焊管2024，Vol.47Issue(4)：32-38,7.DOI:10.19291/j.cnki.1001-3938.2024.04.005
基于IPOA-BP算法的焊接接头抗拉强度预测模型
Tensile Strength Prediction Model of Welded Joint Based on IPOA-BP Algorithm
摘要
Abstract
In order to obtain the tensile strength of X80 pipeline steel girth weld joints more quickly and conveniently,a prediction model for the tensile strength of X80 pipeline steel girth weld joints was constructed using the IPOA-BP algorithm.Logistic chaotic mapping,reverse differential evolution,and firefly algorithm were introduced to improve the optimization ability of the POA algorithm.The model takes welding current,welding voltage,welding heat input,shielding gas flow,welding speed,wire feed speed as input parameters,and joint tensile strength as output parameters.The IPOA-BP model,POA-BP model and BP neural network model are compared.The training set is used to train the model,and the test set is used to verify the model.The mean square error,mean absolute percentage error and R² are used to evaluate the model.The final results show that the IPOA-BP algorithm model is more accurate and has a higher degree of fitting.关键词
抗拉强度预测/焊接接头/改进鹈鹕优化算法/BP神经网络Key words
tensile strength prediction/welding joint/improved pelican optimization algorithm/BP neural network分类
矿业与冶金引用本文复制引用
王程,赵桂敏,郑明高,张骁勇..基于IPOA-BP算法的焊接接头抗拉强度预测模型[J].焊管,2024,47(4):32-38,7.基金项目
国家自然科学基金"油气输送管线钢在线碳配分机理、复相结构特征及塑性增长规律的研究"(项目编号51174165) （项目编号51174165）
陕西省自然科学基础研究计划项目"高钢级管线钢摩擦焊焊接接头形成机理研究"(项目编号2018JM5076). （项目编号2018JM5076）
