关键词: 矿坑涌水量, 偏最小二乘回归, 神经网络, 预报模型

Abstract: There are many and complex factors affecting the gushing water in pit. The forecasting model of water yield of mine mostly takes into account of precipitation, surface water, irrigation and the relation of following variable and independent variable. The authors establish the forecasting model for water yield of mine, combining neural network model with the partial least square method. To deal with independent variables by the partial least square method can not only solve the relationship between independent variables but also to reduce the input dimensions in neural network model. And when the neural network is applied,it can solve the nonlinear problem better，and advance study and expression ability of the model. As the example of water yield of mine in Eighth mine, Hebi City, Henan Province, the model of water yield of mine,coupled with partial least square method and neural network, is founded and the case study shows it has rather high forecasting precision and the extending application value.

Key words: water yield of mine, partial least square method, neural network, forecasting model