煤矿立井基岩含水层涌水危险 ANN 型脆弱性指数预测法

摘要

基于人工神经网络基本原理，对煤矿立井基岩含水层涌水危险性进行了预测。利用归纳分析得出的煤矿立井基岩含水层涌水主控因素，构建了煤矿立井基岩段含水层涌水危险ANN（Artificial Neural Network）型脆弱性指数预测模型，并将其运用于梵王寺煤矿副井基岩含水层涌水危险性预测工程实践。结果表明，脆弱性指数预测模型能够较好地评价立井基岩含水层涌水影响因素，预测结果对基岩段凿井施工具有重要参考价值。%Based on a basic principle of the artificial neural network ,a prediction was conducted on the water inflow danger from the aquifer in the base rock of the mine shaft. The main control factors of the water inflow from the aquifer in the base rock of the mine shaft obtained with an inductive analysis were applied to establish an ANN mode vulnerability index prediction model of the water inflow dan‐ger from the aquifer in the base rock section of the mine shaft. The model was applied to the prediction engineering practices of the water inflow danger in the base rock of an auxiliary shaft in Fanwangsi Mine. The results showed that the vulnerability index prediction model could well evaluate the water inflow influence factors of the aquifer in the base rock of the mine shaft and predicted results would have important reference value to the mine shaft sinking operations at the base rock section.