Herein, we show the prediction of the viscosity of a binary mixture of bitumen and light oil using a feedforward neural network with backpropagation model, as compared to empirical models such as the reworked van der Wijk model (RVDM), modified van der Wijk model (MVDM), and Al-Besharah. The accuracy of the ANN was based on all of the samples, while that of the empirical models was analyzed based on experimental results obtained from rheological studies of three binary mixtures of light oil (API 32°) and bitumen (API 7.39°). The classical Mehrotra–Svrcek model to predict the viscosity of bitumen under temperature and pressure, which estimated bitumen results with an %AAD of 3.86, was used along with either the RVDM or the MVDM to estimate the viscosity of the bitumen and light oil under reservoir temperature and pressure conditions. When both the experimental and literature data were used for comparison to an artificial neural network (ANN) model, the MVDM, RVDM and Al-Besharah had higher R² values.

In this study, we introduce a new method for the prediction of the viscosity of bitumen diluted with light oil under reservoir temperature and pressure. This two-step method works as follows: first, predicting the bitumen viscosity under reservoir temperature and pressure using the classical Mehrotra and Svrcek model, and then subsequently using it in the modified Van Der Wijk (MVDM) model. This model formed from the modification of the original Van Der Wijk model was developed from the consideration of the interactions between like molecules in different binary components of the mixture. In this study, the bitumen viscosity was predicted with an average absolute deviation percentage (AAD%) of 3.86. The accuracy of the MVDM was investigated from the experimental results obtained from the rheological studies of three binary mixtures of light oil (API 32°) and bitumen (API 7.39°). Dead oils were mixed on a mass fraction basis. The viscosity was measured at a temperature range of 45–110 °C and a pressure range of 0.1–6 MPa. For comparison purposes, a reworked Van Der Wijk model (RVDM) was used in the same method and compared to the MVDM. The latter was more accurate than the RVDM with AAD% values of 8.88, 8.02, and 5.07 in predicting the viscosity of the three mixtures of 25, 32.5, and 50% bitumen with light oil. On the other hand, the RVDM had AAD% values of 12.42, 11.43, and 7.87 for the same mixtures, respectively. The applicability of this method was further verified by comparing its accuracy to another reported method using published data and it was found that the MVDM had AAD% values of 1.86, 6.55, and 2.823 when predicting the viscosities of the three mixtures under reservoir temperature and pressure conditions.