發(fā)布時間：2018-05-16 15:51

  本文選題：空氣質(zhì)量 + 時空數(shù)據(jù)　； 參考：《統(tǒng)計研究》2017年08期

【摘要】：空氣質(zhì)量數(shù)據(jù)具有在時間上連續(xù)、空間上相關(guān)的特點,這提高了異常點識別的難度。本文提出在時間維度上運(yùn)用移動平均法,在空間維度上運(yùn)用反距離加權(quán)法對觀測值進(jìn)行預(yù)測并求殘差的解決思路,從而將時空數(shù)據(jù)的異常點識別問題轉(zhuǎn)化為二維殘差值的異常點檢測問題。通過仿真驗證表明新方法具有良好的檢出力。最后將新方法應(yīng)用于北京市實際觀測數(shù)據(jù),取得了滿意的識別效果。
[Abstract]:Air quality data have the characteristics of continuous time and spatial correlation, which makes it more difficult to identify outliers. In this paper, the method of moving average in time dimension and inverse distance weighting method in spatial dimension are put forward to predict the observed value and solve the residual error. Thus, the problem of identifying outliers in spatiotemporal data is transformed into the problem of detecting outliers in two dimensional residuals. Simulation results show that the new method has good detection power. Finally, the new method is applied to the actual observation data in Beijing, and satisfactory recognition results are obtained.
【作者單位】： 天津大學(xué)管理與經(jīng)濟(jì)學(xué)部工業(yè)工程系;天津大學(xué)管理與經(jīng)濟(jì)學(xué)部;
【分類號】：X51

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 李澤城;基于MongoDB的滇池流域非點源污染模擬時空數(shù)據(jù)庫管理系統(tǒng)設(shè)計與實現(xiàn)[D];云南師范大學(xué);2015年

，

本文編號：1897478