论文部分内容阅读
首先提出了一个依据EMD(empirical mode decomposition)方法提取固有模态分量进行SVM建模实现采煤工作面瓦斯涌出量预测的技术方法.利用瓦斯涌出量的历史记录数据,通过EMD分解得出其固有模态函数,即IMF分量,然后,对应于每个固有模态分别利用SVM函数拟合方法进行外推预测,再把不同固有模态的预测结果进行叠加重构合成,获得瓦斯涌出量的理论预测结果.从监测结果的实例分析发现,与常规SVM方法相比,EMD方法的引入能够大幅度提高理论模型的预测精度,并给出监测数据极为吻合的预测结果.实际应用表明,在采煤工作面瓦斯涌出量预测建模中,固有模态的提取和SVM方法的实施都充分利用了样本数据本身驱动的自适应性质,从而为保障优异的预测效果提供了良好的理论基础.
Firstly, a technical method of extracting natural mode component based on empirical mode decomposition (EMD) method to predict gas emission from coal mining face using SVM modeling is proposed.According to historical records of gas emission, EMD Then the intrinsic mode function, namely the IMF component, is then extrapolated using the SVM function fitting method corresponding to each intrinsic mode, and the prediction results of different eigenmodes are superimposed and reconstructed to obtain gas emission The results of theoretical analysis show that compared with the conventional SVM method, the introduction of EMD method can greatly improve the prediction accuracy of the theoretical model and give a very consistent prediction of the monitoring data.Practical applications show that, In the prediction model of gas emission in coalface, both the modal extraction and the implementation of SVM make full use of the adaptive property driven by the sample data itself, which provides a good theoretical basis for ensuring excellent prediction results .