本文選題：角度域共成像點(diǎn)道集 + 地震屬性優(yōu)選��； 參考：《中國石油大學(xué)(華東)》2015年碩士論文

【摘要】：疊前地震數(shù)據(jù)中含有十分豐富的物性、巖性、流體等信息,需要綜合利用多種方法把反映這些儲(chǔ)層信息的參數(shù)提取出來,地震屬性是獲得這些參數(shù)的一個(gè)重要途徑,通過對(duì)提取的地震屬性進(jìn)行分析可以獲得較為準(zhǔn)確的儲(chǔ)層信息。地下地層的厚度、孔隙度、滲透率、飽和度等特征參數(shù)只能通過鉆井信息來確定,而從井中所得到的各種特征參數(shù)僅是井點(diǎn)上的信息,只能確定一些離散位置點(diǎn)上的儲(chǔ)層信息。因此,要了解整個(gè)三維空間上儲(chǔ)層特征參數(shù)分布的規(guī)律,必須依靠地震數(shù)據(jù)在空間上進(jìn)行展開,這就要建立地震數(shù)據(jù)與由測井?dāng)?shù)據(jù)獲得的各種儲(chǔ)層特征參數(shù)之間的關(guān)系。首先,通過相關(guān)性分析發(fā)現(xiàn),相比基于疊后地震數(shù)據(jù)的常規(guī)屬性,在疊前角度域共成像點(diǎn)道集(ADCIGs)上提取的地震屬性包含更加豐富的原始地震信息,疊加會(huì)損失疊前地震資料里微弱的有效信息,因此在ADCIGs上提取常規(guī)地震屬性,同時(shí)提取疊前屬性P、G,與測井?dāng)?shù)據(jù)一起組成樣本集,采用標(biāo)準(zhǔn)差標(biāo)準(zhǔn)化方法對(duì)樣本集數(shù)據(jù)進(jìn)行標(biāo)準(zhǔn)化處理。進(jìn)而,提出了將遺傳算法、支持向量回歸機(jī)算法結(jié)合的屬性優(yōu)選方法(GA-SVR),同時(shí)將禁忌搜索算法(TS)引入到屬性優(yōu)選中,將GA-SVR與TS進(jìn)行對(duì)比、分析,選擇了GA-SVR方法對(duì)樣本集進(jìn)行地震屬性優(yōu)選。最后,利用支持向量回歸機(jī)(SVR)建立屬性優(yōu)選后的樣本集與Gamma、Porosity測井?dāng)?shù)據(jù)等儲(chǔ)層特征參數(shù)之間的非線性估算模型,進(jìn)而對(duì)三維工區(qū)進(jìn)行儲(chǔ)層特征參數(shù)估算。針對(duì)Opendtect中的F3演示數(shù)據(jù)和勝利油田某工區(qū)實(shí)際數(shù)據(jù)進(jìn)行了方法試算,估算結(jié)果在井位置處與實(shí)際測井?dāng)?shù)據(jù)有較高的一致性,在遠(yuǎn)離井的位置,與地震數(shù)據(jù)、地質(zhì)背景相吻合,驗(yàn)證了本研究內(nèi)容的有效性,為油田的石油勘探提供了參考。
[Abstract]:The prestack seismic data contain abundant physical property, lithology, fluid and so on, so it is necessary to extract the parameters reflecting these reservoir information by comprehensive use of many methods. Seismic attribute is an important way to obtain these parameters. More accurate reservoir information can be obtained by analyzing the extracted seismic attributes. The characteristic parameters of underground formation such as thickness porosity permeability and saturation can only be determined by drilling information. Therefore, in order to understand the distribution law of reservoir characteristic parameters in the whole three-dimensional space, we must rely on the spatial expansion of seismic data, which is necessary to establish the relationship between seismic data and various reservoir characteristic parameters obtained from logging data. First of all, through correlation analysis, it is found that compared with the conventional attributes based on post-stack seismic data, the seismic attributes extracted from the common imaging point gathers in prestack angle domain (ADCIGs) contain more abundant original seismic information. Stacking will lose the weak and effective information in the prestack seismic data, so the conventional seismic attributes are extracted from the ADCIGs, and the prestack attributes are extracted to form the sample set together with the logging data. The standard deviation standardization method is used to standardize the sample set data. Furthermore, a new attribute selection method, which combines genetic algorithm and support vector regression algorithm, is proposed. At the same time, Tabu search algorithm is introduced into attribute selection, and GA-SVR and TS are compared and analyzed. The GA-SVR method is selected to select the seismic attributes of the sample set. Finally, the nonlinear estimation model between the sample set after attribute selection and the reservoir characteristic parameters such as Gamma-Porosity logging data is established by using support vector regression machine (SVR), and then the reservoir characteristic parameters are estimated in 3D working area. Based on the F3 demonstration data in Opendtect and the actual data in a certain working area of Shengli Oilfield, the estimated results are in good agreement with the actual logging data at the well location, the location far away from the well, the seismic data and the geological background. The validity of this study is verified and a reference for oil field exploration is provided.
【學(xué)位授予單位】：中國石油大學(xué)(華東)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2015
【分類號(hào)】：P631.4;P618.13

