摘要
针对目前TBM数据挖掘能力和掘进参数优化预测分析的不足以及对未来TBM实现无人驾驶的展望,将最小二乘支持向量机(LS-SVM)机器学习应用到TBM掘进参数预测中,从吉林引松工程TBM掘进数据中提取掘进上升段的刀盘扭矩、刀盘推力、总推力、推进速度这4个重要参数建立LS-SVM预测模型,预测4个参数在稳定段的均值,并讨论了模型训练集大小、参数选取等对预测性能的影响。结果表明,以原始数据中均匀提取的样本、RBF核函数和10折交叉验证建立的LS-SVM模型可以较为准确地预测稳定段中上述4个参数,验证了LS-SVM机器学习预测TBM掘进参数的可行性。
In view of the shortcomings of current TBM data mining capability and tunnelling parameters optimization of TBM performance prediction as well as the unmanned driving in future,this paper examines the feasibility of least squares support vector machine(LS-SVM)technology in the parameter prediction of TBM tunnelling.Four important parameters including the cutter torque,cutter thrust,total propulsion and propulsion speed of the ascending section were extracted from the TBM excavation data of the Yinsong Diversion Project to model,and this study predicted the mean value of the stable section.The influence on model prediction performance was discussed from the aspects of model size and parameter selection.The numerical results show that the LS-SVM model established by uniformly extracted samples from the original data,RBF kernel function and 10-fold cross validation can predict above four parameters in the stable segment more accurately.Therefore,the LS-SVM machine learning method is a scientific and feasible method to predict TBM tunnelling parameters.
作者
张哲铭
李晓瑜
姬建
ZHANG Zheming;LI Xiaoyu;JI Jian(College of Civil and Transportation Engineering,Hohai University,Nanjing 210098,China;Geotechnical Research Institute,Hohai University,Nanjing 210098,China)
出处
《河海大学学报（自然科学版）》
CAS
CSCD
北大核心
2021年第4期373-379,共7页
Journal of Hohai University(Natural Sciences)
基金
国家重点基础研究发展计划(973计划)(2015CB058100)
国家自然科学基金(51879091,52079045)
中央高校基本科研业务费专项(2018B01214)。
