基于AE时间序列的岩爆预测模型

基于AE时间序列的岩爆预测模型 

彭 琦1，张茹1，谢和平1，曲宏略1，龙盎2

(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)

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，岩爆预测果与现场情况一致，证明了该模型工程实用性较强。

关键词：岩爆；AE；小波神经网络；突变理论；预测模型

中图分类号：O 382     文献标识码：A 

Prediction model for rockburst based on acoustic emission time series

PENG Qi1, ZHANG Ru1, XIE He-ping1, QU Hong-lue1, LONG Ang2

(1. College of Water Resources and Hydropower, Sichuan University, Chengdu 610065, China;

2. China Three Gorges Project Corporation, Yichang 443002, China)

Abstract: Based on the features of acoustic emission(AE) time series monitored for rockburst, adopting the wavelet neural networkand catastrophe theory, a new rockburst prediction model is established. Firstly, a wavelet neural network model based on the AEmonitored is established to forecast the future AE. Secondly, a catastrophe prediction model for rockburst is founded based on AEforecasted, A practical example shows that the predicted AE time series has high prediction accuracy；and rockburst prediction areconsistent with field situation. It is shown that the model has the advantages of high forecasting accuracy and strong practicality.Key words: rockburst; acoustic emission(AE); wavelet neural network; catastrophe theory; prediction model

1 引 言

岩爆是高地应力岩石地下工程中的一种常见地质灾害，它是由于在开挖过程中卸荷引起围岩内应力场重新分布，导致储存于硬脆性围岩中的弹性应变能突然释放，使围岩产生爆裂松脱、剥落、弹射甚至抛掷的现象。岩爆的发生和发展都有一定的力学机制，但由于围岩岩体和岩爆破坏影响因素的复杂性，其形成机制一直未能形成统一的定论。