基于主成分灰关联的瓦斯涌出量预测模型

发布时间：2018-06-23 09:31

本文选题：瓦斯涌出量 + 非线性组合预测　；参考：《辽宁工程技术大学学报(自然科学版)》2017年04期

【摘要】：为分析矿井瓦斯涌出的非线性关系、指标间复杂联系和准确预测瓦斯涌出量,基于主成分分析和灰色关联理论在克服指标的共线性、相关性对瓦斯涌出量影响,兼顾二者关联性之上,确定主要指标,建立瓦斯涌出量预测的距离模型和灰色模型,进而基于支持向量机非线性,构建非线性组合预测模型.利用训练样本学习和最小绝对百分比误差确定预测模型参数,并以沈阳某矿某工作面为例,运用已构建模型预测瓦斯涌出量.研究结果表明:日产量、采出率与其他指标的共线性相对较强,煤层间距、临近层厚度及层间岩性与其他指标的共线性相对最弱;该模型绝对百分比误差最大为5.83%,预测精度相对高于各个单项预测模型,大幅降低预测风险.
[Abstract]:In order to analyze the nonlinear relation of mine gas emission, the complex relation between indexes and the accurate prediction of gas emission quantity, based on principal component analysis and grey correlation theory, the collinearity of indexes and the influence of correlation on gas emission are overcome. On the basis of the correlation between them, the main indexes are determined, and the distance model and grey model of gas emission prediction are established, and then the nonlinear combination forecasting model is constructed based on the nonlinear support vector machine (SVM). The prediction model parameters are determined by training sample learning and minimum absolute percentage error. Taking a coal face in a certain mine in Shenyang as an example, the established model is used to predict the gas emission. The results show that the collinearity of daily yield, recovery rate and other indexes is relatively strong, and the colinearity of coal seam spacing, adjacent layer thickness and interlayer lithology is the weakest. The maximum absolute percentage error of the model is 5.83, and the prediction accuracy is higher than that of each single prediction model, and the prediction risk is greatly reduced.
【作者单位】：辽宁工程技术大学系统工程研究所;辽宁工程技术大学工商管理学院;
【基金】：国家自然科学基金项目(71371091) 辽宁省高等学校杰出青年学者成长计划项目(LJQ2012027)
【分类号】：TD712.5

【相似文献】