实时监控下基于二进制蝴蝶优化算法与特征选择的堆石坝压实质量评估研究Compaction quality evaluation of rockfill dam based on feature selection algorithm with b-BOA under real-time monitoring
陈洪春,张显羽,陈文龙,黄文龙,邱伟,游秋森
摘要(Abstract)：
传统压实质量检测手段具有取点有限、效率低、反馈不及时等不足,实时压实质量评估已发展为堆石坝压实质量监控的重要内容。基于碾轮振动加速度信号分析的CMV、CCV指标获取简单快速、难度低,但这些指标与压实质量之间的关系容易受到碾压参数和料源参数的影响。针对上述问题,综合考虑碾压参数、料源参数以及CMV、CCV指标,提出了一种基于二进制蝴蝶优化算法(binary Butterfly Optimization Algorithm, b-BOA)和XGBoost特征选择的堆石坝压实质量评估模型。结合现场碾压试验,将该模型应用于某工程全强风化料区的压实质量评估。结果表明,本文提出的模型考虑碾压参数和料源参数对CMV、CCV指标的影响,可有效地进行特征选择,同时与常用压实质量评估模型相比,具有更佳的评估性能。
关键词(KeyWords)： 堆石坝;压实质量评估;特征选择;二进制蝴蝶优化算法;XGBoost
基金项目(Foundation): 国家自然科学基金(51779169);; 国家自然科学基金雅砻江联合基金(U1765205)
作者(Author): 陈洪春,张显羽,陈文龙,黄文龙,邱伟,游秋森
DOI: 10.13928/j.cnki.wrahe.2021.S2.055
参考文献(References)：
- [1] 中国电力企业联合会.碾压式土石坝施工规范：DL/T 5129—2013[S].北京：电力出版社，2014.
- [2] 钟登华，刘东海，崔博.高心墙堆石坝碾压质量实时监控技术及应用[J].中国科学：技术科学，2011,41(8):1027-1034.
- [3] THUMER H,SANDSTROM A.Continuous compaction control,CCC[C]//European Workshop Compaction of Soils and Granular Materials.Paris:Presses Ponts et Chauss-ees,2000:237-246.
- [4] NOHSE Y,KITANO M.Development of a new type of single drum vibratory roller[C]// Proc.14th Intl.Conf.of the Intl.Soc,Vicksburg:For Terrain-Vehicle Systems,2002:1-10.
- [5] HUA T,YANG X,YAO Q,et al.Assessment of Real-Time Compaction Quality Test Indexes for Rockfill Material Based on Roller Vibratory Acceleration Analysis[J].Advances in Materials Science and Engineering,2018:2879321.
- [6] MICHAEL A.MOONEY,ROBERT V.RINEHART.Field Monitoring of Roller Vibration during Compaction of Subgrade Soil[J].Journal of Geotechnical and Geoenvironmental Engineering,2007,133(3):257-265.
- [7] KROBER W,FLOSS R,WALLRATH W.Dynamic soil stiffness as quality criterion for soil compaction[C]//Geotechnics for roads,rail tracks,and earth structures.Netherlands:A.A.Balkema,Lisse,2001:188-199.
- [8] MEEHAN C L,CACCIOLA D V,TEHRANI F S,et al.Assessing soil compaction using continuous compaction control and location-specific in situ tests[J].Automation in Construction,2017,73:31-44.
- [9] 刘东海，李子龙，王爱国.基于碾压机做功的堆石坝压实质量实时监测与快速评估[J].水利学报，2014,45(10):1223-1230.
- [10] 刘东海，刘志磊，冯友文.堆石坝料压实监测指标影响因素及适用性分析[J].水力发电学报，2019,38(6):1-10.
- [11] 安再展，刘天云，皇甫泽华，等.利用CMV评估堆石料压实质量的神经网络模型[J].水力发电学报，2019,38(12):1-11.
- [12] CHEN T,GUESTRIN C.XGBoost:A Scalable Tree Boosting System[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,San Francisco:Assoc Comp Machinery,2016:785-794.