THE ESTIMATION OF KINEMATIC VISCOSITY OF PETROLEUM CRUDE OILS AND FRACTIONS WITH A NEURAL-NET

This paper illustrates how a neural net, a three-layered Perceptron, c an be trained to estimate viscosities for undefined crude oils and fra ctions. Three Saudi-Arabian crude oils were employed to illustrate the use of the neural net to approximate the relation in a very simple ma nner with no need for a priori knowledge of the system. This empirical correlation was accurate to 98.74% if tested on experimental data not used during training, which is a fivefold improvement on average resu lts obtained by two recently-proposed equations to estimate the viscos ity of hydrocarbons. Although the neural net equation seems to be less transparent than former correlations, a method called backward analys is is proposed to analyze the weight matrix of the neural net in order to gain valuable insight into the viscosity system.