Thixotropy models of waxy crude oil by using non-linear implicit iterative regression method

For the existed problems occurred in parameter regression of thixotropy models for waxy crude oil including error enlargement of experiment data and the difficulty to choose an initial iterative point, the paper, by means of non-linear parameter regression as its frame, introduces the idea of resolving ill-posed problems, with the application of non-linear implicit iterative method which is widely used for ill-posed problem, and obtains the related parameters of Houska thixotropy model at one time. The parameter regression is employed in the case of Zhongyuan Crude with chemical agents (the concentration ratio is 50 mg/kg)added. The obtained numerical results have a good coincidence with the experiment data. The minimum relative error is 2.71%(26 ℃)while the maximum relative error is only 5.09%(21 ℃). The method is both accurate and numerically stable, and the error will be decreased as the temperature increases. So the presented method can be used for parameter regression of thixotropy models for waxy crude oil.