2001 AACC Annual Meeting Charlotte, North Carolina October 14-18, 2001 Charlotte Convention Center

Residence time distribution is an important extrusion processing system parameter. Residence time distribution and mean residence time strongly depend on process variables viz., feed rate, screw speed and feed moisture content. They also have been reported to depend on other process variables like, barrel temperature, die temperature and die diameter. A review of the reported literature on residence time in extruders is made here. Modeling the flows of a wide range of complex food materials containing starches, proteins and lipids in the extruder by simulating the residence time distributions, assuming the extruder as a series of continuous stirred tank reactors or plug flow reactors have been reported. Others have developed relationships for the mean residence time as a function of process variables. It is possible to develop better models by using neural networks for the prediction of mean residence time as a function of the process variables in an extrusion. As an example, data from the literature on extrusion of corn meal were used to model mean residence time as a function of process variables by statistical regression and by neural networks. The neural network models performed better than the regression models.