Deformation forecasting model and its modeling method of super high arch dams during initial operation periods
-
摘要: 针对特高拱坝运行初期温度场的非稳定性和时效的非单调增长性,发展变形监测和预报模型,并提出构建方法。通过主成分上的分层聚类法选取代表性温度测点,将其实测值作为温度变量,引入包含徐变及其恢复项的时效变量表达式,论证其表达谷幅变形的能力。进而考虑库水位、实测温度、组合时效等变量,应用增强回归树方法提出特高拱坝运行初期变形监测和预报模型,并通过后向消减变量建立优化模型。分析各变量对变形的边际效应,得出相对影响,借助部分依赖图,辨识变量间相关关系及其对坝体变形的影响规律,揭示变形机制。将该模型应用于某特高拱坝,验证该模型的可行性和有效性;并将结果与支持向量机、多元回归模型进行对比分析,得出该模型具有显著的优越性。Abstract: The dam temperature field has not been stabilized, and the time dependent effect does not increase non-monotonously for super high arch dams during initial operation period. Therefore a special deformation monitoring and forecasting model was developed, and its modeling method was proposed in the study. The key temperature measurement points were chosen by the hierarchical clustering on principal component method, and the corresponding time series were inputted as thermal predictors. The combined time dependent effect, including creep and its restoration, was introduced as time dependent deformation. This time dependent effect was employed to validate its characterization for the valley contraction deformation. Considering the predictor variables such as reservoir water level, adopted measured temperatures, estimated time dependent effect, the simple boosted regression tree (BRT) based dam deformation prediction model was constructed. Through the backward elimination method, a simplified BRT (SBRT) model only including major predictors was obtained. The marginal effects of variables on deformation were analyzed, and the relative influences can be quantitatively analyzed. With the help of partial dependence plot, the correlations among variables and the influences of variables on deformation can be explored, and the deformation mechanism can be revealed. The model was applied to a super high arch dam, and the case study verified the feasibility of the model. The results were compared with those by the support vector machine model and the traditional multiple regression models, which shows the superiority of the developed model.
-
表 1 模型构建考虑的实测温度变量
Table 1. Measured temperature variables considered in model construction
垂线段 PL15-5 PL15-4 PL15-3 PL15-2 PL15-1 选取的温度计 T6,S3-1,S6-4,S6-9 T14,T15 S5-4,T18 S5-5,T28 T29 表 2 各模型的训练集和预测集的平均绝对误差
Table 2. Comparisons of mean absolute errors among training and prediction sets of constructed models
单位:mm 模型 SBRT SVM HTT HST 训练 预测 训练 预测 训练 预测 训练 预测 平均绝对误差 0.42 0.84 0.95 1.49 1.83 3.07 5.96 6.62 -
[1] 张冲, 王仁坤, 汤雪娟. 溪洛渡特高拱坝蓄水初期工作状态评价[J]. 水利学报,2016,47(1):85-93. (ZHANG Chong, WANG Renkun, TANG Xuejuan. Safety evaluation of Xiluodu ultra-high arch dam during the initial impoundment period[J]. Journal of Hydraulic Engineering, 2016, 47(1): 85-93. (in Chinese) [2] 何柱, 刘耀儒, 杨强, 等. 溪洛渡拱坝谷幅变形机制及变形反演和长期稳定性分析[J]. 岩石力学与工程学报,2018,37(增刊2):4198-4206. (HE Zhu, LIU Yaoru, YANG Qiang, et al. Mechanism of valley deformation of Xiluodu arch dam and back analysis and long-term stability analysis[J]. Chinese Journal of Rock Mechanics and Engineering, 2018, 37(Suppl2): 4198-4206. (in Chinese) [3] 刘宁. 国内外大坝失事分析研究[M]. 武汉: 湖北科学技术出版社, 2002. LIU Ning. Analysis and research of dam-failure[M]. Wuhan: Hubei Science and Technology Press, 2002. (in Chinese) [4] 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. (in Chinese) doi: 10.11660/slfdxb.20160501 [5] 顾冲时, 吴中如. 大坝与坝基安全监控理论和方法及其应用[M]. 南京: 河海大学出版社, 2006. GU Chongshi, WU Zhongru. Theory and method of dam and dam foundation safety monitoring and their applications[M]. Nanjing: Hohai University Press, 2006. (in Chinese) [6] HU J, WU S H. Statistical modeling for deformation analysis of concrete arch dams with influential horizontal cracks[J]. Structural Health Monitoring, 2019, 18(2): 546-562. doi: 10.1177/1475921718760309 [7] SALAZAR F, MORÁN R, TOLEDO M Á, et al. Data-based models for the prediction of dam behaviour: A review and some methodological considerations[J]. Archives of Computational Methods in Engineering, 2017, 24(1): 1-21. doi: 10.1007/s11831-015-9157-9 [8] KANG F, LI J J, DAI J H. Prediction of long-term temperature effect in structural health monitoring of concrete dams using support vector machines with Jaya optimizer and salp swarm algorithms[J]. Advances in Engineering Software, 2019, 131: 60-76. doi: 10.1016/j.advengsoft.2019.03.003 [9] SALAZAR F, TOLEDO M Á, GONZÁLEZ J M, et al. Early detection of anomalies in dam performance: A methodology based on boosted regression trees[J]. Structural Control and Health Monitoring, 2017, 24(11): e2012. doi: 10.1002/stc.2012 [10] 彭卫, 杜时贵. 混凝土徐变效应的计算机分析[J]. 西安公路交通大学学报,1999,19(2):4-5, 9. (PENG Wei, DU Shigui. Computational analysis of concrete creep effect[J]. Journal of Xi'an Highway University, 1999, 19(2): 4-5, 9. (in Chinese) [11] ABDI H, WILLIAMS L J. Principal component analysis[J]. WIREs Computational Statistics, 2010, 2(4): 433-459. doi: 10.1002/wics.101 [12] LI Q B, ZUO Z, HU Y, et al. Smart monitoring of a super high arch dam during the first reservoir-filling phase[J]. Journal of Aerospace Engineering, 2017, 30(2): B4016001. [13] SALAZAR F, TOLEDO M A, OÑATE E, et al. An empirical comparison of machine learning techniques for dam behaviour modelling[J]. Structural Safety, 2015, 56: 9-17. doi: 10.1016/j.strusafe.2015.05.001 [14] 张国新, 周秋景. 特高拱坝封拱后温度回升及影响研究[J]. 水利学报,2015,46(9):1009-1018. (ZHANG Guoxin, ZHOU Qiujing. Study on temperature recovery after arch sealing of super high arch dams and its influence[J]. Journal of Hydraulic Engineering, 2015, 46(9): 1009-1018. (in Chinese) [15] 高俊, 黄耀英, 万智勇, 等. 含冷却水管混凝土坝温度计埋设位置优选[J]. 水利水运工程学报,2017(3):93-98. (GAO Jun, HUANG Yaoying, WAN Zhiyong, et al. Optimization analysis of setting location of thermometer in concrete dam with cooling water pipes[J]. Hydro-Science and Engineering, 2017(3): 93-98. (in Chinese) -