兰州大学学报(自然科学版)
蘭州大學學報(自然科學版)
란주대학학보(자연과학판)
JOURNAL OF LANZHOU UNIVERSITY(NATURAL SCIENCES)
2014年
1期
39-45
,共7页
唐俊%廖朋%郝乐伟%田兵%庞国印%王琪
唐俊%廖朋%郝樂偉%田兵%龐國印%王琪
당준%료붕%학악위%전병%방국인%왕기
灰色关联分析%GM(1, n)模型%孔隙度%长8段%姬塬地区
灰色關聯分析%GM(1, n)模型%孔隙度%長8段%姬塬地區
회색관련분석%GM(1, n)모형%공극도%장8단%희원지구
grey correlation analysis%GM(1,n) model%porosity%Chang 8 member%Jiyuan Area
对姬塬地区C98井长8段(分为长81,长82小层)储层岩心分析孔隙度数据进行筛选、处理,且等间距均值化取值作为特征序列;然后与C98井测井资料孔隙度进行了精细的评价、匹配和拟合,利用灰色关联分析方法计算了孔隙度与各测井曲线的邓氏灰色关联度、灰色绝对关联度、灰色相对关联度和灰色综合关联度,按照关联度大小提取了较大的7个参数: AC, CAL, CNL, DEN, GR, RT, SP作为孔隙度预测的影响因素序列;最后建立GM(1, n)模型对目的层孔隙度进行了模拟和预测,并做了残差和相对误差分析。从整体上来看,模型预测结果与钻井取心分析资料基本趋于一致,达到了纵向上预测的目的,同时也体现了灰色预测方法具有涉及数据量小、操作简便、运算速度快的特点。
對姬塬地區C98井長8段(分為長81,長82小層)儲層巖心分析孔隙度數據進行篩選、處理,且等間距均值化取值作為特徵序列;然後與C98井測井資料孔隙度進行瞭精細的評價、匹配和擬閤,利用灰色關聯分析方法計算瞭孔隙度與各測井麯線的鄧氏灰色關聯度、灰色絕對關聯度、灰色相對關聯度和灰色綜閤關聯度,按照關聯度大小提取瞭較大的7箇參數: AC, CAL, CNL, DEN, GR, RT, SP作為孔隙度預測的影響因素序列;最後建立GM(1, n)模型對目的層孔隙度進行瞭模擬和預測,併做瞭殘差和相對誤差分析。從整體上來看,模型預測結果與鑽井取心分析資料基本趨于一緻,達到瞭縱嚮上預測的目的,同時也體現瞭灰色預測方法具有涉及數據量小、操作簡便、運算速度快的特點。
대희원지구C98정장8단(분위장81,장82소층)저층암심분석공극도수거진행사선、처리,차등간거균치화취치작위특정서렬;연후여C98정측정자료공극도진행료정세적평개、필배화의합,이용회색관련분석방법계산료공극도여각측정곡선적산씨회색관련도、회색절대관련도、회색상대관련도화회색종합관련도,안조관련도대소제취료교대적7개삼수: AC, CAL, CNL, DEN, GR, RT, SP작위공극도예측적영향인소서렬;최후건립GM(1, n)모형대목적층공극도진행료모의화예측,병주료잔차화상대오차분석。종정체상래간,모형예측결과여찬정취심분석자료기본추우일치,체도료종향상예측적목적,동시야체현료회색예측방법구유섭급수거량소、조작간편、운산속도쾌적특점。
The reservoir core porosity analysis data in Chang 8 of Jiyuan C98 well (divided into long 81 and 82 small layer) was filtered and processed so as to result in spacing averaged values as a sequence of features. Then the C98’s well-logging information on its porosity was precisely evaluated, matched and fitted, so that the gray relational analysis method could be applied to calculate the porosity log curve Deng’s, gray relational grade, absolute of gray, gray relative correlation degree and gray comprehensive correlation degree. In accordance with the size of the associated degree, the following seven parameters AC, CAL, CNL, DEN, GR, RT, SP were taken as a sequence of factors affecting the porosity prediction. Lastly, GM(1, n) model was established on simulating and predicting the target layer porosity. Meanwhile, the residual analysis and that of relative error were also conducted. Generally, the result predicted by the model is mostly consistent with drilling coring analysis data, which therefore achieves the purpose of prediction on the vertical. In addition, the advantages of the grey prediction method, i.e. involving a small amount of data, operating easily, with high operation speed, etc, are embodied.