神经网络在矿山竖井工程围岩质量分级中的分析与应用

矿山竖井工程围岩质量分级中影响围岩质量分级的因素众多,各个因素间的非线性作用关系复杂,围岩分级过程中人为因素影响大,分级结果的准确性较差。神经网络通过合适的样本学习,能自动建立各个因素与围岩质量分级间的对应关系,能很好的解决类似矿山竖井围岩质量分级评价。从围岩介质特性、环境条件以及工程因素3个方面系统分析了影响岩体质量分级的因素指标,构建了围岩质量分级的神经网络模型,根据工程实例建立学习样本,经过对网络模型的训练与检验,证实神经网络具有较好的收敛性和稳定性,在岩体质量分级中应用具有很好的实用性。

ANALYSIS AND USE OF NEURAL NETWORK IN CLASSIFICATION OF MINE SHAFT SURROUNDING ROCK-MASS

The factors affecting the engineering classification of rock-mass of Mine Shaft are numerous,among which the nonlinear relation is very complicated,and human factors will produce powerful influence in the engineering classification of rock mass,classification result is worse. Neural network training through suited learning samples can automatically build up corresponding relationship among the factors and engineering classification of rock mass,this ability of neural net work is suitable to classification of rock mass. In this paper,factors affecting classification of rock mass are analyzed through rock-mass mediums character,environment conditions and engineering factor,the neural network model of classification of rock-mass are constructed,and build up training samples,and through the training and testing of the neural network it is proved that neural network can achieve a fast convergence speed and better stability,that can be practically applied in classification of rock mass.