论文部分内容阅读
为更有效预测回采工作面绝对瓦斯涌出量,基于Lyapunov稳定性原理,改进Elman模型的递归部分。选取煤层瓦斯含量、煤层埋藏深度、煤层厚度、煤层倾角、采高、日工作进度、工作面长度、工作面采出率、邻近层瓦斯含量、邻近层厚度、邻近层间距、开采强度和层间岩性作为监测指标,对某矿16个学习样本进行训练,建立隐层递归反馈(HRF)Elman预测模型。利用矿井监测数据检验预测模型。试验结果表明,用HRF Elman模型能够有效地预测出瓦斯涌出量,预测结果相对误差为1.6%~3.41%,平均相对误差为2.45%,相比传统的Elman模型,预测精度和效率都有所提高。
In order to predict the absolute gas emission from mining face more effectively, the recursive part of Elman model is improved based on Lyapunov stability principle. Select the gas content of coal seam, depth of coal seam, coal seam thickness, coal seam dip angle, mining height, daily work progress, working face length, face mining rate, adjacent layer gas content, adjacent layer thickness, adjacent layer spacing, Lithology as a monitoring index, 16 learning samples of a mine were trained to establish a hidden layer recursive feedback (HRF) Elman prediction model. Predicting Model Using Mine Monitoring Data. The experimental results show that the gas emission can be effectively predicted by HRF Elman model. The relative error of prediction results is 1.6% -3.41% and the average relative error is 2.45%. Compared with the traditional Elman model, the prediction accuracy and efficiency are both improve.