首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于LWOA和MKSVM算法的帷幕灌浆施工质量模糊综合动态评价研究
引用本文:刘宗显,余 佳,吴斌平,谢怀宇,任炳昱.基于LWOA和MKSVM算法的帷幕灌浆施工质量模糊综合动态评价研究[J].水利水电技术,2011,51(6):72-83.
作者姓名:刘宗显  余 佳  吴斌平  谢怀宇  任炳昱
作者单位:天津大学 水利工程仿真与安全国家重点实验室,天津 300350
基金项目:国家重点研发计划(2018YFC0406704); 国家自然科学基金(51839007); 国家自然科学基金雅砻江联合基金(U1765205)
摘    要:为了实现施工过程中灌浆质量的动态评价,需要解决由于灌浆孔与检查孔并非一一对应造成的灌浆施工参数与灌浆质量评价指标数据间难以匹配、由于传统事后破坏性试验检测导致的灌浆质量评价指标难以动态获取以及灌浆施工质量难以动态评价的问题。因此,本研究提出基于Lévy飞行改进的鲸鱼优化算法(whale optimization algorithm improved by Lévy flight, LWOA)和混合核支持向量机(mixed kernel function support vector machine, MKSVM)的帷幕灌浆施工质量模糊综合动态评价方法。首先,基于分形理论和改进反距离加权平均方法求解了检查孔处灌浆等效施工参数,进而实现了施工参数与评价指标数据间相互匹配,建立了用于训练模型的数据集。其次,结合LWOA具有参数少、收敛速度快的特点以及混合核能够兼顾局部学习和全局泛化的优势,基于LWOA和MKSVM算法实现了灌后透水率、二次渗透指数、裂隙填充率、岩石质量等级和声波波速等评价指标的高精度动态求解。再次,采用模糊综合评价方法对帷幕灌浆施工质量进行模糊综合动态评价分析。最后,以某水电站坝基帷幕灌浆施工质量模糊综合动态评价分析为例,验证了本方法具有高精度和动态性的特性。

关 键 词:帷幕灌浆  灌浆施工质量  模糊综合动态评价  支持向量机  鲸鱼优化算法  
收稿时间:2019-09-03

Study on LWOA and MKSVM algorithms-based fuzzy comprehensive dynamic evaluation on curtain grouting construction quality
LIU Zongxian,YU Jia,WU Binping,XIE Huaiyu,REN Bingyu.Study on LWOA and MKSVM algorithms-based fuzzy comprehensive dynamic evaluation on curtain grouting construction quality[J].Water Resources and Hydropower Engineering,2011,51(6):72-83.
Authors:LIU Zongxian  YU Jia  WU Binping  XIE Huaiyu  REN Bingyu
Institution:State Key Laboratory of Hydraulic Engineering Simulation and Safety,Tianjin University,Tianjin 300350,China
Abstract:In order to realize the dynamic evaluation of the grouting quality during construction,the problems,i. e. difficult matching between the grouting construction parameters and the data of the grouting quality evaluation indexes due to that the grouting hole and the inspection hole are not one-to-one correspondent,difficult dynamic acquisition of the grouting quality evaluation indexes and difficult dynamic evaluation of grouting quality caused by the destructive test usually made after event,are necessary to be solved. Therefore,the whale optimization algorithm improved by Lévy flight,( LWOA) and the mixed kernel function support vector machine ( MKSVM) -based method for the fuzzy comprehensive dynamic evaluation on the curtain grouting construction is put forward in the study herein. Firstly,the equivalent grouting construction parameters at the inspection hole are solved in accordance with the fractal theory and the improved inverse distance weighted average method,and then the mutual matching between both the construction parameters and the evaluating index data is realized and the date-set for model training is established. Secondly,combined with the characteristics of LWOA,i. e. fewer parameters and faster convergence speed as well as the superiority of that the composite kernel can take both the local learning and the global generalization into account,the high-precision dynamic solutions of the after-grouting permeability,the secondary permeability index,the fracture filling rate,the rock quality grade and the acoustic wave velocity are realized by means of LWOA and MKSVM algorithms. Thirdly,the fuzzy comprehensive dynamic evaluative analysis on the construction quality of curtain grouting is carried out through the fuzzy comprehensive evaluation method. Finally,by taking the fuzzy comprehensive dynamic evaluative analysis on the construction quality of the curtain grouting made for the dam foundation of a hydropower station as the study case,the high accuracy and dynamic characteristics of the proposed method are verified.
Keywords:curtain grouting  grouting construction quality  fuzzy comprehensive dynamic evaluation  support vector machine  whale optimization algorithm  
点击此处可从《水利水电技术》浏览原始摘要信息
点击此处可从《水利水电技术》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号