To Assess the Validity of the Transfer Function Method: A Neural Model for the Optimal Choice of Conduction Transfer Functions

This paper presents a new mathematical approachapplied to the Conduction Transfer Functions (CTFs) of amultilayered wall to predict the reliability of building simula-tions based upon them. Such a procedure can be used todevelop a decision support system that identifies the bestcondition to calculate the best CTFs set. This is a critical pointat the core of ASHRAE calculation methodology founded onthe Transfer Function Method (TFM). To evaluate the perfor-mance of different CTFs sets, the authors built a large amountof data, subsequently employed to train a Neural NetworkClassifier (NNC) able to predict the reliability of a simulationwithout performing it. For this purpose all the multilayeredwalls included in the HVAC ASHRAE Handbook were used,and moreover many other walls typical of Mediterraneanbuilding heritage were added. The results show that theproposed method to optimize CTFs based on NNC is highly accurate, fast and easy to integrate in other buildings simulations tools.