摘要
BP神经网络模型是岩爆预测中的常用模型,为了强化预测效果,选取BFGS算法对BP神经网络模型进行优化。选取应力系数σ_(θ)/σ_(c)、脆性系数σ_(c)/σ_(t)和弹性能量指数W_(et)作为预测指标,国内外46组案例作为样本库,分别建立BFGS-BP神经网络模型和传统BP神经网络模型,对比验证其优化效果,将建好的模型用于锦屏二级水电站和秦岭隧道加以检验,得到一种有应用前景的机器学习预测模型。
BP neural network model is a common model in rock burst prediction.In order to strengthen prediction ef fect,BFGS algorithm is selected to optimize BP neural network model.The stress coefficientσ_(θ)/σ_(c),brittleness coeffi cientσ_(c)/σ_(t)and elastic energy index W_(et)are selected as the prediction indexes,and 46 groups of cases at home and abroad are used as the sample library.The BFGS-BP neural network model and traditional BP neural network model are es tablished respectively,and the optimization effect is verified by comparison.The established model is used in Jinping II Hydropower Station and Qinling Tunnel to test,and a machine learning prediction model with application prospect is obtained.
作者
郭文强
罗军尧
GUO Wenqiang;LUO Junyao(Faculty of Land Resources Engineering,Kunming University of Science and Technology,Kuming Yunan 650093,China)
出处
《工业安全与环保》
2023年第6期7-10,共4页
Industrial Safety and Environmental Protection
基金
云南省教育厅科学研究基金项目(2019Y0036)。
