基于不同模型的煤气管道剩余强度评价方法对比  

Comparison of residual strength evaluation methods of gas pipeline based on different models

作　　者：王战辉[1] 张智芳[1] 李红伟 闫君芝[1] WANG Zhanhui;ZHANG Zhifang;LI Hongwei;YAN Junzhi(School of Chemistry and Chemical Engineering,Yulin University,Yulin719000,China;Key Laboratory of Low Metamorphic Coal Clean Utilization,Yulin719000,China)

机构地区：[1]榆林学院化学与化工学院,陕西榆林719000 [2]浙江闰土股份有限公司,浙江绍兴312300

出　　处：《化工科技》2021年第5期1-5,67,共6页Science & Technology in Chemical Industry

基　　金：国家自然科学基金项目(21763030);陕西省创新能力支撑计划项目(2020TD-032);陕西省教育厅重点科学研究计划(重点实验室)项目(19JS069);榆林市2019年科技计划项目(2019-83-7,2019-102-1);榆林高新区科技计划项目(CXY-2020-02)。

摘　　要：以不同钢级煤气管道为研究对象,借助Matlab软件,通过建立BP、GA-BP和带动量项BP人工神经网络数学模型,对其进行剩余强度预测,并与真实剩余强度进行对比分析;通过引入相对误差和压力比等参数,考察其准确性和适用性。结果表明,从运行时间和最大均方误差角度出发,GA-BP模型是最优的方法;从相对误差和压力比出发,GA-BP模型准确度最高,适用性最好。所得结论对煤气管道的正常运行有一定的指导意义。With the help of Matlab software, the residual strength of gas pipeline with different steel grades is predicted by establishing BP,GA-BP and driving quantity TERM-BP artificial neural network mathematical model, and compared with the real residual strength.The relative error and pressure ratio are introduced to test the accuracy and applicability.The results show that GA-BP model is the best method in terms of running time and maximum mean square error.From the relative error and pressure ratio, GA-BP model has the highest accuracy and the best applicability.The conclusion has a certain guiding significance for the normal operation of gas pipeline.

关 键 词：煤气管道 剩余强度 神经网络 数学模型 

分 类 号：TQ028.8[化学工程]