摘要
为提高露天矿边坡监测数据的解释性与实时响应能力,提出了一种基于支持向量机(SVM)与简单位移平均(SMA)方法相结合的监测数据处理框架。该方法首先利用SVM模型剔除因测量误差、环境扰动等因素引入的异常值,以提高数据的可信度与稳健性;然后,采用SMA平滑处理数据,以趋势项替代逐点位移值并提取整体趋势,减弱单一异常位移值对整体曲线的影响;处理后的数据曲线相比原始数据在趋势显著性和简洁性方面均得到显著提升,更加清晰地反映了边坡位移的实际变化趋势。此方法可有效增强露天矿边坡监测的敏感性和趋势识别能力,从而有助于实现边坡位移的及时预警与有效管理。
To improve the interpretability and real-time response capability of slope monitoring data in open-pit mines,this article proposes a monitoring data processing framework based on the combination of support vector machine(SVM)and simple displacement averaging(SMA)methods.This method first uses an SVM model to remove outliers introduced by measurement errors,environmental disturbances,and other factors,in order to improve the credibility and robustness of the data.Then,SMA smoothing is used to process the data,replacing point by point displacement values and extracting overall trends,reducing the impact of a single abnormal displacement value on the overall curve.The processed data curve has significantly improved in both trend significance and conciseness compared to the original data,reflecting more clearly the actual trend of slope displacement changes.This method effectively enhances the sensitivity and trend recognition ability of open-pit mine slope monitoring,thereby helping to achieve timely warning and effective management of slope displacement.
作者
白继元
肖航
武志高
BAI Jiyuan;XIAO Hang;WU Zhigao(Guoneng Xinjiang Mining Hongshaquan Second Mine Co.,Ltd.,Changji 831100,China)
出处
《露天采矿技术》
2025年第3期13-17,共5页
Opencast Mining Technology
关键词
支持向量机
简单位移平均
趋势项
鲁棒性
数据分析
support vector machine
simple displacement averaging
trend item
robustness
data analysis
作者简介
白继元(1990-),男,山西朔州人,工程师,硕士,从事露天煤矿生产方面的技术工作。E-mail:947446005@qq.com。