Prediction of Friction Factor in Serpentine Micro-Pipes Using Artificial Neural Network

Abstract

Serpentine Micro-fluidic devices have become much relevant especially in biochemical industries and laboratories. Prediction of flow characteristics considering various geometrical parameter of the pipe in corporation has been a serious challenge. This paper present application of Artificial Neural Network (ANN) modeling for prediction of friction factor (fr) which is key among the flow characteristics. Experimental study and ANN modeling were performed to estimate and predict friction factor in serpentine pipes. In this work, nine serpentine micro-pipes with different geometric parameters (diameters (D), straight length (Ls) and curve length (Lc) of the pipe, pipe material (m) and different number of turn (n)) were fabricated and used in the experiment. The different geometric parameters were grouped into number of turns, straight length, curve length, diameter of the pipe and pipe materials with each of the geometrical parameters varied autonomously at a time while the rest of the parameters were kept constant for the experimental determination of the friction factor (fr). The numerical data were applied as input and target set of the ANN model with the data points divided into 40% for training, 30% for validation and 30% for testing and the friction factor (fr) was predicted.