Determining the degree of maturation and thermal evolution in kerogen is vital to the hydrocarbons industry. This study evaluates the thermal maturation of source rock sections in sedimentary basins. Its determination makes it possible to estimate whether a sample from a given depth in an oil well falls within the window of oil or natural gas generation or remains immature. The Spore Coloration Index (SCI) can indicate the maturation of oil- and gas-prone material. The objective of this study is to automate this analysis, thereby achieving significant improvements in time efficiency and reliability. Traditionally, an operator visually identifies the presence of sporomorphs in slides using an optical microscope. When they encounter a sporomorph, the operator visually compares it with standard reference slides. The standard consists of 19 slides of spores of different degrees of maturity, ranging from the lightest to the darkest, from 1.0 to 10.0, in increments of 0.5. Comparison with the standard allows the operator to estimate the Spore Coloration Index (SCI) of a sporomorph. Various tests corroborated the literature's indication that the red channel best correlates with the thermal maturation index. A linear relationship was obtained between the average intensity of the red channel and the SCI (correlation coefficient R2 = 0.97). Images of samples from various wells at different depths were acquired under conditions similar to those standardized for organopalynological slides. Deep Learning-based systems were trained to identify sporomorphs in the images. With the help of an expert operator, the objects were manually outlined in the images to create a reference database. This database was separated into a training set and a validation set, allowing the network to learn and then have its performance evaluated (accuracy ≈ 86 %). The captured images feed the system, which identifies the presence of sporomorphs, measures the SCI value of each one, and generates a histogram of SCI distribution for each case, allowing, for the first time, the direct calculation of the SCI without the need for visual analysis.
扫码关注我们
求助内容:
应助结果提醒方式:
