摘要
井下瓦斯浓度预测是预防煤矿瓦斯事故的重要环节和基础工作。以预测煤矿瓦斯浓度为研究目的,采用灰色系统理论与自记忆原理相结合的方法,将灰色系统理论导出的煤矿瓦斯浓度变化微分方程代入由自记忆原理推导的离散形式自记忆方程,利用最小二乘法求得记忆系数,建立了煤矿瓦斯浓度预测的灰色自记忆模型。结合李雅庄煤矿304综采面瓦斯浓度实测值,由试算法确定最优回溯阶p=7,建立瓦斯浓度预测灰色自记忆模型,并与G(1,1)模型进行对比分析。研究表明,灰色自记忆模型综合了灰色系统理论和自记忆原理的优越性,能够准确拟合与预测出井下瓦斯浓度变化的总体趋势与波动细节,有较好的工程适应性和较高的预测精度,为井下瓦斯浓度预测提供新的途径。
Prediction of the coalmine gas concentration is an important part and the fundamental task of gas accident prevention in coalmine.In order to predict the coalmine gas concentration,the grey self-memorization model is established by combining the grey system theory with the self-memorization theory.Substituting the differential equation deduced from the grey system theory into the discrete selfmemorization equation,the memorization coefficient of the grey self-memorization model for the coalmine gas concentration is calculated by the least-squares method.The model is applied to predict the gas concentration at the 304 comprehensive mining coal face in Liyazhuang coal mine and the result is compared with that of the grey G（1,1） model.The optimal awkward moment of the grey self-memorization model is determined to be seven through trial methods.The research results show that the grey self-memorization model can combine the merits of the self-memorization theory and the grey system theory,to predict the overall trend and the fluctuation of coalmine gas concentration.The proposed method enjoys a good accuracy in forecasting various engineering events,especially,as a new approach to predict the coalmine gas concentration.
出处
《科技导报》
CAS
CSCD
北大核心
2010年第17期58-62,共5页
Science & Technology Review
关键词
瓦斯浓度
灰色自记忆
自记忆原理
动态预测
gas concentration
grey self-memorization
self-memorization theory
dynamic prediction