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 蘇建龍;米鴻;王彥春;張豐麒;唐靜;;基于支持向量機(jī)的非線性彈性阻抗反演方法[J];石油地球物理勘探;2014年04期

2 王開燕;徐清彥;張桂芳;程某存;李培海;;地震屬性分析技術(shù)綜述[J];地球物理學(xué)進(jìn)展;2013年02期

3 朱怡翔;石廣仁;;火山巖巖性的支持向量機(jī)識(shí)別[J];石油學(xué)報(bào);2013年02期

4 王新桐;盧雙舫;肖佃師;;基于聚類分析的地震屬性優(yōu)化及儲(chǔ)層預(yù)測——以敖包塔油田敖9工區(qū)為例[J];石油天然氣學(xué)報(bào);2013年03期

5 張爾華;關(guān)曉巍;張?jiān)?;支持向量機(jī)模型在火山巖儲(chǔ)層預(yù)測中的應(yīng)用——以徐家圍子斷陷徐東斜坡帶為例[J];地球物理學(xué)報(bào);2011年02期

6 肖冬生;;疊前地震屬性在濁積巖儲(chǔ)層預(yù)測中的應(yīng)用——以興隆臺(tái)-馬圈子地區(qū)沙三中下亞段為例[J];石油天然氣學(xué)報(bào);2010年06期

7 袁野;劉洋;;地震屬性優(yōu)化與預(yù)測新進(jìn)展[J];勘探地球物理進(jìn)展;2010年04期

8 唐小彪;;基于對(duì)應(yīng)分析的支持向量機(jī)回歸在地震儲(chǔ)層厚度預(yù)測中的應(yīng)用[J];物探與化探;2009年04期

9 唐耀華;張向君;高靜懷;;基于地震屬性優(yōu)選與支持向量機(jī)的油氣預(yù)測方法[J];石油地球物理勘探;2009年01期

10 楊培杰;印興耀;;基于支持向量機(jī)的疊前地震反演方法[J];中國石油大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年01期

相關(guān)碩士學(xué)位論文 前3條

1 于文彬;基于偏最小二乘及最小二乘支持向量機(jī)算法在儲(chǔ)層參數(shù)預(yù)測上的應(yīng)用[D];吉林大學(xué);2011年

2 倪鳳田;基于地震屬性分析的儲(chǔ)層預(yù)測方法研究[D];中國石油大學(xué);2008年

3 倪艷;Isomap算法在地震屬性參數(shù)降維中的應(yīng)用[D];成都理工大學(xué);2007年

，

本文編號(hào)：1821390