XRF data mining algorithm-based metamorphic rock lithology identification method

Abstract

The invention relates to a metamorphic rock lithology identification method based on an XRF data mining algorithm, and the method comprises the following steps: S1, sorting slice identification data according to lithology classification, extracting XRF (element logging) data of corresponding depths of slices, selecting elements Si, Al, Fe, Ca, Mg, Na and K in seven common rock-forming minerals, and extracting the selected elements from the selected elements; carrying out dimensionality reduction treatment on the seven common rock-forming mineral elements by utilizing a principal component analysis (PCA) method, and preferably selecting several elements with relatively high correlation with the lithology as sensitive elements; s3, sensitive elements selected by PCA serve as samples, and a sample file is formed; s4, training the sample file by using a random forest algorithm, and generating a discrimination forest; and S5, processing actual data by using the generated discrimination forest, and outputting a processing result. The method is simple to operate, easy to popularize and apply, and suitable for on-site geological personnel to quickly identify the lithology of the metamorphic rock reservoir. Specific processes are shown in abstract drawings of the specification.

Description

technical field [0001] The invention relates to the field of oil and gas field exploration, in particular to a metamorphic rock lithology identification method based on an XRF data data mining algorithm. Background technique [0002] The buried hill reservoirs are mainly metamorphic rock reservoirs, and the lithology and components of metamorphic rock reservoirs are very complex. The existing conventional lithology identification methods are difficult to subdivide the lithology of metamorphic rocks, and lithology identification is the reservoir. The basis of evaluation is very important for subsequent evaluation and analysis. At present, there are relatively mature methods for the identification of metamorphic rock buried hill entry (that is, buried hill interface clipping), the core of which is mainly based on relevant logging data (such as XRF, Whole rock analysis, etc.) to identify metamorphic rocks and other lithologies (such as mudstone, glutenite, etc.), and the subdiv...

AI Technical Summary

Problems solved by technology

[0002] Buried hill reservoirs are mainly metamorphic rock reservoirs, and the lithology and components of metamorphic rock reservoirs are very complex. The basis of evaluation is very important for the follow-up evaluation and analysis. At present, there are relatively mature methods for the identification of metamorphic rock buried hills (that is, the interface of buried hills). The core is mainly based on relevant mud logging data (such as XRF, Whole-rock analysis, etc.) to identify metamorphic rocks and other lithologies (such as mudstone, glutenite, etc.), and the subdivision of metamorphic rock lithology is mainly based on well logging data. method of gender segmentation

Images