Data-Analytics Based Investigation of Controlling Parameters of CO2 Sequestration in Depleted Unconventional Reservoirs

The Paris Agreement on climate change in 2015 changed the entire landscape of the oil and gas industry. Fortunately, the oil and gas industry is the only industry in the world with the technology and experience of dealing with enormous volumes of fluid, such as CO2 and storing these volumes in deep underground as a means to mitigate climate change. Several organizations have used laboratory experiments and numerical reservoir simulators to model and discern the behavior of CO2 sequestration. Nonetheless, these methods have high computational cost. Besides, studies have been done on the use of data-driven statistical techniques, these studies mainly focused on production optimization in unconventional reservoirs. In this study, a data-analytics based investigation was carried out to develop insights and analyze the primary variables that affect CO2 sequestration process in unconventional reservoirs. This study revealed a relationship between reservoir, operational parameters, and the performance metric. At the same time, accurately predicting the volume of CO2 sequestered using statistical and machine-learning algorithms. Interestingly, the most influential parameters are the ones affecting the stimulated reservoir volume. According to our results, operational parameters are more dominant than reservoir parameters in driving high performance and stimulated reservoir volume fracture permeability is the most important parameter in order to get high-performance. Our findings will aid in designing these sequestration projects sustainably.