基于SVM的TBM盾构施工滚刀更换周期预测  

Prediction of TBM Shield Disc Cutters Replacement Cycle Based on SVM

在线阅读下载全文

作  者:王华桢 WANG Huazhen(China Railway 18th Bureau Group Third Engineering Co.,Ltd.,Zhuozhou,Hebei 072750,China)

机构地区:[1]中铁十八局集团第三工程有限公司,河北涿州072750

出  处:《施工技术(中英文)》2025年第1期9-14,共6页Construction Technology

基  金:中国铁建股份有限公司2024年度科技研究开发计划及资助课题(2024-C1)。

摘  要:滚刀作为TBM的重要部件,其磨损严重影响掘进效率和安全性,准确预测更换周期至关重要,然而传统方法主要依赖经验判断和简单数学模型,在复杂地质条件下效果欠佳。为此提出利用支持向量机(SVM)智能算法,结合实际工程数据,包括地质条件、推力、扭矩、转速等,建立滚刀更换周期的预测模型。重点分析了影响滚刀磨损的主要因素,并选择了线性、多项式和径向基函数(RBF)3种SVM核函数进行模型优化。研究数据来源包含13 080个样本,按80%用于训练,20%用于测试。结果显示,径向基函数SVM核函数模型在不同地层条件下预测准确率均超过80%,优于其他核函数。模型的准确率、精度和误分类率进一步验证了其在不同地质条件下的可靠性。基于SVM的预测模型能捕捉掘进过程中复杂非线性关系,具有较强的泛化能力。The hob,as an important part of TBM,its wear will seriously affect the tunneling efficiency and safety.Therefore,it is crucial to accurately predict the replacement cycle.However,the traditional method,which mainly relies on empirical judgment and simple mathematical models,is not as effective as expected under complex geological conditions.A prediction model of hob replacement cycle was established by using a support vector machine(SVM)intelligent algorithm and combined with actual engineering data,such as geological conditions,thrust,torque,speed,etc.in this paper.The main factors affecting hob wear were analyzed.And three SVM kernel functions—linear function,polynomial function and radial basis function(RBF)—were chosen for model optimization.The study includes 13,080 samples,with 80%for training and 20%for testing.The results indicate that the prediction accuracy of the RBF SVM kernel function model is more than 80%under different formation conditions,which is better than other kernel functions.The accuracy,precision and misclassification rate of the model further verify its reliability under different geological conditions.The prediction model based on SVM can capture the complex nonlinear relationship in the tunneling process and has comparatively strong generalization ability.

关 键 词:隧道 盾构 支持向量机 机器学习 换刀 预测 

分 类 号:U445[建筑科学—桥梁与隧道工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象