1、Comprehensive evaluation of constructionsafetyintherehabilitationphaseofancient buildings based on NCMFPNZHAO Ping1,2,LIU Guang-chuan1,2,ZHOU Ting-ting3,CAO Jin-feng1,2,AO Qiang1,2(1 School of Civil Engineering,Xi an University of ArchitectureandTechnology,Xi an710055,China;2NationalExperimental Tea
2、ching Center for Civil Engineering VirtualSimulation,Xi an University of Architecture and Technology,Xian 710055,China;3 School of Management,XianUniversity of Architecture and Technology,Xian 710055,China)Abstract:The purpose of this article is to prevent constructionsafety accidents in the repair
3、stage of ancient buildings,and tofind out the key factors that affect their construction safetymanagementForthisreason,thispaperproposesacomprehensive evaluation method for construction safety based onthe Normal Cloud Model(NCM)and Fuzzy Petri Net(FPN)Firstly,the article analyzes the construction ch
4、aracteristics andrisk characteristics of the ancient building in the repair phase,establishes a tripod accident cause model(Tripod-Delta)undermulti-factorcoupling,andobtainsthecouplinginfluencemechanism of personnel,materials and equipment,environment,technology,and management factors on constructio
5、n safetyFrom this,an indicator system is constructed from five aspects:personnel,materials and equipment,environment,technology,and management Then,the above index system is convertedinto a construction safety multi-factor coupled FPN networkstructure NCM is used to determine the initial state of th
6、e FPNindex,the FPN redundant index nodes are reduced through theinverse search strategy,and the fuzzy inference algorithm is usedto obtain the final evaluation matrix In the end,a diagnosismodel of obstacle factors is constructed to diagnose the evaluationprocess and evaluation results Finally,the t
7、eam carried out acase study on the construction safety assessment during the repairphase of the Xiyue Temple in Huayin The results show that theevaluation result of the example is almost safe,which is almostconsistent with the on-site situation Collaborative managementand materials and equipment are
8、 the key factors that affect theconstruction safety of the ancient building in the repair stageTheNCM FPNcomprehensiveevaluationmethodofconstruction safety in the restoration stage of ancient buildingscan fully express the coupling characteristics of the constructionsafety risks in the restoration s
