Prediction of mining subsidence under thin bedrocks and thick unconsolidated layers based on field measurement and artificial neural networks

Abstract

The deformation characteristics of subsidence and movement induced by mining under thin bedrocks and thick unconsolidated layers are researched using field measurement and the prediction method of artificial neural networks (ANN). Firstly, the occurrence characteristics of thin bedrock and thick unconsolidated layers were analyzed in a research coal field. Based on the measured data, the characteristics of ground movement show that the surface subsidence deformation of mining under thin bedrock is more intensive than that of mining under normal thickness bedrock. Such is evident through the settlement time concentrating, the maximum surface subsidence being greater than the thickness of coal seam, the distribution of ground movement and deformation being concentrated, the range extension being wide, the active period being intensive and concentrated, the surface damage being very serious, and the crack development being significant. A quantitative prediction method is made on mining subsidence under thin bedrocks and thick unconsolidated layers by means of ANN. The improved neural network was used for modeling and predicting the mining subsidence. The ANN output can reflect the change trend of ground movement and deformation. The forecasting results are in good agreement with the real observation results.