摘要
针对传统的主元分析存在的缺点和样本的特征,结合煤层瓦斯涌出影响因素,建立了基于权重的改进型主元分析和支持向量回归机的煤层瓦斯涌出量的预测模型.首先,采用改进型主元分析法对影响瓦斯涌出量的指标分析处理得到主成分即输入变量,然后建立以主要成分为输入变量,以瓦斯涌出量为输出变量的基于支持向量回归机的预测模型.实验表明,该模型能够消除输入变量间相关性,减少输入变量个数,有效解决样本少、模型复杂的问题,并且提高了预测精度.
Due to the shortcomings of traditional principal component analysis and the characteristics of the samples, and combined with the influence factors of coal seam gas emission, the prediction model of coal seam gas emission was established, which based on the improved principal component analysis of weight and support vector regression machine. Firstly, it used the improved principal component analy- sis method to analysis the influence factors, obtain the principal component, secondly, it used principal components as input variables and gas emission quantity as output variables , then established the predic- tion model based on support vector regression machine. Experiment results show that the model can be used to eliminate the correlation between input variables and reduce the number of input variables,it can effectively solve the problems of small sample and complex model, it also improves the accuracy of pre- diction.
作者
张文东
胡彧
ZHANG Wen-dong;HU Yu(Institude of Measuring and Controlling Technology,Taiyuan University of Technology, Taiyuan 030024, Chin)
出处
《中北大学学报（自然科学版）》
CAS
2018年第3期303-309,共7页
Journal of North University of China(Natural Science Edition)