9、tage of ancient buildings,anddetermine the key factors of safety management The evaluationresults are objective and accurateIt can be used for thecomprehensive evaluation of construction safety in the restorationstage of ancient buildingsKey words:safetyengineering;ancientbuildingrepair;construction
10、 safety;Tripod-Delta;Normal CloudModel(NCM);Fuzzy Petri Net(FPN)文章编号:1009-6094(2023)04-1031-07海洋大气环境下架空管道腐蚀失重预测*骆正山,杨赛星,王小完(西安建筑科技大学管理学院,西安 710055)摘要:为提高海洋大气环境下油气管道腐蚀失重预测精度,提出了一种基于 IWOA DGM(1,1,)的架空管道腐蚀失重预测模型。首先用佳点集理论优化鲸鱼优化算法(Whale Optimization Algorithm,WOA)初始种群的布局以增强寻优能力,并用非线性自适应权重和改进黄金正弦算法避免WOA 模
11、型的随机性,以提高收敛速度。然后利用指数变换、变权弱化缓冲算子和新陈代谢改进离散灰色模型(DiscreteGrey Model,DGM(1,1),指数变换可以提高原始数据的光滑度,变权弱化缓冲算子可以抵消冲击扰动的干扰,加入新陈代谢可以实现动态的中长期预测。再用改进鲸鱼优化算法(Improve Whale Optimization Algorithm,IWOA)对 改 进DGM(1,1,)模型中的参数 进行寻优,最后建立改进鲸鱼优化算法的优化离散灰色预测模型(IWOA DGM(1,1,),并用该模型对西沙 Q235 碳钢管道腐蚀失重情况进行预测。结果表明,改进模型的灰色绝对关联度、平均绝对百分
12、比误差和拟合度分别为 98.99%、2.43%和 98.73%,各项评价指标均优于其他对比模型,且预测精度得到了极大的提高,证明该模型有较好的鲁棒性。关键词:安全工程;海洋大气环境;架空管道;腐蚀失重;改进鲸鱼优化算法(IWOA);DGM(1,1)模型中图分类号:X937文献标志码:ADOI:10.13637/j issn 1009-6094.2021.2312*收稿日期:2021 12 24作者简介:骆正山,教授,博士,从事油气管道风险评估、建模与方法、信息管理与信息系统等研究,。基金项目:国家自然科学基金项目(41877527);陕西省社科基金项目(2018S34)0引言海洋管道是海洋油气
13、运输的生命线,但长期处于海洋大气环境中的管道腐蚀非常严重,甚至造成管道腐蚀破裂泄漏风险 1,腐蚀破坏是引发管道恶性事故的主要原因 2 4。因此,随着海洋大气环境下管道设施建设的推进,如何精准地研究海洋大气环境下油气管道腐蚀失重的变化趋势,准确预测管道的使用寿命尤为重要,对管道的安全使用有重大的保障作用。灰色模型在金属及管道腐蚀的预测等小数据预1301第 23 卷第 4 期2023 年 4 月安全 与 环 境 学 报Journal of Safety and EnvironmentVol 23No 4Apr,2023测方面应用非常广泛。1997 年,李长荣等5 首次将传统灰色模型法(Grey M
14、odel,GM(1,1)用于管道腐蚀泄露预测,得到较好的预测精度。Zhu 等6 建立了非流体 GM(1,1)模型,预测铜在被污染大气中的腐蚀速率。俞树荣等7、张新生等8 基于灰色模型预测油气管道的腐蚀寿命,但模型的预测精度受到随机确定背景值的影响。秦谢勋等9 运用改进的蜂群算法对 GM(1,1)模型优化背景权值,建立了改进蜂群算法优化的指数变换灰色模型,预测管道寿命,但该模型的长期预测效果较差。李昊燃等10 加入构建背景值等多方面的优化灰色模型,对管道腐蚀速率进行预测,再用马尔科夫模型修正预测结果,解决了传统模型中长期动态预测精度较差的问题。Zhi 等11 建立了正则化遗传算法的非线性灰色贝努
15、利模型,实现了碳钢腐蚀的长期预测,但模型预测精度很大程度上依赖于其参数的选择。骆正山等12 建立了灰色径向基函数腐蚀剩余寿命预测模型,但GM(1,1)模型预测的精度依赖于建模数据的选取。姜峰等 13、张新生等 14 基于非等间距 GM(1,1)模型预测管道腐蚀速率,扩大了 GM(1,1)模型的有效预测范围。上述研究大都集中于埋地管道及海底管道的研究,对架空管道研究甚少,且在众多改进方法中,较少涉及动态预测模型的改进方法。由于 GM(1,1)模型有时误差很大,无论怎么修正初始值、优化时间和相应系数等都无法得到满意的拟合精度。鉴于此,本文提出一种基于改进鲸鱼优化算法(Improve Whale O
16、ptimization Algorithm,IWOA)的优化离 散 灰 色 预 测 模 型(DiscreteGreyModel,DGM(1,1,)的架空管道腐蚀失重预测模型。用IWOA 模型优化 DGM(1,1,)模型中的参数,构建 IWOA DGM(1,1,)管道腐蚀失重预测模型。以西沙某架空管道的腐蚀失重数据为实例,开展架空管道腐蚀失重预测研究,并与原有模型进行统计学指标对比和验证,以期提高架空管道腐蚀失重预测精度,为架空管道的安全防护提供技术支撑和决策依据。1相关原理1.1鲸鱼优化算法(WOA)鲸鱼优化算法(Whale Optimization Algorithm,WOA)是启发式优化新算法15,算法有包围捕食、螺旋更新和搜寻猎物 3 个阶段16。该算法采用参数向量 A 和随机概率 P 来确定鲸鱼更新的位置,P 0,1,当 P0.5 时,鲸鱼以螺旋式接近猎物,当P 0.5 且|A|1 时,鲸鱼无法获取到猎物的有效信息,此时执行全局搜索,当 P 0.5 且|A|1 时,鲸鱼局部包围猎物。1.2DGM(1,1)预测模型设 非 负 原 始 数 据 序 列 X(0)=x(0)(1),x(