基于粒子群優(yōu)化最小二乘支持向量機(jī)的非線性AVO反演
發(fā)布時(shí)間：2019-02-19 21:35
【摘要】：為了求解非線性AVO反演問題,本文提出基于粒子群算法和最小二乘支持向量機(jī)的非線性AVO反演方法,并用粒子群算法優(yōu)化最小二乘支持向量機(jī)的參數(shù)。即首先通過精確Zoeppritz方程正演得到角道集,并進(jìn)行動(dòng)校正和部分角度疊加;然后運(yùn)用最小二乘支持向量機(jī)方法建立反射振幅與彈性參數(shù)之間的非線性模型;最后以此非線性模型對地震道集數(shù)據(jù)進(jìn)行反演。模型數(shù)據(jù)和實(shí)際資料的反演結(jié)果表明,該方法克服了常規(guī)廣義線性AVO反演在遠(yuǎn)炮檢距及彈性參數(shù)縱向變化大等情況下的缺陷,可直接從實(shí)際地震道集數(shù)據(jù)中提取較高精度的地層彈性參數(shù),具有快速穩(wěn)健、抗噪能力強(qiáng)的優(yōu)點(diǎn)。
[Abstract]:In order to solve the nonlinear AVO inversion problem, a nonlinear AVO inversion method based on particle swarm optimization (PSO) and least square support vector machine (LS-SVM) is proposed in this paper. The parameters of LS-SVM are optimized by particle swarm optimization (PSO). First, the angular gathers are obtained by accurate Zoeppritz equation, then the nonlinear model between reflection amplitude and elastic parameters is established by using least square support vector machine (LS-SVM), and the NMO and partial angle superposition are carried out. Finally, the nonlinear model is used for inversion of seismic gather data. The inversion results of the model data and the actual data show that the method overcomes the defects of the conventional generalized linear AVO inversion under the condition of large longitudinal variation of the far offset and elastic parameters. The formation elastic parameters can be extracted directly from the actual seismic gather data, which has the advantages of fast stability and strong anti-noise ability.
【作者單位】： 中國地質(zhì)大學(xué)(北京)地球物理與信息技術(shù)學(xué)院;中國石油華北油田公司勘探開發(fā)研究院;中國石化勘探分公司研究院;
【基金】：國家高技術(shù)研究發(fā)展計(jì)劃項(xiàng)目(2013AA064201) 國家“十二五”油氣田及煤層氣科技重大專項(xiàng)專題(2011ZX05033-004)聯(lián)合資助
【分類號(hào)】：P631.4
[Abstract]:In order to solve the nonlinear AVO inversion problem, a nonlinear AVO inversion method based on particle swarm optimization (PSO) and least square support vector machine (LS-SVM) is proposed in this paper. The parameters of LS-SVM are optimized by particle swarm optimization (PSO). First, the angular gathers are obtained by accurate Zoeppritz equation, then the nonlinear model between reflection amplitude and elastic parameters is established by using least square support vector machine (LS-SVM), and the NMO and partial angle superposition are carried out. Finally, the nonlinear model is used for inversion of seismic gather data. The inversion results of the model data and the actual data show that the method overcomes the defects of the conventional generalized linear AVO inversion under the condition of large longitudinal variation of the far offset and elastic parameters. The formation elastic parameters can be extracted directly from the actual seismic gather data, which has the advantages of fast stability and strong anti-noise ability.
【作者單位】： 中國地質(zhì)大學(xué)(北京)地球物理與信息技術(shù)學(xué)院;中國石油華北油田公司勘探開發(fā)研究院;中國石化勘探分公司研究院;
【基金】：國家高技術(shù)研究發(fā)展計(jì)劃項(xiàng)目(2013AA064201) 國家“十二五”油氣田及煤層氣科技重大專項(xiàng)專題(2011ZX05033-004)聯(lián)合資助
【分類號(hào)】：P631.4
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 傅文敏;AVO模擬分析泥盆系礁體勘探的困難和可能性[J];重慶石油高等�？茖W(xué)校學(xué)報(bào);1997年01期
2 朱兆林,王永剛,曹丹平;裂縫性儲(chǔ)層AVO檢測方法綜述[J];勘探地球物理進(jìn)展;2004年02期
3 胡曉麗;譚大龍;;孔隙形狀對AVO響應(yīng)影響的研究[J];巖性油氣藏;2010年03期
4 李中元;北極地區(qū)天然氣水合物的AVO屬性研究[J];石油物探;2004年S1期
5 何濤;史，
本文編號(hào)：2426899
本文鏈接：http://sikaile.net/kejilunwen/diqiudizhi/2426899.html
最近更新
教材專著
