基于改進(jìn)蟻群追蹤策略的地震層位自動(dòng)識(shí)別方法
本文選題:層位自動(dòng)追蹤 + 地震數(shù)據(jù)預(yù)處理。 參考:《石油地球物理勘探》2017年03期
【摘要】:目前層位拾取和標(biāo)定更多地是依靠人工或機(jī)器輔助的方式進(jìn)行,如同相軸追蹤、基于神經(jīng)網(wǎng)絡(luò)和圖像邊緣提取技術(shù)的層位拾取,業(yè)界的地震層位追蹤方法偏重于三維層位追蹤和剖面自動(dòng)追蹤算法。存在效率低、需要人為指定種子點(diǎn)、訓(xùn)練追蹤時(shí)間長(zhǎng)等缺陷,且每次只能完成一個(gè)層位(面)追蹤,無(wú)法對(duì)層間關(guān)系進(jìn)行解釋。為此,提出了基于改進(jìn)蟻群追蹤策略的地震層位自動(dòng)識(shí)別方法。首先采用改進(jìn)的加權(quán)中值濾波對(duì)地震數(shù)據(jù)進(jìn)行預(yù)處理,在增強(qiáng)同相軸連續(xù)性的基礎(chǔ)上,減少了傳統(tǒng)種子點(diǎn)生長(zhǎng)算法中的人為參與;基于蟻群算法的基本思想,引入支持向量機(jī)(SVM)技術(shù)對(duì)地震數(shù)據(jù)進(jìn)行分類處理,同時(shí)綜合考慮地震振幅、瞬時(shí)相位、層位傾角、信息素濃度等多種信息,納入蟻群層位追蹤的評(píng)價(jià)函數(shù),采用改進(jìn)的蟻群搜索算法實(shí)現(xiàn)地震層位自動(dòng)追蹤。通過(guò)對(duì)追蹤區(qū)域進(jìn)行劃分,實(shí)現(xiàn)區(qū)域內(nèi)小層加密,提供多種合理的層位加密策略,完成地震層位追蹤后期處理。實(shí)際資料應(yīng)用結(jié)果檢驗(yàn)了方法的有效性,為層序地層分析、沉積體系域解釋、精細(xì)儲(chǔ)層描述提供了技術(shù)思路。
[Abstract]:At present, layer pickup and calibration mostly depend on manual or machine aided methods, such as phase axis tracking, layer picking based on neural network and image edge extraction technology. Seismic horizon tracing methods in the industry are focused on three-dimensional horizon tracing and automatic section tracing algorithms. There are some defects such as low efficiency, artificial seed points, long training tracking time, and only one layer (plane) trace can be completed at a time, which can not explain the relationship between layers. Therefore, an automatic seismic horizon recognition method based on improved ant colony tracking strategy is proposed. Firstly, the improved weighted median filter is used to preprocess the seismic data, which reduces the artificial participation in the traditional seed point growth algorithm on the basis of enhancing the continuity of the in-phase axis, and based on the basic idea of ant colony algorithm, The support vector machine (SVM) technique is introduced to classify and process seismic data. At the same time, seismic amplitude, instantaneous phase, horizon dip angle, pheromone concentration and other information are considered synthetically, which is included in the evaluation function of ant colony horizon tracing. An improved ant colony search algorithm is used to realize automatic seismic horizon tracing. By dividing the tracing area to realize the layer encryption in the region and provide a variety of reasonable layer encryption strategies to complete the post-processing of seismic horizon tracing. The application results of practical data verify the validity of the method and provide technical ideas for sequence stratigraphic analysis, interpretation of sedimentary system tracts and fine reservoir description.
【作者單位】: 中國(guó)石油大學(xué)(北京)克拉瑪依校區(qū);東營(yíng)市人力資源和社會(huì)保障局信息中心;中國(guó)石油大學(xué)(華東)計(jì)算機(jī)與通信工程學(xué)院;中國(guó)石化勝利油田有限公司物探研究院;
【基金】:國(guó)家重點(diǎn)基礎(chǔ)研究發(fā)展計(jì)劃(“973”計(jì)劃)項(xiàng)目(2013CB228604)資助
【分類號(hào)】:P631.4
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 P.布瓦;邵祝華;;應(yīng)用判別因數(shù)分析法分辨地震層位[J];石油物探譯叢;1981年03期
2 張銀鳳,黃延祜;模糊聚類分析方法在地震層位自動(dòng)對(duì)比中的應(yīng)用[J];石油物探;1989年02期
3 張劍秋,張福炎;地震層位信息三維可視化[J];石油地球物理勘探;1998年01期
4 鄭公營(yíng);曾婷婷;;地震層位自動(dòng)追蹤技術(shù)研究[J];物探化探計(jì)算技術(shù);2013年06期
5 馮智慧;劉財(cái);馮fE;王典;王兆國(guó);;基于互四階累積量一維切片的地震層位自動(dòng)拾取方法[J];石油地球物理勘探;2011年01期
6 謝輝;;基于地震層位趨勢(shì)約束的小層系地質(zhì)層位隨機(jī)模擬[J];科技創(chuàng)新與應(yīng)用;2013年31期
7 Franklin Kemp;趙蓓蓓;;應(yīng)用神經(jīng)網(wǎng)絡(luò)分支和邊界值搜索法的地震層位追蹤系統(tǒng)[J];地質(zhì)科學(xué)譯叢;1993年04期
8 齊志洋;沈偉正;徐振旺;強(qiáng)小軍;楊青;;地震層位曲率屬性在裂縫預(yù)測(cè)中的應(yīng)用研究[J];石油化工應(yīng)用;2014年03期
9 閻賡實(shí);唐小山;;反褶積對(duì)地震層位時(shí)域比對(duì)的影響[J];石油天然氣學(xué)報(bào);2008年02期
10 錢榮毅;唐科學(xué);;利用Java-X3D實(shí)現(xiàn)地震層位三維可視化[J];物探與化探;2005年06期
相關(guān)會(huì)議論文 前1條
1 鄧廷權(quán);吳從p,
本文編號(hào):1989976
本文鏈接:http://sikaile.net/kejilunwen/kuangye/1989976.html