基于ELM模型的岩体质量分级及应用

Classification of rock mass quality in a mine based on ELM model and its application

为提高工程岩体质量分级的效率和准确率,提出一种岩体质量分级的极限学习机(ELM)模型。选取5项指标进行分析,采用现场30例岩体质量数据进行验证。将样本70%划分为训练集,30%划分为测试集,训练集用于优化ELM参数得到最优分级模型,然后将测试集作为最优模型输入,得到分级结果。训练集与测试集精度分别达96. 52%、83. 43%,满足工程需要。将该模型应用于某金矿岩体质量分级实例中,分级结果与现场实际情况一致。

In order to improve the efficiency and accuracy of engineering rock mass quality classification,extreme learning machine(ELM)classification model for rock mass quality was proposed.Five classification indexes were chosen and analyzed,and thirty sampling data were collected from field for verification.70%measured samples were chosen as the training set.30%samples were chosen as the test set.The training set was used to optimize ELM model parameter so as to obtain the optimal classification model.Then,the test set was taken as input of the optimal model,and the classification results were obtained.The classification accuracy of the training set and the test set are 96.52%and 83.43%respectively,meeting the demand of engineering practice.Finally,the model was applied to the rock mass quality classification of a gold mine and the classification result was consistent with the actual situation.