Method for calculating crude oil density by utilizing geochemical pyrolysis spectrogram and machine learning

Abstract

The invention provides a method for calculating crude oil density by utilizing a geochemical pyrolysis spectrogram and machine learning, which comprises the following specific steps of: standardizing the spectrogram according to the geochemical logging pyrolysis spectrogram; extracting the characteristic parameters of crude oil in the spectrogram from the standardized spectrogram, wherein the characteristic parameters comprise area sizes corresponding to different time periods, the sample degradation degree calculated by using the peak connecting line area and the valley bottom connecting line area of the spectrogram, and the occurrence time of the maximum peak; establishing a training sample set according to the extracted geochemical pyrolysis spectrograms under different densities; training the sample set through a machine learning model, and calculating the crude oil density through the extracted spectrogram features. According to the method, the characteristics of the geochemical pyrolysis spectrogram are extracted, the machine learning model is established according to the characteristic sample set of the geochemical pyrolysis spectrogram, and then the crude oil sample density is calculated through the geochemical pyrolysis spectrogram.

Description

technical field [0001] The invention belongs to the technical field of petroleum exploration, and in particular relates to a method for calculating the density of crude oil by using geochemical pyrolysis spectrum and machine learning. Background technique [0002] By improving the recognition rate of oil and gas layers in the mud logging stage, timely identify oil and gas layers when the drill bit is drilling, and provide technical support for subsequent exploration and construction operations by judging oil quality types, thereby improving the efficiency of exploration and construction operations. [0003] Using mud logging technology to identify and judge oil and gas layers has always been a technical problem in mud logging operations. Rapid judgment of reservoir crude oil properties, interpretation of oil, gas and water in mud logging, cable pressure measurement and sampling, selection of formation testing tools and processes, and accumulation models Research plays an imp...

AI Technical Summary

Problems solved by technology

The chromatographic data with high precision and little influence of the external environment and its derived parameters have not been fully utilized

[0007] 3. When the current method interprets and analyzes the geochemical pyrolysis spectrum, it does not fully mine the spectrum information, only does qualitative analysis, does not perform quantitative analysis, and does not have high utilization of spectrum sample data

Images