Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

In order to solve a variety of problems that restrict the controlling and prediction of moisture during feed extrusion processing, the research about moisture on-line prediction system of extruded feed is put forward. After analyzing of the extrusion process of feed, finding that extruding is the key process to moisture control. Then moisture on-line prediction model of extruder based on BP neural networks was established. The input of the network are ingredient feeding speed, steam feeding speed, water feeding speed, screw speed, and chamber temperature, while taking the moisture content of the pellet as output. The linear equation between prediction output and true value is above 0.95 which indicates that prediction model is available. Moisture on-line prediction system of extruded feed has definite practical value. It can be an operation guidance of the moisture control during extrusion process, and provide a new way of modeling to feed processing industry.