Waveform Indication Simulation of Rock Mineral Vomposition Prediction and Unsupervised Neural Network Clustering Drillability Analysis
摘要
Rock drillability is a comprehensive response of rock hardness, elastic–plastic properties, abrasiveness, and rock structure. Essentially, it is closely related to the physical and mechanical properties, mineral composition, and content of rocks. Cluster analysis of rock mineral composition and content can be used to predict the drillability of stratums. Seismic waveform indication inversion (or simulation) has the significance of waveform phase control, which is more scientific and advanced than geostatistical inversion. Based on three-dimensional seismic data, a waveform indication simulation algorithm is used to predict the three-dimensional spatial distribution of rock mineral composition curves. Using the combination of well and seismic and waveform indication simulation to obtain rock mineral components, unsupervised neural network clustering analysis is used to divide the strata into three categories. Compared with the drilling speed of the drilling machinery, semi quantitative prediction is made for the three-dimensional spatial distribution of difficult to drill strata, sub difficult to drill strata, and easy to drill strata. Extract drillability profiles along the trajectory to be drilled, guide drill bit selection, and determine construction parameters. Through practical examples, it has been verified that the approach proposed in this article is feasible and has high accuracy.