| 184 | 0 | 267 |
| 下载次数 | 被引频次 | 阅读次数 |
“三合一”自建房虽便于业主生产、储存、经营活动，但因其功能复合，其结构安全性常存在较多严重的问题。为提前预警此类建筑的安全隐患，本研究基于机器学习算法构建了“三合一”自建房结构安全预警模型，并先后应用集成思想和智能优化算法，对预警性能进行了2次提升。首先，采用独热编码、过采样等方法对某地获取的数据进行预处理；其次，选择总体准确率、召回率和AUC值3项指标分别选出最佳的基分类器；之后，使用袋装法和提升法等集成思想优化预警模型的性能；然后，应用鲸鱼、金豺、粒子群等智能优化算法对预警模型性能进一步优化；最后，综合以上模型运算结果，挖掘预警指标中的关键指标。结果表明：(1)经鲸鱼算法优化的Bagging(KNN)模型可更高效地对“三合一”自建房的结构安全进行预警，其召回率为0.802；(2)经鲸鱼算法优化的Boosting(SVM)模型具有更稳定的预警鲁棒性，其AUC值为0.933；(3)默认参数的XGB模型整体预警效率更佳，其总体准确率为0.915；(4)建筑年份、砖混结构、地上层数、建筑面积等14个指标是“三合一”自建房结构安全预警的关键指标。
Abstract：Although a “three-in-one” self-built house is convenient for the owner to produce, store and operate, there are often many serious problems in its structural safety due to its complex functions. In order to warn the potential safety hazard of such buildings in advance, this study constructed a “three-in-one” self-built housing structure safety intelligent early warning model based on machine learning algorithm, and applied the integration idea and intelligent algorithm to improve the early warning performance twice. Firstly, the data obtained in a certain place were preprocessed by means of one-hot coding and oversampling. Secondly, the overall accuracy, recall rate and AUC value were selected to select the best base classifier. Then, the integrated ideas such as bag method and lifting method were used to improve the early warning performance. The intelligent optimization algorithms such as WOA, GJO and PSO were applied to further improve the early warning performance. Finally, based on the above model calculation results, the key indicators in the early warning indicators were mined.The results are as follows:(1)The Bagging(KNN) model optimized by the whale algorithm can more efficiently warn the structural safety of the “three-in-one” self-built houses, with a recall rate of 0. 802；(2)the Boosting(SVM) model optimized by the whale algorithm has more stable early warning robustness, and the AUC value is 0. 933；(3)the XGB model with default parameters has better overall early warning efficiency, and the overall accuracy rate is 0. 915；(4) 14 indicators such as building year,brick-concrete structure,number of upper floors and building area are the key indicators for the early warning of “three-in-one” self-built houses.
[1]北京市安委会.北京“三合一”场所大排查大清理大整治[J].领导决策信息，2017(48):20-21.Beijing Municipal Security Commission. Beijing’s “three in one” place large investigation, large clean-up and large renovation[J]. Information for Decisions Magazine,2017(48):20-21.
[2]娄鹏.浅谈“三合一”场所火灾隐患存在原因及整治对策[J].科技风，2017(13):276.LOU P. Discussion on the causes of fire hazards in “three-inone” places and the countermeasures[J]. Technology Wind,2017(13):276.
[3]张葭伊，吕淑然.“三合一”场所消防现状分析及管理对策探讨[J].安全，2018, 39(11):56-59.ZHANG J Y, LV S R. Analysis on the present situation of fire control in the “three in one” place and discussion on the management countermeasures[J]. Safety, 2018, 39(11):56-59.
[4]何荣.“三合一”自建房火灾试验及灭火技战术研究[J].消防科学与技术，2023, 42(3):351-355.HE R. Study on fire test and fire fighting tactics of three-in-one building[J]. Fire Science and Technology, 2023, 42(3):351-355.
[5]朱杰，陈洋，蔡乐刚.基于既有房屋信息系统的安全管理预警技术研究与应用[J].安全与环境工程，2018, 25(3):166-171.ZHU J, CHEN Y, CAI L G. Research and application of prewarning technology for existing houses safety management based on information system[J]. Safety and Environmental Engineering, 2018, 25(3):166-171.
[6]张璐.群租房火灾预防预警机制失效及其治理研究——基于X区“9·01”事故调查[D].昆明：云南大学，2020.ZHANG L. Study on the Failure and Prevention of Group Fire Rental Fire Prevention and Early Warning Mechanism—Based on “9·01” Accident Investigation in Area X[D]. Kunming:Yunnan University, 2020.
[7]段在鹏，李帆，邱少辉，等.地铁沿线老旧房屋结构安全预警模型[J].中国安全生产科学技术，2022, 18(3):162-167.DUAN Z P, LI F, QIU S H, et al. Early warning model for structural safety of old buildings along metro lines[J]. Journal of Safety Science and Technology, 2022, 18(3):162-167.
[8]何流.基于物联网技术的危险房屋管控研究[D].杭州：浙江大学，2021.HE L. Research on Dangerous House Management and Control Based on Internet of Things Technology[D]. Hangzhou:Zhejiang University, 2021.
[9]翟渊博，王龙，王术春.陕西关中地区自建房安全性能调查分析[J].建筑安全，2023, 38(2):15-18.ZHAI Y B, WANG L, WANG S C. Investigation and analysis on safety performance of self-built houses in Guanzhong Area of Shaanxi Province[J]. Construction Safety, 2023, 38(2):15-18.
