论文部分内容阅读
本文基于人工神经网络,设计了一个长周期储存式压力容器安全分析评价系统。按压力容器自身的特点进行建模,选择三种类型缺陷为模拟对象,利用有限元应力分析进行应力计算,获取各种状态的应力数据作为训练的样本数据,并选用带二次动量项的BP算法对样本数据进行学习,进而建立长周期压力容器安全评价智能系统软件。最后将该评价智能系统软件计算的结果与GB/T 19624—2004《在用含缺陷压力容器安全评定》计算结果进行比对,系统准确性高。利用该软件使压力容器安全评价变得方便、快捷、简单。
Based on artificial neural network, this paper designs a long-period storage pressure vessel safety analysis and evaluation system. According to the characteristics of the pressure vessel, the three types of defects were chosen as the simulation objects. The stress was calculated by finite element stress analysis, and the stress data of various states were obtained as the training sample data. BP with second momentum Algorithm to learn the sample data, and then establish long-period pressure vessel safety evaluation intelligent system software. Finally, the results of the intelligent system software evaluation and the GB / T 19624-2004 “in the safety assessment of flaws with pressure vessels” results for comparison, the system accuracy. The software makes pressure vessel safety evaluation easy, quick and easy.