Abstract
Lithium has become a valuable resource for a variety of applications, such as for batteries, in the quest for greater sustainability of the utilization of resources. These challenges have led to the demand for enhanced field operations and making the operations more sustainable while maintaining production levels. The 4th Industrial Revolution is having enormous impact on the oil and gas industry, and with the rise of the demand for lithium for battery and energy technologies, hydrocarbon reservoirs have become an attractive source for these valuable resources. Several reservoirs around the world have been demonstrated to contain significant quantities of lithium in their produced brines.
In this work, we present a new AI optimization approach for the optimization of lithium recovery from reservoir operations while maintaining reservoir oil production targets. The approach is novel in the both the AI framework utilized as well as its integration into a global optimization approach based on genetic algorithm to optimize both oil production and lithium recovery. The AI framework incorporates a deep learning LSTM algorithm for estimating oil, brine and lithium recovery based water injection levels. The deep learning model is then incorporated into a global optimization framework to optimize the water injection levels to maximize lithium recovery while maintaining reservoir oil production levels.
We provided a new AI optimization framework for lithium recovery from reservoir brine from an oil and gas reservoir. The framework enables optimization of lithium recovery while maintaining hydrocarbon recovery rates. The framework was successfully demonstrated on the Volve field, outlining the potential significant increase in lithium recovery rates from an optimized injection process. This framework may provide significant opportunities in enhancing reservoir brine utilization, and contributed subsequently to enhancing sustainability in reservoir operations.
