Scholarlink

<p>The WZ6-1 structure in the Beibu Gulf Basin of the South China Sea is broken, and there are many fault blocks. The degree of fault sealing is difficult to determine, and the law of oil and gas enrichment is difficult to understand. The conventional single seismic attribute method is ineffective in predicting the petroliferous area of the structure, and the number of dry wells is large. Through the comprehensive study of drilling and seismic data, six types of seismic attributes along the layer are selected, and the prediction method of the petroliferous area using multi-seismic attributes nonlinear fusion PCA-SOM model is proposed. The PCA-SOM model is used to predict the petroliferous area of the WZ6-1 structure. The results show that: (1) The SOM neural network can not only synthesize more seismic attribute variable information of oil and gas bearing formation but also effectively mine the statistical characteristics of nonlinear response between seismic attributes and oil and gas bearing formation. Moreover, it is not necessary to determine the weight value of multiple seismic attribute variables, to reduce the ambiguity of prediction results and improve the reliability of prediction. (2) The principal component after principal component analysis (PCA) of seismic attribute variable data is used as the input parameter of the SOM neural network, which weakens the correlation redundancy between multi-seismic attribute variable data, affects the weight vector of SOM neural network and the range of superior neighborhood, and equivalently amplifies the oil and gas information contained in seismic attribute variables. (3) The prediction of the petroliferous area in the WZ6-1 structure has achieved good results, which not only confirms that the prediction results are consistent with the geological conditions revealed by actual drilling, but also consistent with the law of oil and gas geological accumulation. The prediction method of the petroliferous area by using the PCA-SOM model not only provides a new way to improve the prediction reliability of petroliferous area in complex fault block structure areas by using seismic attribute data but also provides a reference for the exploration and development of similar complex fault block structure area.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of the petroliferous area in WZ6-1 structure of Beibu Gulf Basin in the South China Sea based on PCA-SOM model

  • 期刊论文

  • 发表日期：

    2025年2月14日

查看全文
  • Zhilong Chen,
  • Renyi Wang,
  • Biao Xu,
  • Jianghang Zhu

摘要

The WZ6-1 structure in the Beibu Gulf Basin of the South China Sea is broken, and there are many fault blocks. The degree of fault sealing is difficult to determine, and the law of oil and gas enrichment is difficult to understand. The conventional single seismic attribute method is ineffective in predicting the petroliferous area of the structure, and the number of dry wells is large. Through the comprehensive study of drilling and seismic data, six types of seismic attributes along the layer are selected, and the prediction method of the petroliferous area using multi-seismic attributes nonlinear fusion PCA-SOM model is proposed. The PCA-SOM model is used to predict the petroliferous area of the WZ6-1 structure. The results show that: (1) The SOM neural network can not only synthesize more seismic attribute variable information of oil and gas bearing formation but also effectively mine the statistical characteristics of nonlinear response between seismic attributes and oil and gas bearing formation. Moreover, it is not necessary to determine the weight value of multiple seismic attribute variables, to reduce the ambiguity of prediction results and improve the reliability of prediction. (2) The principal component after principal component analysis (PCA) of seismic attribute variable data is used as the input parameter of the SOM neural network, which weakens the correlation redundancy between multi-seismic attribute variable data, affects the weight vector of SOM neural network and the range of superior neighborhood, and equivalently amplifies the oil and gas information contained in seismic attribute variables. (3) The prediction of the petroliferous area in the WZ6-1 structure has achieved good results, which not only confirms that the prediction results are consistent with the geological conditions revealed by actual drilling, but also consistent with the law of oil and gas geological accumulation. The prediction method of the petroliferous area by using the PCA-SOM model not only provides a new way to improve the prediction reliability of petroliferous area in complex fault block structure areas by using seismic attribute data but also provides a reference for the exploration and development of similar complex fault block structure area.