论文部分内容阅读
传统的预测底板破坏深度方法主要有力学及数理统计、定性比较分析等,然而这些方法的预测值往往达不到预期的效果或与实际值差距较大。通过对比检验样本的预测误差可得出基于灰色理论的神经网络预测模型的精度高于BP神经网络的预测结果。故此次选用灰色理论与神经网络相结合的方法建立模型预测青东矿104采区10煤底板破坏深度,预测结果为16.86m。
The traditional methods for predicting the failure depth of the floor mainly include mechanical and mathematical statistics, qualitative comparative analysis, etc. However, the predicted values of these methods often fail to achieve the expected results or have a large gap with the actual values. By comparing the prediction errors of the test samples, it can be concluded that the accuracy of the neural network prediction model based on the gray theory is higher than that of the BP neural network. Therefore, the gray theory and neural network are used to establish the model to predict the destruction depth of 10 coal floor in the 104 mining area of Qingdong Mine, the prediction result is 16.86m.