Experimental investigation and intelligent modeling of pore structure changes in type III kerogen-rich shale artificially matured by hydrous and anhydrous pyrolysis

ENERGY(2023)

引用 1|浏览3
暂无评分
摘要
The occurrence and enrichment of shale plays are highly controlled by pore characteristics of the formation. In this study, an immature sample rich in kerogen type III from the Damoguaihe formation, Hailar Basin in China is subjected to hydrous and anhydrous pyrolysis (HP and AHP) across an extensive temperature range (300-450 degrees C). Next, low-pressure N2 adsorption analysis was conducted on the pyrolyzate of each maturity stage to study the pore structures' evolution during thermal maturity simulation. Moreover, deconvolution and fractal dimension analyses were implemented to study pore families and the complexity of pores within the shale samples. Finally, radial basis function (RBF) neural networks optimized by four evolutionary approaches were applied for modeling N2 adsorption data obtained from the HP and AHP samples. According to the results, the original shale sample and all pyrolyzates obtained with HP and AHP scenarios exhibited type IV isotherm with H3 hysteresis loops. As a whole, BET surface area, micro-, meso- and total pore volume of HP pyrolyzates were higher than AHP ones. The unheated shale sample had seven families containing three mesopore and four macropore groups. Although all pyrolyzates obtained from the pyrolysis had families with similar means, their pore volumes were entirely different, which proves that the pore structure of samples undergoes changes during thermal maturation and the presence of water can also enforce these changes. Both fractal dimensions showed a direct relationship with BET surface area and a negative correlation with the average pore diameter of shale samples. The RBF model optimized by differential evolution (DE) delivered a mean absolute percent relative error (MAPRE) value of 4.84% and determination coefficient (R2) of 0.9946 for the total data set, which outperforms other RBF models in predicting N2 adsorption/desorption tests of pyrolyzates. The outcome of sensitivity analysis suggested that the N2 adsorption/desorption behavior of the pyrolyzates was mostly affected by relative pressure and the pyrolysis type (HP or AHP). Ultimately, the results clearly revealed that the effect of water on the pores' alteration of shales with type III kerogen is greater than the effect of temperature or thermal maturity itself.
更多
查看译文
关键词
Damoguaihe formation,Anhydrous and hydrous pyrolysis,N2 adsorption analysis,Deconvolution analysis,Fractal dimension,Artificial neural networks
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要