Abstract: In the pipeline operation, the phenomenon of wax deposition often occurs, which will lead to the reduction of the volume of transmission and even the occurrence of condensation accident. For Songshan⁃Huludao thermal insulation crude oil pipeline conveying Qingji mixed oil, a self⁃made indoor thermal insulation ring experimental device wasdeveloped, and a stepwise regression model and a support vector machine model of wax deposition rate were established. The results of comparison and verification showed that the stability and accuracy of the support vector machine model were higher than that of the stepwise regression model. The wax deposition rate model was used to predict the wax deposition in different seasons, which provided a theoretical basis for the determination of the optimal paraffin removal period.

Key words: Insulated crude oil pipeline, Stepwise regression, Wax deposition, Support vector machine, Ring experimental device

摘要： 在管道运行中，时常伴随着蜡沉积现象的发生，这种现象会导致输量减少，严重时还会出现凝管事故。针对输送庆吉混油的松山⁃葫芦岛保温原油管段，自制了室内保温环道实验装置，建立了蜡沉积速率的逐步回归模型和支持向量机模型，对比验证结果表明，支持向量机模型的稳定性和准确性高于逐步回归模型。应用建立的蜡沉积速率模型对现场管段进行了不同季节的蜡沉积预测，为管道现场最优清蜡周期的确定提供理论参考。

关键词: 保温原油管道, 逐步回归, 蜡沉积, 支持向量机, 环道实验装置