Neural Network Based on Ant Colony Clustering Algorithm Applied to Predict the Stability of the Roof in Coal Mining

The colliery roof collapse accident is one of the mine disasters .The influence factors have the characteristic of variety, non-linear, incertitude, etc., which make traditional neural prediction have to process a large amount of convoluted data. This paper gets cluster center and subjection degree through combine ant colony with fuzzy C mean value clustering algorithm, establishes the dynamic system distinguish mode that suits for non-linear form with recurrent fuzzy neural network, it obviously reduces the calculation complexity of the algorithm