冯子强,李登华,丁勇

• 1:南京理工大学理学院
• 2:南京水利科学研究院
• 3:水利部水库大坝安全重点实验室

摘要(Abstract)：

【目的】变形是反映大坝结构性态最直观的效应量，构建科学合理的变形预测模型是保障大坝安全健康运行的重要手段。针对传统大坝变形预测模型预测精度低、误报率高等问题导致的错误报警现象，【方法】选取不同预测模型和聚类算法集成，构建了一种Blending-Clustering集成学习的大坝变形预测模型，该模型以Blending对单一预测模型集成提升预测精度为核心，并通过Clustering聚类优选预测值改善模型稳定性。以新疆某面板堆石坝变形监测数据为实例分析，通过多模型预测性能比较，对所提出模型的预测精度和稳定性进行全面评估。【结果】结果显示：Blending-Clustering模型将预测模型和聚类算法集成，均方根误差(RMSE)和归一化平均百分比误差(nMAPE)明显降低，模型的预测精度得到显著提高；回归相关系数(R~2)得到提升，模型具备更强的拟合能力；在面板堆石坝上22个测点变形数据集上的预测评价指标波动范围更小，模型的泛化性和稳定性得到有效增强。【结论】结果表明：Blending-Clustering集成预测模型对于预测精度、泛化性和稳定性均有明显提升，在实际工程具有一定的应用价值。

关键词(KeyWords)： 大坝;变形;预测模型;Blending集成;Clustering集成;模型融合

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划项目(2022YFC3005502);; 国家自然科学基金项目(51979174);; 中央级公益性科研院所基本科研业务费专项资金项目(Y321004)

