机械科学与技术2025，Vol.44Issue(9)：1491-1498,8.DOI:10.13433/j.cnki.1003-8728.20230332
热浸镀锌钢管涂层厚度预测与工艺参数优化
Prediction of Coating Thickness and Optimization of Processing Parameters for Hot-dip Galvanization Steel Pipes
摘要
Abstract
The galvanized steel pipe is used as the research objective,and the thickness parameters of 167 steel pipes were used as the modeling data and validation data.The sensitivity evaluation factors included the lead-up speed,lead-out speed,draw time,discharge time,press-down time,zinc liquid temperature,zinc dipping time,external blowing pressure and wind ring position.A prediction model for galvanized coating thickness of steel pipes based on the SVM model was established,and a sensitivity analysis of the calculation model was carried out by using the Sobol method to determine the influence of the processing parameters on the galvanized coating thickness.Four swarm intelligence optimization algorithms were used to optimize the SVM model,the prediction accuracy was analyzed,and the optimal combination of zinc coating thickness parameters was given by searching the optimal model.The results show that wind ring position,zinc dipping time,zinc liquid temperature,press down time and extraction speed are important parameters that affect the prediction results of the model.The SVM model by using the Golden jackal optimization(GJO)has a fast convergence speed and good prediction ability,which is the best model among four models.The model of GJO-SVM is optimized by using the genetic algorithm to obtain the optimal processing parameters.For the practical production,under the requirement of the standard,zinc material can be saved and efficiency can be improved.关键词
涂层厚度/支持向量机/金豹优化算法/全局敏感性分析/工艺参数Key words
coating thickness/support vector machine/Golden jackal optimization/global sensitivity analysis/process parameters分类
矿业与冶金引用本文复制引用
张俊红,孟峻巍,李相东,戴胡伟,张学玲,于洋洋..热浸镀锌钢管涂层厚度预测与工艺参数优化[J].机械科学与技术,2025,44(9):1491-1498,8.基金项目
天津市科技重大专项与工程项目(21ZXCCSN00020) （21ZXCCSN00020）
