A neuro-fuzzy modeling tool (ANFIS) has been used to dynamically model cross flow ultrafiltration of milk. It aims to predict permeate flux and total hydraulic resistance as a function of transmembrane pressure, PH, temperature, fat, molecular weight cut off, and processing time. Dynamic modeling of ultrafiltration performance of colloidal systems (such as milk) is very important for designing of a new process and better understanding of the present process. Such processes show complex non-linear behavior due to unknown interactions between compounds of a colloidal system. In this paper, ANFIS, Multilayer perceptron (MLP), and FIS were applied to compare results. The ANFIS approximation gave advantages over the other methods. The results showed that there is an excellent agreement between the checked data (not used in training) and modeled data, with average errors very low. Also the trained ANFIS is able to accurately capture the non-linear dynamics of milk ultrafiltration even for a new condition that has not been used in the training process (tested data). As well as ANFIS is compared to Multilayer perceptron (MLP). As well as the MATLAB codes and instructions used to implement the method.

Keywords