摘要
针对影响瓦斯涌出量的因素复杂多样化以及各因素之间的非线性问题,采用径向基核函数把支持向量机算法中的低维空间向量集映射到高维空间,进而建立基于实验数据的煤矿瓦斯涌出量预测模型。样本数据分为训练样本、测试样本和校验样本,结合MATLAB强大的运算功能,进行仿真研究。结果显示:整个系统具有较强的逼近和容错能力,以及较快的收敛速度,对煤矿瓦斯涌出量具有较好的预报效果。
Aiming at the complexity and diversity of the factors influencing coal mine gas emission, and the nonlinear problems among each factor, the set of low-dimensional vector space of support vector machine algorithm was mapped to high dimensional space based on RBF kernel function. Then the prediction model of coal mine gas emission was set up through experimental data. The data was divided into training sample, test sample and check-up sample. Combined with the vast calculation function of MATLAB software, the research on simulation was carried out. The results showed that the whole system, which possessed strong approximation capability, fault-tolerant ability and faster convergence rate, had a better prediction result on the emission of coal mine gas.
出处
《煤矿机械》
北大核心
2011年第2期78-80,共3页
Coal Mine Machinery
基金
中央高校基本科研事业费专项资金项目(DL09AB03)
佳木斯大学科学技术研究重点项目(Lz2010-013)
关键词
支持向量机
瓦斯涌出
预测
support vector machine
gas emission
prediction
作者简介
王巍（1975-），女，黑龙江哈尔滨人，博士研究生，副教授，研究方向：工业工程．