[10]陈立，谭光宇，关杰，等.居民自建房空斗墙砌体结构检测与鉴定研究[J].建筑结构，2023, 53(6):131-136.CHEN L, TAN G Y, GUAN J, et al. Research on detection and identification of masonry structure of rowlock cavity wall of residential self-built house[J]. Building Structure, 2023, 53(6):131-136.
[11]RAHMAN M, HASSAN M R, HOSSAIN M T. Seismic performance analysis of mud house in the context of Bangladesh[J]. International Journal of Scientific and Engineering Research, 2013, 4:1612-1617.
[12]BENDAK S, ALHAMMADI A A. A multi-criteria decisionmaking approach to minimising fire risk in detached house designs[J]. Journal of Engineering, Design and Technology,2019, 17(6):1146-1160.
[13]刘艳辉，黄俊宝，肖锐铧，等.基于随机森林的福建省区域滑坡灾害预警模型研究[J].工程地质学报，2022,30(03):944-955.LIU Y H,HUANG J B,XIAO R H,et al. Study on early warning model for regional landslides based on random forest in Fujian Province[J]. Journal of Engineering Geology,2022,30(03):944-955.
[14]MIAO D, LYU Y, YU K, et al. Research on coal mine hidden danger analysis and risk early warning technology based on data mining in China[J]. Process Safety and Environmental Protection, 2023, 171:1-17.
[15]THIRUGNANAM H, RAMESH M V, RANGAN V P. Enhancing the reliability of landslide early warning systems by machine learning[J]. Landslides, 2020, 17(9):2231-2246.
[16]陈铭建.既有民用建筑安全鉴定方法及工程应用[D].南宁：广西大学，2007.CHEN M J. The Existing Civil Buidings’ Appraisement of Safety and Works Application[D]. Nanning:Guangxi University, 2007.
[17]中华人民共和国住房和城乡建设部.危险房屋鉴定标准：JGJ125—2016[S].北京：中国建筑工业出版社，2016.Ministry of Housing and Urban-Rural Development of the People’s Republic of China. Standard for Dangerous Building Appraisal:JGJ 125—2016[S]. Beijing:China Architecture&Building Press, 2016.
[18]LUOR D C. A comparative assessment of data standardization on support vector machine for classification problems[J]. Intelligent Data Analysis, 2015, 19(3):529-546.
[19]梁杰，陈嘉豪，张雪芹，等.基于独热编码和卷积神经网络的异常检测[J].清华大学学报（自然科学版），2019, 59(7):523-529.LIANG J, CHEN J H, ZHANG X Q, et al. One-hot encoding and convolutional neural network based anomaly detection[J].Journal of Tsinghua University(Science and Technology),2019, 59(7):523-529.
[20]CHAWLA N V, BOWYER K W, HALL L O, et al.SMOTE:Synthetic minority over-sampling technique[J]. Journal of Artificial Intelligence Research, 2002, 16:321-357.
[21]杨道龙.房屋结构智能监测系统的设计与实现[D].南京：南京邮电大学，2018.YANG D L. Design and Realization of House Structure Intelligent Monitoring System[D]. Nanjing:Nanjing University of Posts and Telecommunications, 2018.
[22]段在鹏，张灿，谢汉青，等.面向小样本安全生产数据的事故集成预警模型[J].安全与环境学报，2022, 22(1):51-57.DUAN Z P, ZHANG C, XIE H Q, et al. An integrated accident warning model for small sample production safety data[J].Journal of Safety and Environment, 2022, 22(1):51-57.
[23]方匡南，吴见彬，朱建平，等.随机森林方法研究综述[J].统计与信息论坛，2011, 26(3):32-38.FANG K N, WU J B, ZHU J P, et al. A review of technologies on random forests[J]. Statistics&Information Forum,2011, 26(3):32-38.
[24]连克强.基于Boosting的集成树算法研究与分析[D].北京：中国地质大学（北京），2018.LIAN K Q. The Study and Application of Ensemble of Trees Based on Boosting[D]. Beijing:China University of Geosciences, 2018.
[25]李超，汪磊，陈洋，等.基于贝叶斯集成学习算法的土体先期固结压力预测模型[J].地球科学，2023, 48(5):1780-1792.LI C, WANG L, CHEN Y, et al. Prediction model of soils’preconsolidation pressure based on Bayesian ensemble learning algorithm[J]. Earth Science, 2023, 48(5):1780-1792.
[26]CHOPRA N, MOHSIN ANSARI M. Golden jackal optimization:A novel nature-inspired optimizer for engineering applications[J]. Expert Systems with Applications, 2022, 198:116924.
[27]MARINI F,WALCZAK B. Particle swarm optimization(PSO).A tutorial[J]. Chemometrics and Intelligent Laboratory Systems, 2015, 149:153-165.
[28]NASIRI J, KHIYABANI F M. A whale optimization algorithm(WOA)approach for clustering[J]. Cogent Mathematics&Statistics,2018,5(1):1483565.
[29]WILKINSON L. Revising the Pareto chart[J]. The American Statistician, 2006, 60(4):332-334.
基本信息:
DOI：10.13578/j.cnki.issn.1671-1556.20230709
中图分类号:TU746
引用信息:
[1]段在鹏,李炯,郑宏涛等.“三合一”自建房结构安全智能预警模型[J].安全与环境工程,2024,31(05):8-17+27.DOI:10.13578/j.cnki.issn.1671-1556.20230709.
基金信息:
国家社会科学基金项目(17CGL049); 福建省社会科学基金项目(FJ2022B052)