基于最小二乘支持向量机的瓦斯涌出量预测模型  

The Gas Emission Prediction Model Based on Least Squares Support Vector Machine

作　　者：张幸幸[1] 苗星[1] 

机构地区：[1]河南师范大学计算机与信息工程学院,河南新乡453007

出　　处：《科技视界》2014年第7期287-288,共2页Science & Technology Vision

摘　　要：瓦斯涌出量的准确预测对于通风系统的设计、瓦斯防治、安全管理有着重要意义,可以有效减轻采煤工作面的危险程度,同时提高煤矿业在燃料市场中的竞争能力。本文简述了瓦斯涌出量预测的价值和曾出现的各种方法,提出了基于遗传算法(Genetic Algorithms,GA)和最小二乘支持向量机(least square support vector machine,LS-SVM)的短期瓦斯涌出量预测方法,以某煤矿回采工作面瓦斯涌出量与影响因素为例,建立了GA-LSSVM预测模型,根据上述煤矿的数据进行实例验证,结果表明文中的方法显著优于神经网络的预测结果。Accurate prediction of gas emission has important significance to the design of the ventilation system, gas control and safety management, and it can reduce the danger degree of coal mining work face effectively, At the same time ,it also improves the market competition ability of the coal in the fuel. This paper expounds the value of gas emission prediction and a variety of methods having appeared, and proposed a short-term gas emission prediction method based on Genetic algorithm （GA） and least squares support vector machine （LS-SVM）.Taken gas emission quantity and influenced factors in a coal mine working face for an example, the GA-LS-SVM forecasting model is established. According to the data of coal mine, the results show that the method of this paper was superior to that of the neural network prediction results.

关 键 词：瓦斯涌出量预测 遗传算法 最小二乘支持向量机 神经网络 

分 类 号：TD712.5[矿业工程—矿井通风与安全]