作者(Author): 冯子强,李登华,丁勇

DOI: 10.13928/j.cnki.wrahe.2024.04.006

参考文献(References)：

• [1] 顾冲时，苏怀智，王少伟.高混凝土坝长期变形特性计算模型及监控方法研究进展[J].水力发电学报，2016,35(5):1-14.GU Chongshi,SU Huaizhi,WANG Shaowei.Advances in calculation models and monitoring methods for long-term deformation behavior of concrete dams[J].Journal of Hydroelectric Engineering,2016,35(5):1-14.
• [2] 漆一宁，苏怀智，姚可夫，等.耦合时序特征分解筛选的大坝变形分析模型[J].水力发电学报，2023,42(7):56-68.QI Yining,SU Huaizhi,YAO Kefu,et al.Dam deformation analysis model based on coupling time series characteristic decomposition screening[J].Journal of Hydroelectric Engineering,2023,42(7):56-68.
• [3] RANKOVIC V,GRUJOVIC N,DIVAC D,et al.Development of support vector regression identification model for prediction of dam structural behaviour [J].Structural Safety,2014,48:33-39.
• [4] XING Yin,CHEN Yang,HUANG Saipeng,et al.Research on dam deformation prediction model based on optimized SVM [J].Processes,2022,10(9):1842.
• [5] 刘志，刘泽，杨金辉，等.基于BBO-SVM的大坝变形预测模型与性能验证[J].水利水电技术，2020,51(8):62-68.LIU Zhi,LIU Ze,YANG Jinhui,et al.Dam deformation prediction model and performance verification based on BBO-SVM[J].Water Resources and Hydropower Technology,2020,51 (8):62-68.
• [6] 杨杰，吴中如.大坝变形监测的BP网络模型与预报研究[J].西安理工大学学报，2001,17(1):1-5.YANG Jie,WU Zhongru.Dam deformation monitoring model and forecast based on BP Algorithmof Artificial Neural Networks[J].Journal of Xi′AN University of Technology,2001,17(1):1-5.
• [7] 康文豪，徐天奇，王阳光，等.基于CEEMDAN-精细复合多尺度熵和Stacking集成学习的短期风电功率预测[J].水利水电技术(中英文),2022,53(2):163-172.KANG Wenhao,XU Tianqi,WANG Yangguang,et al.Short term wind power prediction based on CEEMDAN fine composite multi-scale entropy and Stacking Ensemble learning[J].Water Resources and Hydropower Technology,2022,53 (2):163-172.
• [8] 康俊锋，胡祚晨，陈优良.基于布谷鸟搜索算法优化LSTM的大坝变形预测[J].排灌机械工程学报，2022,40(9):902-907.KANG Junfeng,HU Zuochen,CHEN Youliang.Dam deformation prediction based on optimization of LSTM by using cuckoo search algorithm[J].Journal of Drainage and Irrigation Machinery Engineering,2022,40(9):902-907.
• [9] LEI W,WANG J,JI T,et al.Dam deformation early warning model based on cluster analysis and spatiotemporal data fusion [J].Measurement,2022,204:112109.
• [10] 邓思源，周兰庭，王飞，等.大坝变形的XGBoost-LSTM变权组合预测模型及应用[J].长江科学院院报，2022,39(10):72-79.DENG Siyuan,ZHOU Lanting,WANG Fei,et al.XGBoost-LSTM combinatorial model with variable weight for dam deformation prediction and its application[J].Journal of Yangtze River Scientific Research Institute,2022,39(10):72-79.
• [11] 张石，郑东健，陈卓研.基于改进PSO-RF算法的大坝变形预测模型[J].水利水电科技进展，2022,42(6):39-44.ZHANG Shi,ZHENG Dongjian,CHEN Zhuoyan.Dam deformation prediction model based on improved PSO-RF algorithm[J].Advances in Water Conservancy And Hydropower Technology,2022,42(6):39-44.
• [12] 刘丹，吕倩，胡少华，等.大坝变形GA-LSTM组合预测模型研究[J].安全与环境学报，2023,23(7):2246-2253.LIU Dan,LYU Qian,HU Shaohua,et al.Research on GA-LSTM combined prediction model of dam deformation[J].Journal of Safety and Environment,2023,23(7):2246-2253.
• [13] 詹明强，陈波，刘庭赫，等.基于变权重组合预测模型的混凝土坝变形预测研究[J].水电能源科学，2022,40(9):115-119.ZHAN Mingqiang,CHEN Bo,LIU Tinghe,et al.Study on deformation prediction of concrete dam based on variable weight combination prediction model[J].Hydropower Energy Science,2022,40(9):115-119.
• [14] 于鹏，顾冲时.大坝安全监测的组合预测模型[J].人民黄河，2006,28(1):67-68.YU Peng,GU Chongshi.Combination forecasting model of dam safety monitoring[J].Yellow River,2006,28(1):67-68.
• [15] 胡毅，瞿博阳，梁静，等.进化集成学习算法综述[J].智能科学与技术学报，2021,3(1):18-33.HU Yi,QU Boyang,LIANG Jing,et al.Evolutionary ensemble learning algorithms[J].Journal of Intelligent Science and Technology,20213(1):18-33.
