基于粗糙集理论和支持向量机的岩爆预测

来源期刊:中南大学学报(自然科学版)2017年第5期

论文作者:李宁 王李管 贾明涛

文章页码:1268 - 1276

关键词:岩爆预测;支持向量机;粒子群算法;粗糙集理论

Key words:rockburst prediction; support vector machine; particle swarm optimization algorithm; rough set theory

摘    要:为了提高不同环境和地质条件下对岩爆预测的准确性,在综合岩爆影响因素的基础上,选取岩石取样处的埋深、岩石单轴抗压强度,岩石单轴抗压强度与抗拉强度比值、围岩最大切向应力与岩石单轴抗压强度比值、岩石弹性变形能指数作为评判指标建立岩爆烈度预测决策表,根据粗糙集理论中的属性约简算法,确定特定地质条件下岩爆的主要影响因素,删除冗余数据,再使用粒子群算法优化支持向量机的参数,通过核函数将岩爆主控因素映射到高维空间,拟合主控因素与岩爆烈度之间的非线性映射关系,建立基于粗糙集理论和粒子群支持向量机(RS-PSOSVM)的岩爆预测模型,并将该模型应用于大相岭隧道的岩爆预测。研究结果表明:该模型具有较高准确率和和较强稳定性;岩爆预测结果与实际结果一致,验证了该模型的可行性。

Abstract: In order to improve the accuracy of rockburst prediction in different environments and geologies, on the basis of comprehensive influence factors of rockburst, the intensity of rockburst prediction decision table was established according to the evaluation indicators, including burial depth of the rock sample,rocks’ uniaxial compressive strength, the ratio of the uniaxial compressive strength to the uniaxial tensile strength of rock, the ratio of the maximum tangential stresses on cavern boundaries to the uniaxial compressive strength of rock and the elastic energy index of rock. By the attribute reduction of rough set theory(RS), the main factors of rockburst under specific geological conditions were determined, and redundant data were removed. Using particle swarm optimization (PSO) to optimize parameters of support vector machine (SVM), the main control factors of rock burst were mapped to high-dimensional space through kernel function, and the nonlinear relationship between main control factors and intensity of rockburst was fitted. Finally, a rockburst prediction model based on set theory (RS), particle swarm optimization (PSO) and support vector machine (SVM) was established. The model was applied in Daxiangling tunnel. The results show that this model has high accuracy and stability. The predict results agree well with the actual results.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号