Prediction of wax deposition rate of crude oil using statistical and neural network methods is presented. Wax deposition in the pipeline is an important issue for crude oil transportation and is a complicated process. It depends on the dynamic viscosity of the crude oil, shear stress, temperature gradient, concentration gradient of wax molecules on the wall. By virtue of oil wax deposition data from laboratory experiments, we train the neural network to learn the non-linear relationship between the influential factors and the wax deposition rate. Comparison of prediction results from a linear regression model and the neural network model is carried out to examine the feasibility of using neural network methods to predict wax deposition.

This content is only available via PDF.
You do not currently have access to this content.