A Novel Method for the Inversion of the Virtual Well-log Interval Transit Time
By analyzing the relation between well-log data and seismic data,a novel method for predicting the virtual well-log interval transit time based on particle swarm optimization and support vector machine (PSO-SVM) is proposed.A prediction model for the virtual well-log interval transit time is established using the data of seismic of the un-drilled well,seismic and well-log of the drilled well by training the SVM,which is optimized by PSO algorithm.The proposed method is applied to the well YONG of Junggar Basin and the experimental results show it has higher prediction accuracy ,faster convergence speed and better generalization than BP neural network approach.Obtaining the geological characteristic parameters mainly depends on the seismic and the well-log information1,2 at present.For the un-drilled well,it is difficult to predict various of geological parameters as only the seismic layer velocity is available.As the seismic and well-log data are both the comprehensive response of the same geologic body and have the strong correlation,the mapping relation between them of the drilled well can be used to reconstruct the virtual well-log interval transit time of the un-drilled well.Since the traditional method such as the regression method cant solve the nonlinear mapping problem well,a novel method for the inversion of virtual well-log interval transit time is proposed using the data of seismic of the un-drilled well,seismic and well-log of the drilled well by the PSO-SVM algorithm.
Hai Ma Yanjiang Wang
College of Information and Control Engineering,China University of Petroleum,Dongying,Shandong 257061,P.R.China
国际会议
9th International Conference on Signal Processing(第九届国际信号处理学术会议)(ICSP08)
北京
英文
2008-10-26（万方平台首次上网日期，不代表论文的发表时间）