Study On Wax Deposition Rate Model Of Thermal Insulation Crude Oil Pipeline

With the development of national economy,the use of energy is increasing year by year.As the main energy source,petroleum is not renewable,so how to use it scientifically becomes extremely urgent.In recent years,some pipelines in China have been in the state of low throughput,which aggravates the occurrence of wax deposition.In the process of oil product transportation,the oil temperature will gradually decrease along the pipeline,and the inner wall of the pipeline will produce a wax layer,which will reduce the flow area of the pipeline,resulting in a reduction in the amount of oil transported.In this paper,the problem of wax deposition of Qing-ji mixed oil with Songshan-Huludao insulated crude oil pipe section is carried out,and the waxing point,wax peak and wax content are analyzed by differential scanning calorimetry(DSC).The measurement of the wax characteristic parameters,the precipitation process of the wax crystal was observed by a polarizing microscope,the wax point of the oil was obtained,and the oil was measured by a rheometer to obtain the fluid consistency and flow characteristics of the oil under the non-Newtonian fluid.Index and other parameters.In the case of knowing the on-site pipeline,the indoor loop simulation wax deposition law device was established with reference to previous experience.Due to the complicated working conditions on the site,it is impossible to create exactly the same conditions in the laboratory.Based on the similarity of shear stress,a series of experimental schemes have been developed.A series of data were obtained from the laboratory experiment of wax deposition loop in the laboratory.Using SPSS software,the model of stepwise regression wax deposition rate was obtained,and then the MATLAB software was used toobtain the support vector machine model.Through laboratory experiments,it is concluded that the wax deposition rate increases first and then decreases with the increase of oil temperature and wall shear stress,and increases with the increase of flow and oil wall temperature difference.The variance order and the range analysis method were used to judge the main order of factors affecting the deposition rate of Qingji mixed oil wax: oil temperature> oil wall temperature difference> flow.Using two models to predict the data in the laboratory,the support vector machine model has better stability and accuracy.Then,the situation of the site is predicted.In the three seasons of winter,spring and summer,the data of the scene is brought into the step vector machine model and the support vector machine model to obtain the wax deposition rate.The results show that stepwise regression method can predict the trend of wax deposition better,and the predicted value is larger than that of SVM method,and the predicted value of SVM method is more stable.