文献类型：期刊文献

英文题名：Influence of Thick Plate Bending Process on Material Strength Distribution in Hydrogenation Reactor Shells

作者：Li, You[1];Chen, Zhiping[1];Jiao, Peng[1];Zhang, Delin[1];Xu, Dong[1];Ma, He[1];Huang, Song[2]

机构：[1]Zhejiang Univ, Inst Proc Equipment, 38 Zheda Rd, Hangzhou 310027, Zhejiang, Peoples R China;[2]East China Univ Sci & Technol, Sch Mech & Power Engn, Shanghai 200237, Peoples R China

年份：2020

卷号：29

期号：8

起止页码：5158

外文期刊名：JOURNAL OF MATERIALS ENGINEERING AND PERFORMANCE

收录：;EI(收录号：20203309057487);WOS:【SCI-EXPANDED(收录号:WOS:000559177000001)】；

基金：This work was supported by National Key Basic Research and Development Project of China (973 Project No. 2015CB057603) and National Natural Science Foundation of China (Grant No. 51905173).

语种：英文

外文关键词：2; 25Cr-1Mo-0; 25 V; artificial neural network; high-temperature strength; hydrogenation reactor; manufacturing residual influence; thick plate bending

摘要：Thick plate bending process (warm bending and tempering) has a profound impact on the material strength distribution (MSD) in hydrogenation reactor shells. To date, few studies have studied the thick plate bending process. In this work, an artificial neural network (ANN) combined with finite element analysis (FEA) was utilized to investigate the impact of thick plate bending on the MSD of reactor shells. First, tensile tests of 0-10% pre-strained 2.25Cr-1Mo-0.25 V specimens were subjected to 390 to 510 degrees C. The results obtained from this experiment were used to develop ANN with two inputs (temperature and plastic strain) to predict the strength of pre-deformed steel. Subsequently, the plastic strain distribution of reactor shells after warm bending was obtained via FEA. We then inputted the FEA results into well-established ANN to predict the MSD of un-tempered reactor shells. The MSD of an actual tempered reactor shell was measured to study the synergic effect of warm bending and tempering on MSD variation. Results showed that the average absolute relative errors between the proposed ANN and tensile test results were below 4%. The absolute relative errors of the proposed prediction method varied from 0.24 to 7.88%. The proposed method is therefore reliable in the lightweight design of the hydrogenation reactor.