【摘要】：針對礦井瓦斯監(jiān)測數(shù)據(jù)包含異常數(shù)據(jù)、存在數(shù)據(jù)缺失及數(shù)據(jù)含噪等特征,提出了瓦斯監(jiān)測數(shù)據(jù)預(yù)處理方法。首先利用移動平均線處理法或自回歸模型處理法進(jìn)行異常數(shù)據(jù)替代,然后采用三次指數(shù)平滑法補(bǔ)齊缺失數(shù)據(jù),最后通過小波軟閾值法進(jìn)行數(shù)據(jù)消噪處理。實(shí)例分析表明,該方法可在不改變瓦斯監(jiān)測數(shù)據(jù)統(tǒng)計特征的基礎(chǔ)上,消除異常數(shù)據(jù)的干擾,保證監(jiān)測數(shù)據(jù)的完整性,使監(jiān)測數(shù)據(jù)表現(xiàn)特征平滑、波動性較小。
[Abstract]:Aiming at the characteristics of mine gas monitoring data including abnormal data, lack of data and data noise, the preprocessing method of gas monitoring data is put forward. First, the method of moving average line processing or autoregressive model processing is used to replace the abnormal data, then the cubic exponential smoothing method is used to correct the missing data, and finally the wavelet soft threshold method is used to remove the noise of the data. The analysis of examples shows that the method can eliminate the interference of abnormal data and ensure the integrity of monitoring data without changing the statistical characteristics of gas monitoring data, so that the characteristics of monitoring data are smooth and the volatility is low.
【作者單位】： 西安科技大學(xué)能源學(xué)院;西安科技大學(xué)教育部西部礦井開采及災(zāi)害防治重點(diǎn)實(shí)驗(yàn)室;天地(常州)自動化股份有限公司;
【基金】：國家自然科學(xué)基金資助項(xiàng)目(51104116) 西安科技大學(xué)博士啟動基金資助項(xiàng)目(2012QDJ030)
【分類號】：TD712
，

本文編號：2181915