基于AE时间序列的岩爆预测模型  

Prediction model for rockburst based on acoustic emission time series

作　　者：彭琦[1] 张茹[1] 谢和平[1] 曲宏略[1] 龙盎[2] 

机构地区：[1]四川大学水利水电学院,成都610065 [2]中国长江三峡工程开发总公司,湖北宜昌443002

出　　处：《岩土力学》2009年第5期1436-1440,共5页Rock and Soil Mechanics

基　　金：国家基础研究发展规划(973)项目(No.2002CB412705);国家自然科学基金创新群体项目(No.50221402);国家自然科学基金委雅砻江水电开发联合研究基金重点项目(No.50639100);国家自然科学基金项目:“岩石灾变破裂过程中能量耗散与释放规律的研究”(No.50579042)

摘　　要：根据现场岩爆监测中声发射(AE)时间序列的特点,采用小波神经网络与突变理论,建立了一种新的岩爆预测模型。该模型首先针对监测到的声发射建立小波神经网络模型,对声发射时间序列进行了拟合与预测;再运用突变理论对预测的声发射建立了岩爆突变预测模型。通过实例分析表明,声发射的预测精度较高,岩爆预测结果与现场情况一致,证明了该模型工程实用性较强。Based on the features of acoustic emission（AE） time series monitored for rockburst, adopting the wavelet neural network and catastrophe theory, a new rockburst prediction model is established. Firstly, a wavelet neural network model based on the AE monitored is established to forecast the future AE. Secondly, a catastrophe prediction model for rockburst is founded based on AE forecasted. A practical example shows that the predicted AE time series has high prediction accuracy; and rockburst prediction are consistent with field situation. It is shown that the model has the advantages of high forecasting accuracy and strong practicality.

关 键 词：岩爆 AE 小波神经网络 突变理论 预测模型 

分 类 号：O382[理学—流体力学]