普适性结蜡模型相关参数的优化方法

Optimization on related parameters of unified wax deposition model

为制定经济、合理的清管方案,需建立蜡沉积模型来预测蜡沉积速率和厚度。基于目前工程常用的普适性结蜡模型,借助蜡沉积实验数据计算蜡沉积各因素之间的灰色相对关联度,通过分析蜡沉积各影响因素之间的多重共线性,证明普适性结蜡模型的变量选用合理;并基于1stOpt软件对普适性结蜡模型相关参数进行优化研究。结果表明:基于Levenberg-Marquardt算法、单纯形法、模拟退火算法、遗传算法和微粒群算法建立的普适性结蜡模型几乎一致且精度高于线性变换法。

In order to formulate an economical and reasonable pigging plan,a wax deposition model needs to be established to predict the deposition rate and thickness.Based on the unified wax deposition model commonly used in engineering at present,with the aid of wax deposition experimental data,the gray relative degree of wax deposition between the various factors of wax deposition is calculated with the aid of wax deposition experimental data,and the multi-collinearity between the various influencing factors of wax deposition is verified that the independent variables of the unified wax deposition model are reasonably selected;and the relevant parameters of unified wax deposition model is optimized based on the 1 stOpt software.The results show that the unified wax deposition model is established based on the Levenberg-Marquardt algorithm,the simplex method,the simulated annealing algorithm,the genetic algorithm and the particle swarm algorithm.The adaptive wax deposition model is almost the same and the accuracy is higher than that of the linear transformation method.