Evaluation of micromixing in helically coiled microreactors using artificial intelligence approaches

Abstract

Artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models were employed to evaluate micromixing in micro-helically coiled tubes. For this purpose, the value of segregation index (Xs) in Villermaux/Dushman reaction was obtained in twelve helically microchannels. The Reynolds number (Re), curvature ratio (δ), torsion (ϒ), and the ratio of the volume flow rate of alkaline solution to the acid solution were used as the model input data. The validity of the models was evaluated through one-fourth of the total experimental data, which were not applied in the training procedure. The mean relative error (MRE), mean square error (MSE), and absolute fraction of variance (R2) for ANN model was 0.83%, 1.65 × 10−10, and 0.9994 respectively. The corresponding calculated values for ANFIS were 1.14%, 5.08 × 10−10, and 0.9980. The estimation precision for both models are appropriate and the results indicated that the ANN approach has higher precision than ANFIS.