Prediction of silver nanoparticles’ diameter in montmorillonite/chitosan bionanocomposites by using artificial neural networks

Artificial neural networks (ANNs) are computational tools that have found comprehensive utilization in solving many complex real world problems. Major benefits in using ANNs are their remarkable information-processing characteristics pertinent mainly to high parallelism, nonlinearity, fault and noise tolerance, and learning and generalization capabilities. An ANN approach is used to model the size of silver nanoparticles (Ag-NPs) in montmorillonite/chitosan bionanocomposites layers as a function of the silver nitrate concentration, reaction of temperature, chitosan percentage, and d-spacing of clay layers. The best ANN model is found and this final model is capable of predicting the size of nanosilver for a wide range of conditions with a mean absolute error of less than 0.004 and a regression error of about 1. Results obtained showed good ability predictive of neural network model for the prediction of the size of Ag-NPs in chemical reduction methods.