Application of optimized wavelet neural network to the evaluation of coal seam-roof stability based on genetic algorithm

Roof stability prediction features randomness, complexity and instability, and precise prediction places much value on theoretical and practical application. In light of the latest research, this paper, according to complicated nonlinear mapping relation between various influence factors and stability of intensity, turns out that the GA-WNN model of roof stability prediction, with optimized wavelet neural network based on genetic algorithm, is constructed. This method overcomes some drawbacks of BP algorithm, such as local minimum and over-fitting. The Practical simulation results show that GA-WNN model can effectively increase the diagnostic accuracy of the network and improve the speed of convergence. This model applies to estimation of coal roof stability.