Indian Journal of Science and Technology

Abstract

Background/Objectives: The non-uniform flow of fluid over parallel channels having a common inlet and outlet is called flowmaldistribution.Theproblemofflowmaldistributionispredominantinheatexchangersandaffects theirperformances by increasing the pressure drop over the channels and has non-uniform mass flow rate. Methods/Statistical Analysis:In the present study a numerical model using commercial code ANSYS FLUENT © of a cross flow heat exchanger is presented and validated with experimental results. The main objective of the study is to optimize the heat exchanger to reduce the mal-distribution of the fluid. A Soft computing technique, Artificial Neural Network (ANN) is used to predict and optimize the size of the heat exchanger. The neural network is trained by results obtained from the numerical simulations for differentRceosnudlittsi:oIntsh. as been found that the flow mal distribution is minimum when the individual channel diameter is minimum. Increasing the channel diameter increases the non-uniformity in the mass flow rate. It has been ascertained from the mean standard deviation based on Neural network prediction that reducing the diameter of the channels 11 and 12 plays a major role in reducing the flow mal distribution inside the heat exchanger. Conclusion/Application: The trained neural network predicts the mass flow rate. Based on these results the heat exchanger is optimized to minimize the flow mal distribution. This Neural network can further be implemented for any design.
Keywords: Artificial Neural Network, Flow Mal-distribution, Heat Exchanger, Mass Flow Rate, Numerical Simulations