A New Method Based on Binary Regression for Time-depth Conversion

A set of buried-hill draping bioclastic dolostone developed in Shahejie Formation in C oilfield of Nanpu depression, Because of the effects of regional geological background, it generally has the following characteristics which create more difficulties in highly efficient development of horizontal wells, such as: complicated reservoir structures, strong lateral velocity variation. As the velocity is of strong lateral variation, it is difficult to accurately calculate the reservoir's structure with traditional method for time-depth conversion. According to the detailed study of time-depth relationship in boreholes, compaction effect and thickness of volcanic rock are confirmed as the two factors affecting the velocity. Based on analysis of traditional method for time-depth conversion. An integrated workflow has been established in this study to improve the prediction accuracy of reservoir depth. According to the detailed study of time-depth relationship in drilled boreholes, compaction effect and thickness of high-speed volcanic rock are confirmed as the two factors affecting the velocity, so we focus on the distribution of the two factors. Initial processing in the time domain was designed to attenuate noise and enhance the signal in the data. This improvement has made it possible to interpret and map above-mentioned two factors in seismic. Elaborate interpretation can depict the reflection time of Ordovician strata which can represent the factor of compaction effect. There existed complex types of volcanic rocks, including volcanic lava and pyroclastic rocks in the area. Based on the mechanism of volcanic eruption and the seismic data, a fact is attained that only the high-speed pyroclastic rocks, namely the effective volcanic rock, account for the lateral velocity variation, from which the relationship between interval velocity of volcanic layer and thickness ofpyroclastic rock is established innovatively to define the second velocity influence factor. On the premise of above effort, which helps us get the distribution of the two factors, we introduce the binary regression to get the average velocity of the reservoir. The average velocity is regarded as the dependent variable of the binary regression while compaction effect and thickness of effective volcanic rock are regarded as the two independent variables. Through the ultra factor of thickness of effective volcanic rock, the new time-depth conversion approach takes full consideration of the lateral variation of velocity, which makes it more suitable for the time-depth conversion of the volcanic-covered reservoir. Application of the field data shows that the depth calculated from the new method matches the actual drilling results very well, and the calculation results could objectively reflect the structural form of the oil-bearing reservoirs.