Method And Application Of Heterogeneity Detection For Coalbed Methane Reservoir

Coalbed methane(CBM)is a green,clean and unconventional oil and gas resource.According to the current exploration results,China's CBM reserves are huge,and CBM will play an important role in China's future energy supply.The results of previous studies indicate CBM reservoir generally have high heterogeneity characteristics.The heterogeneity of CBM reservoir strongly controls CBM target area selection,infill well deployment,and production capacity of CBM wells.Therefore,it has great significance to reveal the heterogeneity characteristics of CBM.Higher-order statistics is a mathematical tool that describes high-order(second-order and higher)statistical features of stochastic processes.Higher-order statistics of seismic data can provide detailed information on horizontal orientation changes in reservoir.Firstly,this thesis carries out numerical simulation of seismic,analyzes the characteristics of the corresponding high-order statistics parameters(skewness and kurtosis)of different models,and explores the fitting methods of experimental variogram.Then calculated the high-order statistical parameters of the coal seam No.10 in the target layer and evaluated its heterogeneity.Combined with logging data,it was obtained that the comprehensive parameter I0 of reservoir heterogeneity in the coal seam No.10.Finally,a variety of seismic attributes were extracted and screened.The improved neural network algorithm was used to establish a prediction model for the evaluation parameters of seismic attributes and heterogeneity,and the I0 plane distribution of the coal seam No.10 was obtained,thereby implementing the heterogeneity of the reservoir.This thesis mainly achieved the following research results.(1)Changes of fracture development density and filling velocity rate of coalbed methane reservoir will change the skewness and kurtosis parameters.In the area where structural faults developed,the skewness and kurtosis parameters all showed low values,indicating that the CBM reservoir was highly heterogeneous.In the flattened area,the skewness and kurtosis parameters appeared to be higher values,where the CBM reservoir was less heterogeneous.Most of the high gas producing wells are located close to the fracture development area,and the kurtosis and skewness values are medium-high values.The low gas production wells were mostly located in the flattened area and were far away from the fracture development area,which have higher skewness and kurtosis value;the dry wells were located near the structural faults with low skewness and kurtosis values.(2)A new variogram fitting method based on improved quantum particle swarm optimization algorithm is proposed.By comparing with the method of weighted least squares fitting,the results show that this method has a better effect in fitting the experimental variogram.(3)The I0 values in most of the coal seam No.10 are low-medium,which indicates that most of the coal seam No.10 show low to moderate heterogeneity.In areas where tectonic fractures are developed,the I0 values are all high,indicating that these regions have strong heterogeneity.In the area far from the structural fault,the I0 value gradually decreases,and the heterogeneity gradually decreases.Based on the actual gas production conditions of the drilling wells,most of the high gas production wells are located in the low-medium heterogeneity region close to the faults,and the low gas production wells are mostly distributed in the regions where the heterogeneity is weak.The dry wells are mostly distributed in the structural area where the fault develops.(4)CBM reservoir is characterized by strong heterogeneity and need to be comprehensively tested by a variety of methods to guide the layout of wells and increase the productivity of coalbed methane.