Rheology of ethylene- and propylene-glycol ice slurries: Experiments and ANN model.

An experimental investigation combined with a numerical study is performed to characterize the rheological behavior of ice slurries. Two additives, namely ethylene glycol and propylene glycol, are considered at three initial concentrations Xa?=5, 14 and 24%. The ice fraction is varied from 5 to 65%. Flow ramp tests are carried out using a hybrid HR-2 rheometer. The Herschel–Bulkley model is then employed to predict the rheological behavior of ice. Using a least-square approach, the flow index n and the consistency index k are deduced from the rheograms. The ice slurries exhibit either a shear-thinning or a shear-thickening behavior depending on the operating conditions. An experimental database is produced based on the present experiments and on experimental data retrieved from the literature. An Artificial Neural Network (ANN) model is then developed and validated using this database and appears to be a valuable tool for predicting the rheological behavior of ice slurries.