Prediction model for the scale of karst cave based on back propagation artificial neural network and its application

In complex karst region, the size of karst cave is affected by many factors, such as geological structure, properties of soluble rock and groundwater hydrodynamic system and so on, which is characterized by high complexity and nonlinearity. Through the study of the occurrence and development of karst caves in karst area, the control factors affecting the scale of karst cave are determined and quantitatively analyzed, for which the data of proved caves are collected. In order to solve the problem with data fuzziness and descriptive formation of the karst caves, in this paper, the method of Back Propagation (BP) artificial neural network is employed to achieve the prediction of the scale of karst caves. As a BP neural network model is self-organization and self-adaptive, it is expected to handle the nonlinearity of sample data. The model is designed, tested, and applied, based on the MATLAB R2012a software. The results show that BP artificial neural network prediction model for the scale of karst cave is of high accuracy with its algorithm of good convergence.