论文部分内容阅读
在W地区,储层孔隙度预测技术是将支持向量机方法与分频法相结合构成一种新的地震非线性预测技术。支持向量机是一种基于统计理论的新型机器学习算法,它是建立输入与输出的一个隐形的映射关系,在我们的问题中,应用测井数据作为支持向量机的输入向量,通过支持向量机优选参数,对实际样本数据构造核函数,实现全局最优解。分频法采用短时快速傅利叶变换(SFFT)将地震数据进行时频转换,形成高分辨的分频数据体,作为独立的输入数据体,以提高储层孔隙度反演的分辨率和精度。因此,构成的这种联合预测技术,充分发挥了支持向量机和分频法的优势,这种全新的储层孔隙度预测技术适应储层地质参数高维非线性特点。依据建立的储层孔隙孔隙度预测技术,利用W地区实际地震数据形成的分频数据体和测井数据,对朱海组储层孔隙度进行了预测,获得了高分辨率和高精度储层孔隙度剖面和储层孔隙度数据体。在W地区,预测的朱海组储层孔隙度与测井孔隙度具有良好的一致性,其相关系数值为(0.816～0.945),平均值为0.882,因此,预测的储层孔隙度具有较高的可信度和准确率。在W地区成功应用表明,其方法技术具有通用性和较好的推广性,它的适用面广,预测效果好,易于实现,它将油气储层孔隙度预测推向一个新的发展水平。
In the W area, reservoir porosity prediction technology combines support vector machine method with frequency division method to form a new nonlinear seismic prediction technology. Support vector machine (SVM) is a new type of machine learning algorithm based on statistical theory. It establishes an invisible mapping relationship between input and output. In our problem, we use well logging data as the input vector to support vector machine (SVM) Optimum parameters, the actual sample data to construct the kernel function to achieve the global optimal solution. The crossover method uses short-time fast Fourier transform (SFFT) to convert the seismic data to time-frequency, forming a high-resolution frequency-divided data body as an independent input data body to improve the resolution and accuracy of reservoir porosity inversion. Therefore, the proposed joint forecasting technique fully takes advantage of SVM and frequency-division method. This new reservoir porosity prediction technology adapts to the high-dimensional nonlinear characteristics of reservoir geological parameters. Based on the established porosity prediction technology for reservoir porosity, the porosity of the reservoir in Zhuhai Formation is predicted by using the frequency-divided data body and log data from the actual seismic data in W region. The reservoir with high resolution and high precision is obtained Porosity Profile and Reservoir Porosity Data Body. In the W area, the predicted reservoir porosity of the Zhuhai Formation is well consistent with the log porosity with a correlation coefficient of (0.816-0.945) with an average of 0.882. Therefore, the predicted reservoir porosity is more High credibility and accuracy. The successful application in the W region shows that the method has the advantages of universality and good popularization. It has wide applicability, good prediction effect and easy realization. It will push the prediction of oil and gas reservoir porosity to a new level.