Measurement of palm oil viscosity after heat treatment has been done. The sample is palm oil packaging and bulk ones Falling body method is used as an instrument to determine the coefficient of viscosity which had previously been calibrated using water and REFPROP. The measurement results showed that the coefficient of viscosity of palm cooking oil that has been used 10 times for 30 minuted will rise by 76% for oil packs A and about 45% for a sample of bulk, respectively. For palm oil used once but its has been using for one hour, the viscosity will be comparable to 6 times the use of the cooking oil used for 30 minutes. Thus based on the masurement of viscosity value, the palm oil will have a viscosity values are relatively small when used often, but in a short time than is used only once, but in a relatively longer.

A novel approach is presented for efficiently training a neural network (NN)-based surrogate model when the training data set is to be generated using a computationally intensive high-fidelity computational model. The approach consists in using a Gaussian Process (GP), and more specifically, its acquisition function, to adaptively sample the parameter space of interest and generate the minimum amount of training data needed to achieve the desired level of approximation accuracy. The overall approach is explained and illustrated with numerical experiments associated with the prediction of the lift-over-drag ratio for a NACA airfoil in a large, two-dimensional parameter space of free-stream Mach number and free-stream angle of attack. The obtained numerical results demonstrate the superior accuracy delivered by the proposed training over standard trainings using uniform and random samplings.