CORRELATION TO PREDICT WAX APPEARANCE TEMPERATURE AND CHARACTERISATION OF CRUDE OIL

Abstract

Wax precipitation in oil pipelines and industrial equipment is a severe concern in the petroleum sector as it can cause well bores to block and reduces the efficiency of oil/gas production and transportation processes. The solid particles (wax) will increase the pressure drop in the tubing and eventually cause plugging. Thus, predicting the Wax Appearance Temperature (WAT) is crucial, the temperature at which the first wax droplets are discovered. Gamma Distribution Function (GDF) parameters were used to develop a characterisation to determine the nature of hydrocarbon mixtures. The parameters of GDF were calculated using the characterisation developed in Python. The oil was characterised by using the calculated GDF parameters until C50+. Based on the estimated value of the parameters of GDF, the type of hydrocarbon mixtures were classified into heavy oil/biodegraded, asphaltenic, paraffinic, waxy, and light oils. The WAT of crude oils was predicted using data analysis techniques of regression analysis. Root Mean Square Error (RMSE) and R-squared were used to evaluate the model. Multiple regression was the best data analysis method to correlate WAT, giving the highest R-squared value with a lower RMSE than simple regression. This study will assist in predicting the nature of the crude oil. Correct prediction of crude oil’s nature will help manage flow assurance.

Keywords

• gamma distribution function
• data analysis
• flow assurance
• wax precipitation
• characterisation