摘要
高压氢能储运设备的质量分级对保障氢能产业链安全至关重要。针对现有评价体系不完善的问题,构建了基于ANP-TOPSIS和BP神经网络的分级预警模型。首先,依据GBT 33145-2016标准选取抗拉强度、循环次数等10项关键指标,通过ANP确定指标权重;其次,采用TOPSIS方法计算设备贴近度并划分质量等级(A/B/C三级);最后,基于分级数据训练BP神经网络实现智能预警。实验表明:该模型在仿真数据实验中整体准确率达到94%,在实测数据对比实验中,整体准确率达到了97%,对高风险等级(C级)设备的召回率分别高达97%和96%。研究成果为高压储氢设备的质量管控提供了理论依据和实用工具。
The quality classification of high-pressure hydrogen energy storage and transportation equipment is very important to ensure the safety of hydrogen energy industry chain.Aiming at the problem that the existing evaluation system is not perfect,this paper constructs a hierarchical early warning model based on anp-topsis and BP neural network.Firstly,according to GBT 33145-2016 standard,10 key indicators such as tensile strength and number of cycles are selected,and the index weight is determined through ANP;Secondly,the TOPSIS method is used to calculate the equipment proximity and classify the quality level(a/b/c);Finally,BP neural network is trained based on hierarchical data to realize intelligent early warning.The experimental results show that the overall accuracy of the model is 94%in the simulation data experiment,97%in the measured data comparison experiment,and the recall rate of high-risk level(Level C)equipment is 97%and 96%respectively.The research results provide a theoretical basis and practical tools for the quality control of high-pressure hydrogen storage equipment.
作者
左从瑞
李光明
王璇涛
朱宪宇
易意业
ZUO Congrui;LI Guangming;WANG Xuantao;ZHU Xianyu;YI Yiye(Hunan Institute of Metrology and Test,Changsha 410014,China;Department of Intelligent Manufacturing Engineering,Changsha Vocational and Technical College,Changsha 410217,China;Department of Mechanical and Energy Engineering,Shaoyang University,Shaoyang 422004,China)
出处
《计量科学与技术》
2026年第1期36-45,共10页
Metrology Science and Technology
基金
湖南省自然科学基金项目“高压氢能储运设备质量分级评价方法研究”(2024JJ8306)
湖南省自然科学基金项目“基于D-S证据理论的多源数据联合电力设备故障诊断算法研究”(2024JJ7492)
湖南省教育厅科学研究项目“基于信息融合的电力设备故障诊断技术研究”(24C1448)。
关键词
计量学
高压氢能储运设备
质量分级
ANP-TOPSIS模型
BP神经网络
metrology
high pressure hydrogen energy storage and transportation equipment
quality classification
anp-topsis model
BP neural network
