Stability Classification of Artificial Dam of Groundwater Reservoir Based on Neural Network

Xiang-song KONG, Jie FANG, Shuan WANG, Xiao-lin HAO, Hao-chen ZHANG, Han-sen ZHANG

Abstract


The artificial dam is a key part of the groundwater reservoir of the coal mine, but the engineering geological conditions of the artificial dam are very complicated, and the artificial dam is easy to be unstable, leading to engineering accidents. Therefore, the classification of artificial dam stability is of great significance to the design and maintenance of the dam. Based on the principles of clarity, importance and independence, combined with the actual situation of the mining area, 10 stability classification indexes were determined. The basic principle of artificial neural network is expounded. BP neural network is improved by additional momentum method. The stability identification model of artificial neural network is established by using MATLAB software. 30 artificial dams are selected as samples to study and train the model. The model was applied to five artificial dams for inspection, and the recognition accuracy of the results was 100%. It can be seen that the model has a high degree of type recognition, and the nonlinear mapping effect is very good, which is suitable for classifying the stability of the groundwater reservoir artificial dam.

Keywords


Neural network, Classification index, Stability classification, Artificial dam


DOI
10.12783/dtcse/aicae2019/31446

Full Text:

PDF

Refbacks

  • There are currently no refbacks.