• [16] 蔡雁，吴敏，王绍丽，等.基于多模型集成的铁矿粉库存量预测方法[J].中南大学学报(自然科学版),2011,42(11):3399-3407.CAI Yan,WU Min,WANG Shaoli,et al.Prediction method of iron ore powder inventory based on multi-model integration[J].Journal of Central South University (Natural Science Edition),2011,42(11):3399-3407.
• [17] 徐耀松，王德兴.基于加权灰色关联投影的Bagging-Blending多模型融合短期电力负荷预测[J].信息与控制，2022,51(2):188-200.XU Yaosong,WANG Dexing.Bagging-Blending multi-model fusion short-term power load forecasting based on weighted gray relational projection[J].Information and Control,2022,51(2):188-200.
• [18] 张晓.基于聚类集成的大坝异常数据检测研究[D].郑州：郑州大学，2021.ZHANG Xiao.Research on dam anomaly data detection based on clustering ensemble[D].Zhengzhou:Zhengzhou University,2021.
• [19] 黄耀英，何一洋，沈振中，等.大坝监测量最佳统计模型优选方法[J].水利学报，2022,53(2):154-164.HUANG Yaoying,HE Yiyang,SHEN Zhenzhong,et al.Optimization method for the best statistical model of dam monitoring quantity[J].Journal of Hydraulic Engineering,2022,53(2):154-164.
• [20] 余龙靖，王冉，刘丰恺.基于Blending多卷积神经网络模型融合的滚动轴承声学故障诊断方法[J].失效分析与预防，2021,16(4):238-245.YU Longjing,WANG Ran,LIU Fengkai.Rolling bearing acousticfault diagnosis method based on Blending multi-convolution alneural network model fusion[J].Failure Analysis And Prevention,2021,16(4):238-245.
• [21] 廖彬，黄静莱，王鑫，等.SCEA:一种适应高维海量数据的并行聚类集成算法[J].电子学报，2021,49(6):1077-1087.LIAO Bin,HUANG Jinglai,WANG Xin,et al.SCEA:A parallel clustering ensemble algorithm for high-dimensional massive data[J].Journal of Electronics,2021,49(6):1077-1087.
• [22] 武新章，王泽宇，代伟，等.基于异质聚类与Stacking的双集成光伏发电功率预测[J].电网技术，2023,47(1):275-284.WU Xinzhang,WANG Zeyu,DAI Wei,et al.Bi-integrated photovoltaic power prediction based on heterogeneous clustering andStacking[J].Power grid technology,2023,47(1):275-284.
• [23] FRED A L N,JAIN A K.Combining multiple clusterings using evidence accumulation [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(6):835-850.
• [24] HSU D.Comparison of integrated clustering methods for accurate and stable prediction of building energy consumption data [J].Applied Energy,2015,160(3):153-163.
• [25] CHEN Siyu,GU Chongshi,LIN Chaoning,et al.Prediction of arch dam deformation via correlated multi-target stacking [J].Applied Mathematical Modelling,2021,91:1175-1193.
• [26] 吴中如.水工建筑物安全监控理论及其应用[M].北京：高等教育出版社，2003.WU Zhongru.Hydraulic building safety monitoring theory and its application[M].Beijing:Higher Education Press,2003.
• [27] 戴波，何启.大坝变形监测统计模型与混沌优化ELM组合模型[J].水利水运工程学报，2016(6):9-15.DAI Bo,HE Qi.Dam Deformation monitoring statistical model and chaotic optimization ELM combined model[J].Hydro-Science and Engineering,2016(6):9-15.
• [28] 张梦迪，徐庆，刘振鸿，等.基于动态滑动窗口BP神经网络的水质时间序列预测[J].环境工程技术学报，2022,12(3):809-815.ZHANG Mengdi,XU Qing,LIU Zhenhong,et al.Water quality time series prediction based on dynamic sliding window BP neural network[J].Journal of Environmental Engineering Technology,202212(3):809-815.
• [29] 勾玄，肖先勇，基于经验模式分解与LSTM神经网络的短期电价预测模型[J].西安理工大学学报，2020,36(1):129-134.GOU Xuan,XIAO Xianyong.Short-term electricity price forecasting model based on empirical mode decomposition and LSTM neural network[J].Journal of XiAN University of Technology,2020,36(1):129-134.
• [30] 贾俊平，何晓群，金勇进.统计学[M].7版.北京：中国人民大学出版社，2018.JIA Junping,HE Xiaoqun,JIN Yongjin.Statistics[M].7 Edition.Beijing:Renmin University of China Press,2018.
• [31] 孙韩玉，黄丽霞，张雪英，等.基于双通道卷积门控循环网络的语音情感识别[J].计算机工程与应用，2023,59(2):170-177.SUN Hanyu,HUANG Lixia,ZHANG Xueying,et al.Speech emotion recognition based on dual-channel convolutional gated recurrent network[J].Computer engineering and application,2023,59(2):170-177.