Analysis Of Influencing Factors Of Thermal And Physical Parameters Of Deep Formation Rock Based On Well Songke II

With the increasing tension of global energy and environmental problems,it is urgent to explore new,green and sustainable energy.Dry hot rock is a new type of geothermal energy,with wide distribution and huge reserves.Dry hot rock is a rock mass with high temperature,low porosity and permeability,and lack of fluid.Due to the lack of fluid,the heat stored in the rock needs to be exploited by artificial fracturing and other technologies to form enhanced geothermal system(EGS).EGS is a very complicated project,and the thermal conductivity of rock is an important aspect.Generally speaking,the occurrence depth of dry hot rock resources is deep,and the general drilling depth can not meet the research of dry hot rock.Songke-2 well is the deepest exploration well in China,with a depth of 7018 m and complete coring.Based on the good geothermal background of Songliao Basin,we study the thermal conductivity of deep formation rock in Songliao basin.Thermal conductivity is an important part of the rock's thermal physical properties,and its size is affected by many factors.Therefore,the thermal conductivity is also anisotropic like other properties of rock.Methods,analyze the reasons for the formation of rock anisotropy and provide the real situation of the anisotropy of the thermal conductivity of rock in the deep formation of Songke 2.Next,the author obtains the porosity,density,wave velocity and mineral composition of rock samples,temperature and pressure by corresponding means,and analyzes the relationship between the thermal conductivity of rock and these factors.It is found that in the depth of 3200m-4500 m,sandstone has a good correlation with porosity and wave velocity,while the mudstone has not.For the in-situ thermal conductivity of rock,the influence of temperature and pressure conditions is also very important.In this study,the empirical formula proposed by the predecessors was used to correct the thermal conductivity of rock under the temperature and pressure.The thermal conductivity is reduced by 30% after temperature correction.For pressure,the overall thermal conductivity after pressure correction tends to increase,but the effect of pressure is much smaller than temperature.Based on the analysis results of influencing factors,the author predicts the in-situ thermal conductivity of the rock through multiple linear regression methods,and calculates the thermal conductivity of the rock skeleton based on the ECS element capture log.Both methods have obvious shortcomings.The rock deposition environment in different regions is different,so the empirical formula in this article may not be applicable to other regions,and the latter's skeleton thermal conductivity calculated by mineral composition differs greatly from the measured thermal conductivity.In view of the shortcomings of the two methods mentioned above,the author puts forward the optimization BP neural network algorithm based on genetic algorithm(GA-BP neural network algorithm)to predict the thermal conductivity by combining the logging data and the existing physical property data,and describes the specific construction process and parameter selection of the neural network algorithm in detail.The comparison between the predicted results and the measured results shows the rationality and applicability of the method.It provides a basis for the prediction of the thermal conductivity of rocks in the deep formation of Songliao basin without core.