A COMPARISON OF THE ENERGY AND EXERGY PERFORMANCE OF R1234YF AND R134a IN A COMPRESSION STAGE USING COMPUTATIONAL INTELLIGENCE TECHNIQUES

This paper presents a scheme for the modeling the energy and exergy performance of a reciprocating compressor operating with R1234yf and R134a fluids; the compression process model is developed using the Artificial Neural Network (ANN), which is based on artificial intelligence techniques that act as a black box model. The model was created only from experimental data and provided evidence that it can be extended to systems working with R1234yf as long as data is available. The selected network has three hidden layers, this becomes a special configuration never used before in this field. The input variables are: suction pressure, suction temperature, discharge pressure, and compressor rotation speed and molecular weight. The output parameters are: energy consumption, exergy destruction and exergy efficiency. The models are experimentally validated, and then, they are used in a computational simulation in order to stablish a comparative approach on the energy and exergy performance between these both refrigerants.