基于数据驱动的文物环境风险状态评估方法研究  

Study on the environmental risk state assessment method based on the data-driven mode

作　　者：王琨[1] 梁茹 侍洪波 谭帅 蔡兰坤[2] 吴来明[3,4] 徐方圆[3,4] WANG Kun;LIANG Ru;SHI Hongbo;TAN Shuai;CAI Lankun;WU Laiming;XU Fangyuan(Key Laboratory of Smart Manufacturing in Energy Chemical Process(East China University of Science and Technology),Ministry of Education,Shanghai 200237,China;National Key Laboratory of Environmental Risk Assessment and Control for Environmental Protection Chemical Processes(East China University of Science and Technology),Shanghai 200237,China;Key Scientific Research Base of Museum Environment(Shanghai Museum),National Cultural Heritage Administration,Shanghai 200231,China;Shanghai Museum,Shanghai 200231,China)

机构地区：[1]能源化工过程智能制造教育部重点实验室(华东理工大学),上海200237 [2]国家环境保护化工过程环境风险评价与控制重点实验室(华东理工大学),上海200237 [3]馆藏文物保存环境国家文物局重点科研基地(上海博物馆),上海200231 [4]上海博物馆,上海200231

出　　处：《文物保护与考古科学》2024年第5期143-150,共8页Sciences of Conservation and Archaeology

基　　金：国家重点研发计划(2020YFC1522502,2020YFC1522505)资助。

摘　　要：在博物馆文物保护中,对环境风险监测和风险变量的及时定位有重要意义。然而,由于各个文物展馆的差异性,很难将某一展馆的环境风险划分标准应用于其他展馆,甚至无法应用到同一展馆的不同类型文物。除此之外,当风险出现时,博物馆无法及时锁定影响因素,延长了风险排除所需的时间。因此,结合上海博物馆江南文化艺术临展的连续监测数据,提出了一种针对纸质文物的环境风险状态评估二级监测框架,并按需求划分为“优、良、中、差”四个等级进行风险监测。首先,使用“优”数据集建立主成分分析(principal component analysis,PCA)模型并构建控制限用以挑选“优”数据,“差”数据集通过已有的PCA模型得到判断“差”等级的控制限;然后,使用“非优非差”数据集建立ReliefF-PCA模型并得到“良”“中”等级划分的控制限;最后,对应PCA模型和ReliefF-PCA模型分别使用贡献图和基于ReliefF算法计算权值的加权贡献图,以此量化风险,初步定位对环境指标存在影响的风险变量。通过两个评价模型构建的监测框架,为文物保存环境质量的评估和后续馆藏环境调控提供了一种更为科学的方案,有助于提高整个文物保护系统的效率。For the conservation of museum objects on display,it is important to monitor environmental risks and locate risk variables in a timely way.However,due to the differences of exhibition halls for various cultural relics,it is difficult to apply the environmental risk classification criteria for a particular exhibition hall to other exhibitions,or even to different types of cultural relics in the same exhibition hall.In addition,when a risk appears,a museum may not find the influencing factors in time,which prolongs the length of time of the risk.Therefore,in combination with the continuous monitoring data of the temporary exhibition“Jiangnan Culture and Art”in Shanghai Museum,a two-level monitoring framework for environmental risk state assessment of paper cultural relics has been proposed.The risk monitoring is carried out by four levels,namely“excellent,good,medium and poor”according to the demands.Firstly,a principal component analysis(PCA)model is established using the“excellent”dataset and the control limit is constructed to select“excellent”data.The control limit of“poor”data is obtained using the existing PCA model.Then,a ReliefF-PCA model is established using the“non-excellent and non-poor”dataset and the control limit of“good”and“medium”classification is obtained.Finally,a contribution plot and a weighted contribution plot based on the ReliefF algorithm are used for the corresponding PCA model and ReliefF-PCA model respectively,thus quantifying the risk and initially locating the risk variables that have an impact on the environmental indicators.The monitoring framework constructed by the two evaluation models provides a more scientific scheme for the evaluation of the environmental quality of cultural relic preservation and the subsequent adjustment of the museum environment.This is helpful for improving the efficiency of the whole cultural relic conservation system.

关 键 词：环境监测 风险评估 文物保护 PCA算法 贡献图 变量定位 

分 类 号：K85[历史地理—考古学及博物馆学] G264[历史地理—历史学] TP274[文化科学]