基于RS-GWO-GRNN的充填管道失效风险研究  

Study on Failure Risk of Backfill Pipeline Based on RS-GWO-GRNN

作　　者：骆正山[1] 王文辉 张新生[1] LUO Zhengshan;WANG Wenhui;ZHANG Xinsheng(School of Management, Xi'an University of Architecture &Technology, Xi'an 710055, China)

机构地区：[1]西安建筑科技大学管理学院

出　　处：《有色金属工程》2019年第6期76-83,共8页Nonferrous Metals Engineering

基　　金：国家自然科学基金项目资助(61271278);陕西省重点学科建设专项资金资助项目(E08001)~~

摘　　要：为克服充填管道失效风险评判指标间的复杂性,传统方法预测精度低及适用性差等缺陷,提出基于粗糙集(RS)和灰狼优化(GWO)算法融合广义回归神经网络(GRNN)的充填管道失效风险评价模型。选取10项风险评价指标,通过属性约简提取影响充填管道失效的主要风险因素,运用GWO优化GRNN的参数,构建预测模型,以国内某具体矿山充填系统为例进行实证研究,结果表明:与其它预测模型相比,RS-GWO-GRNN模型的预测精度更高,泛化能力更强,为充填管道失效风险研究提供了新思路,具有较好的借鉴意义。In order to overcome the complexity of the evaluation index of failure risk for backfill pipeline and the defects of traditional methods such as low prediction accuracy and poor applicability, the paper presents a new method of backfill pipeline failure risk assessment model called generalized regression neural network (GRNN) based on rough set (RS) and GWO algorithm. Ten risk evaluation indexes were selected, the main risk factors affecting filling pipeline failure were extracted through attribute reduction, and GWO was used to optimize the parameters of GRNN to build a forecasting model, taking a specific domestic mine filling system as an example for empirical research. The results show that compared with other prediction models, RS-GWO-GRNN model has higher prediction accuracy and stronger generalization ability, which provides a new idea for the research on the risk of backfill pipeline failure with good reference significance.

关 键 词：粗糙集(RS)理论 灰狼优化(GWO)算法 广义回归神经网络(GRNN) 充填管道 失效风险 

分 类 号：TD853.34[矿业工程—金属矿开采]