本文選題：碳酸鹽巖　切入點：約束稀疏脈沖反演　出處：《成都理工大學(xué)》2015年碩士論文　論文類型：學(xué)位論文

【摘要】：碳酸鹽巖儲集層是全球極為重要的產(chǎn)油氣層,其油氣儲量占已發(fā)現(xiàn)總資源量的50%以上,世界上大部分大中型油氣田的主力產(chǎn)油氣層均為碳酸鹽巖地層�？傮w來說,碳酸鹽巖儲層類型較多,內(nèi)部非均質(zhì)性強,分布發(fā)育受沉積環(huán)境和成巖作用的控制,這些特征大大增加了儲層預(yù)測的難度。波阻抗反演能夠?qū)y井資料和地震資料有效的結(jié)合起來,推算出目標(biāo)地層的波阻抗,再利用儲層和圍巖在波阻抗上的差異,將儲層識別出來,是一種為數(shù)不多的較為有效的碳酸鹽巖儲層預(yù)測手段。本文所研究的區(qū)域位于川西孝泉-新場構(gòu)造帶,該區(qū)域儲層厚度薄、地震分辨率低,常規(guī)的波阻抗反演(如約束稀疏脈沖反演)受到地震分辨率限制無法精確展示儲層與圍巖的波阻抗差異。因此,本文引入了以地質(zhì)統(tǒng)計學(xué)為基礎(chǔ)的隨機反演方法。隨機反演方法以地質(zhì)框架模型、測井和地震資料為依據(jù),對儲層油氣藏的空間分布規(guī)律、相關(guān)性以及物性參數(shù)進行隨機模擬,獲得一組等概率的儲層油氣藏參數(shù)模型,利用已知資料分析確定儲層油氣藏參數(shù)的空間分布規(guī)律(直方圖)和空間相關(guān)性(變差函數(shù)),生成即滿足測井資料,又滿足地震資料的儲層油氣藏參數(shù)模型,對不確定性進行定量分析,獲得比常規(guī)地震反演分辨率更高且又與地質(zhì)特征吻合的波阻抗和巖性信息。本文的主要是針對疊后遞推反演方法中的約束稀疏脈沖反演和地質(zhì)統(tǒng)計學(xué)反演中的隨機反演來進行實際應(yīng)用中的研究。首先介紹了地震波阻抗反演的現(xiàn)狀和發(fā)展趨勢;詳細(xì)研究了波阻抗反演的基本理論、分類以及不同反演技術(shù)的局限性;然后針對川西地區(qū)碳酸鹽巖儲層的自身特點,選擇了約束稀疏脈沖反演和基于地質(zhì)統(tǒng)計學(xué)的隨機地震反演,深入研究分析了兩種方法的原理、各項關(guān)鍵技術(shù)(包括地震及井?dāng)?shù)據(jù)處理,子波提取與合成記錄,初始模型建立,反演質(zhì)量監(jiān)控等;以及在地質(zhì)統(tǒng)計學(xué)隨機地震反演中用到的概率密度函數(shù)和變差函數(shù))和各自的優(yōu)勢特點;最后用Jason反演軟件的Inver Trace-plus模塊進行了約束稀疏脈沖反演,而Stat Mod MC模塊進行了隨機地震模擬以及地質(zhì)統(tǒng)計學(xué)隨機反演,在反演過程中,各項反演參數(shù)占了重要地位,通過反演的結(jié)果,可以對儲層大致范圍進行描述和預(yù)測。在實際應(yīng)用的基礎(chǔ)上,分析各種反演方法的主要影響因素,分析對比進行波阻抗反演得到的結(jié)果的差異。對比約束稀疏脈沖反演的結(jié)果,可以看到隨機地震反演結(jié)果的垂向分辨率有了非常明顯的提高。
[Abstract]:Carbonate reservoir is one of the most important oil and gas producing zones in the world. Its oil and gas reserves account for more than 50% of the total discovered resources. The main oil-producing reservoirs in most large and medium-sized oil and gas fields in the world are carbonate strata. There are many types of carbonate reservoirs, the internal heterogeneity is strong, the distribution and development of carbonate reservoirs are controlled by sedimentary environment and diagenesis. These characteristics greatly increase the difficulty of reservoir prediction. Wave impedance inversion can effectively combine logging data with seismic data to calculate the wave impedance of target formation, and then make use of the difference between reservoir and surrounding rock in wave impedance. Identifying the reservoir is one of the few effective means of carbonate reservoir prediction. The area studied in this paper is located in the Xiaoquan-Xinchang structural belt of western Sichuan, where the reservoir thickness is thin and the seismic resolution is low. Conventional wave impedance inversion (such as constrained sparse pulse inversion) cannot accurately show the difference between reservoir and surrounding rock wave impedance due to seismic resolution limitation. In this paper, a stochastic inversion method based on geostatistics is introduced. Based on geological frame model, logging and seismic data, the spatial distribution, correlation and physical parameters of reservoir reservoir are simulated randomly. A set of equal probability reservoir parameter models are obtained, and the spatial distribution law (histogram) and spatial correlation (variation function) of reservoir parameters are determined by analyzing the known data. The parameter model of reservoir oil and gas reservoir is satisfied with seismic data, and the uncertainty is analyzed quantitatively. Wave impedance and lithology information with higher resolution than conventional seismic inversion and consistent with geological characteristics are obtained. This paper focuses on constrained sparse pulse inversion and geostatistical inversion in poststack recursive inversion. The present situation and development trend of seismic wave impedance inversion are introduced in this paper. The basic theory, classification and limitation of different inversion techniques of wave impedance inversion are studied in detail. The constrained sparse pulse inversion and stochastic seismic inversion based on geostatistics are selected. The principles of the two methods are deeply studied and analyzed. The key techniques are seismic and well data processing, wavelet extraction and synthesis, initial modeling. Inversion quality monitoring and so on; probability density function and variation function used in geostatistics random seismic inversion) and their advantages. Finally, the constrained sparse pulse inversion is carried out using Inver Trace-plus module of Jason inversion software. The Stat Mod MC module carries out random seismic simulation and geostatistical stochastic inversion. In the inversion process, the inversion parameters play an important role. On the basis of practical application, the main influencing factors of various inversion methods are analyzed, and the differences of the results obtained from wave impedance inversion are analyzed. It can be seen that the vertical resolution of random seismic inversion results has been greatly improved.
【學(xué)位授予單位】：成都理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：P618.13;P631.4

【參考文獻】

相關(guān)期刊論文 前4條

1 賈小樂;何登發(fā);童曉光;王兆明;;全球大油氣田分布特征[J];中國石油勘探;2011年03期

2 白彥彬,楊長春,井西利;地質(zhì)規(guī)律約束下的波阻抗反演[J];石油物探;2002年01期

3 馮斌,田波,趙光宇,史秋賢,方大鈞;Jason測井約束反演在樁106地區(qū)河道含油氣砂體預(yù)測中的應(yīng)用[J];浙江大學(xué)學(xué)報(工學(xué)版);2003年06期

4 葉云飛;劉春成;劉志斌;張益明;何峰;;地質(zhì)統(tǒng)計學(xué)反演技術(shù)研究與應(yīng)用[J];物探化探計算技術(shù);2014年04期

，

本文編號：1627324