PREDICTION OF DOUGH RHEOLOGICAL PROPERTIES USING NEURAL NETWORKS

A neural network was designed to predict the rheological properties of dough from the torque developed during mixing. Dough rheological prop erties were determined using traditional equipment such as farinograph and extensigraph. The back-propagation neural network was designed an d trained with the acquired mixer torque curve (input) and the measure d rheological properties (output). The trained neural network accurate ly predicted the rheological properties (>94%) based on the mixer torq ue curve. The ability to measure the rheology of every batch of dough enables online process control by modifying subsequent process conditi ons. This development has significant potential to improve product qua lity and reduce cost by minimizing process variability during dough mi xing.