基于神经网络的巷道围岩稳定性分类研究进展

Research Progress of Roadway Surrounding Rock Stability Classification Based on Neural Networks

系统阐述了人工神经网络的基本原理,分析了巷道围岩稳定性分类中的指标选取及分类方法。目前的大部分研究在确定分类指标的基础上,将巷道围岩稳定性分为4种或5种类型,建立高精度模型判别围岩稳定性类型成为研究的关键点。从基于BP神经网络、RBF神经网络、与其他方法结合的神经网络巷道围岩稳定性分类三方面剖析了目前的研究现状。指出非线性动态系统、多方法联合应用、分类指标科学化是基于神经网络的巷道围岩稳定性分类的发展方向。

This paper systematically elaborated the basic principle of the artificial neural network and analyzed the indexselection and classification methods in the stability classification of the roadway surrounding rock. Based on the classificationindexes presently determined in most researches, the stability of the roadway surrounding rock was classified into 4 or 5 types,building a high-accuracy model to determine the stability type of surrounding rock has become a key point in research. The paperalso dissected the current research status from three stability classifications of roadway surrounding rock based on BP neuralnetwork, RBF neural network and the neural network combined with other methods. The paper pointed out that the nonlineardynamic system, the combined application of many methods and the scientific classification indexes were the developmentdirections of the stability classification of the roadway surrounding rock based on neural network.