摘要
通过引入遗传算法优化BP神经网络,构建GA-BP神经网络模型。选取围岩最大切向力与岩石单轴抗压强度比(应力集中系数)、岩石单轴抗压强度与单轴抗拉强度比(脆性系数)和弹性能量指数作为输入指标,构建岩爆烈度分类预测体系。选取104组工程岩爆实例,其中84组作为训练集,20组作为测试集进行验证,结果表明,GA-BP神经网络模型的分类预测准确率能够达到95%,优于BP神经网络模型的80%。在工程试验中,GA-BP神经网络模型分类预测效果较好(准确率90%),可为岩爆烈度分类预测研究作为参考。
By introducing genetic algorithm to optimize BP neural network,GA-BP neural network model is constructed.The maximum tangential force and uniaxial compressive strength ratio of rock(stress concentration coefficient),uniaxial compressive strength and uniaxial tensile strength ratio of rock(brittle⁃ness coefficient)and elastic energy index are selected as input indexes to construct a classification and pre⁃diction system of rockburst intensity.The results show that GA-BP neural network model has a classification prediction accuracy of 95%,which is better than 80%of BP neural network model.In engineering experi⁃ments,GA-BP neural network model has a good classification and prediction effect(90%accuracy),which can be used as a reference for the classification and prediction of rockburst intensity.
作者
滕涛
王国军
周伟胜
倪智伟
景杨凡
TENG Tao;WANG Guojun;ZHOU Weisheng;NI Zhiwei;JING Yangfan(Shandong Zhengyuan Construction Engineering Co.,Ltd.;Sinosteel Maanshan General Institute of Mining Research Co.,Ltd.;State Key Laboratory of Metal Mine Safety and Health;Huawei national Engineering Research Center for High Efficiency Recycling of Metal Mineral Resources Co.,Ltd.)
出处
《现代矿业》
CAS
2023年第9期278-281,共4页
Modern Mining
关键词
遗传算法
BP
神经网络
岩爆烈度
分类预测
genetic algorithm
BP neural network
rockburst intensity
classification prediction
作者简介
滕涛(1979-),男,工程师,250000山东省济南市。
