基于BP神经网络模型的地下采煤区地表沉降预测

Prediction of Ground Subsidence of Underground Coal Mining Area Based on BP Neural Network Model

地下采煤区地表沉降预测是保障采煤区可持续发展的重要技术环节,本文以国家能源集团榆家梁煤矿52406工作面29期实测地表沉降值为数据源,以BP神经网络模型开展地表沉降预测分析。结果表明,地下采煤区地表沉降BP神经网络预测值与实测值吻合度高,模型精度为一级,为地下采煤区地表沉降预测研究与应用提供了实践应用。

Prediction of ground subsidence of underground coal-mining areas is essential to safeguard sustainable development of the coal mine. This paper analyses prediction of subsidence settlement based on BP neural network model(NNM) using the data on ground subsidence measured at 29 periods at the working face#52406 of Yujialiang Coal Mine,CNY Energy. The results prove that for the ground subsidence of underground coal mining areas,the predicted values with the BP neural network are highly consistent with the actually measured values,and the model is Level-1 in precision,providing an approach and practices for research and application to predict ground subsidence in underground coal mining areas.