و دومین کنگره بین المللی عمران، معماری و شهرسازی آسیا در سال 1399
مشخصات نویسندگان مقاله:

Ahmad sayadi - Master of civil engineering shiraz azad university branch

خلاصه مقاله:
The Trombe wall system is a passive solar building technique that uses different characteristics in various configurations to provide an alternative method for cooling or heating the building. Mechanical heating, ventilation, and air conditioning systems need a source of energy, electricity or burning fuel to provide a proper Human Comfort Index which cause air pollution. So, these modern technologies for heating and cooling can be widely used instead of commercial methods. Hence, there are different methods to optimize Trombe wall design and improve performance of Trombe walls. In addition, some experimental works have been done for validation and verification of numerical and computational works. In this paper an algorithm has been developed to optimize performance of modified Trombe wall. A neural network has been applied to available experimental data and a smart algorithm has been trained. This trained system can intelligently determine the amount of water inside the wall. Therefore, based on the outdoor temperature, outdoor humidity and desired Human Comfort Index, the number of cells that need to be filled with water can be calculated. In summary, using advanced numerical methods such as training algorithms and networks can help structures to behave as “an Intelligent Structures”.

کلمات کلیدی:
Trombe Wall, Neural Network, Intelligent Structure, Human Comfort Index

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/1116137/