老油田的勘探已经进入到以薄层为主的隐蔽油藏勘探阶段,厚度作为储层的主要特征参数,反映了物性、物源方向、沉积环境、沉积背景等。因此,储层厚度的精确求取关系到油藏勘探的成败。有关薄储层厚度的计算已有不少成熟的方法,如反射振幅法、反演法等,这些方法因适应性低或过程复杂,没有形成一种完全有效的厚度求取方法。近几年发展起来的频谱分解技术,是一种无井约束的高分辨率储层预测技术,也是储层厚度求取方法的补充和完善。频谱分解技术根据薄层调谐原理,利用全频段的信息,将地震数据从时间域转换到频率域,突出高频信号,相对提高地震资料的主频,从而提高了对薄储层厚度的识别能力。从基于离散傅里叶变换和小波变换两种原理的频谱分解技术出发,对两种方法处理的结果进行分析,并基于处理得到的不同参数,如调谐频率、最大能量等,提出了三种求取储层厚度的方法,并在实际应用中取得了较好效果。 The exploration of old oilfields has entered the subtle reservoir exploration stage mainly with thin layers. As the main characteristic parameter of reservoir, thickness reflects the physical properties, provenance direction, sedimentary environment, sedimentary background and so on. Therefore, the accurate calculation of reservoir thickness is related to the success of reservoir exploration. There are many mature methods for calculating the thickness of thin reservoirs, such as the reflection amplitude method and the inversion method. These methods have not formed a completely effective method for determining the thickness because of their low adaptability or complicated process. The spectral decomposition technique developed in recent years is a well-constrained, high-resolution reservoir prediction technique, which is also a complement and improvement to the reservoir thickness calculation method. According to the principle of thin-layer tuning, the spectral decomposition technique uses the information of the whole frequency band to transform the seismic data from the time domain to the frequency domain, highlighting the high-frequency signals and relatively increasing the dominant frequency of the seismic data, so as to improve the identification of thin reservoir thicknesses . Based on the spectral decomposition techniques based on the two principles of Discrete Fourier Transform and Wavelet Transform, the results of the two methods are analyzed. Based on the different parameters processed, such as tuning frequency and maximum energy, three methods are proposed Take the method of reservoir thickness, and achieved good results in practical application.