Towards a Sustainable Building

Calculate Thermal Loads and Find the Optimal Insulation Thickness for Your Building!

This web application allows you to calculate the thermal loads of your specific building and determine the optimal thermal insulation thickness instantly. The models employed here are powered by Artificial Neural Networks (ANNs), ensuring accuracy and efficiency.
Three ANN models with carefully designed architectures are utilized: one calculates the heating load, another predicts the cooling load and the third determines conversion coefficients. The first two models compute thermal loads based on the materials selected by the user. The third model adapts these thermal loads to match the geometry of various residential buildings, ensuring tailored results. These ANN models deliver precise heating and cooling load predictions with minimal Mean Squared Error (MSE).
Finally, the models are integrated to identify the economical thermal insulation thickness. "Economical thickness" is defined as the point where the cost of adding one centimeter of insulation exceeds the savings from reduced energy consumption over the building’s lifetime. This balance is determined using a detailed mathematical relationship: U_f × E_p × (EU_i+1 - EU_i) > I_p
Where U_f is the useful life, E_p is the energy price according to the fuel consumption tariff, I_p is the price of insulation in one centimeter of thickness, EU_i and EU_i+1 are the energy consumption base on i centimeter and i+1 centimeter insulation, respectively.

To perform the calculations, three categories of input data are required:

1. Building Geometry: You need to input details about the building's geometry, limited to the gross floor area, wall area, and window-to-wall ratio.
2. Material Properties: Specify the properties of the wall, windows, and HVAC system used in the building.
3. Expenditures: Enter the price of different energy types used by the HVAC system and the cost of thermal insulation materials.

This model is currently tailored for use in Tehran, Iran. It operates under the following assumptions:

• The heating setpoint temperature is assuming 20°C.
• The cooling setpoint temperature is assuming 25°C.
• The R_roof, and the R_floor connected with the first residential floor from the ground side, are assumed to be about 2.50 (m²K/W).

A comprehensive version of this model is under development and will support other cities and climate conditions across Iran. You can learn more about the models and access the corresponding Python files in the "More Information" section of this website.