压力容器2025，Vol.42Issue(1)：78-84,7.DOI:10.3969/j.issn.1001-4837.2025.01.010
基于静/动态因素协同的承压设备损伤智能识别技术
Intelligent damage recognition technology for pressure equipment based on the collaboration of static and dynamic factors
摘要
Abstract
Based on the standards for damage mode recognition technology for pressure equipment and risk-based inspection(RBI)cases,an intelligent damage recognition model that integrates standards and expert experience was developed.The model is composed of two sub-models,one is for static factors and the other is for dynamic factors.The combination of these two models was used for the recognition of damage modes of pressure equipment.The model was trained with 9 522 RBI cases and validated with 2 459 cases.The results show that the recognition rates for 19 out of 29 damage modes are above 0.75,the recognition rates for 14 damage modes exceed 0.85,and those for 12 damage modes reach over 0.93.Meanwhile,the causes for the low recognition rates for some damage modes were analyzed and measures for further enhancing the model's recognition rates were proposed.Furthermore,the model was utilized to analyze damage modes labeled as"unknown corrosion"in 7 063 RBI cases,aiming to ascertain possible damage modes for these instances of unknown corrosion.This technology has been successfully applied and validated on a specific atmospheric-vacuum distillation unit,yielding favorable application outcomes.关键词
承压设备/损伤模式/智能识别/机理分析/工况数据Key words
pressure equipment/damage model/intelligent recognition/mechanism analysis/working condition data分类
机械制造引用本文复制引用
朱建新,徐越遥,高振宇,乔松,胡久韶..基于静/动态因素协同的承压设备损伤智能识别技术[J].压力容器,2025,42(1):78-84,7.基金项目
国家重点研发计划项目(2022YFB3306400) （2022YFB3306400）
合肥通用机械研究院有限公司青年基金项目(2024011175) （2024011175）
