回采工作面瓦斯涌出量遗传投影寻踪回归预测

发布时间：2018-03-21 05:55

本文选题：回采工作面　切入点：瓦斯涌出量　出处：《中国安全科学学报》2015年03期 　论文类型：期刊论文

【摘要】：为快速、准确预测回采工作面瓦斯涌出量,基于投影降维思想,建立一种遗传算法(GA)投影寻踪回归预测方法。选取煤层瓦斯原始含量、埋藏深度、煤层厚度、煤层倾角、工作面长度、推进速度、采出率、临近层瓦斯含量、临近层厚度、临近层层间距、岩层岩性、开采深度作为评价因子,对某矿15个学习样本进行训练,建立GA投影寻踪回归预测模型。利用该矿3个实测样本对模型进行检验,并与主成分分析和BP神经网络方法结果进行对比。研究表明:利用GA投影寻踪回归预测回采工作面瓦斯涌出量,平均误差为3.43%,最大误差为5.7%,精度优于其他2种方法。
[Abstract]:In order to predict gas emission quickly and accurately, a genetic algorithm (GA) projection pursuit regression prediction method is established based on the idea of reducing dimension of projection. The original content of gas in coal seam, buried depth, thickness of coal seam, inclination angle of coal seam are selected. Working face length, propelling speed, recovery rate, gas content in adjacent strata, thickness of adjacent layer, adjacent layer spacing, lithology of strata and mining depth are used as evaluation factors to train 15 study samples in a certain mine. The prediction model of GA projection pursuit regression was established, and the model was tested by three measured samples of the mine. The results are compared with the results of principal component analysis and BP neural network. The results show that the average error is 3.43, the maximum error is 5.7, and the accuracy of GA projection pursuit regression is better than the other two methods.
【作者单位】：辽宁工程技术大学安全科学与工程学院;矿山热动力灾害与防治教育部重点实验室;
【基金】：国家自然科学基金青年科学基金资助(51404126)
【分类号】：TD712.5

【参考文献】