Integrated modeling of multi-scale transport in coal and its application for coalbed methane recovery

Exploitation of coalbed methane (CBM) acts as a synergy between increased natural gas demand and reduced greenhouse gas emissions to the atmosphere. However, substantial CBM resources lie undeveloped because of unfavorable economic conditions. Accurate forecast of production based on a thorough understanding of physical mechanisms is the key to increase and sustain CBM recovery. CBM production involves complex multi-scale phenomena ranging from desorption and diffusion in nanoporous matrix at molecular level to Darcy flow of free gas in cleats at macroscopic level. Current computational modeling built on fracture flow and neglecting multi-scale coupled flow phenomena underestimates long-term production performance. In fact, experimental evidence indicated that matrix experienced a much greater increase in gas deliverability than cleats. In this work, apparent matrix permeability was derived to characterize diffusion rate and directly coupled into current modeling to decipher matrix deliverability and multi-scale flow. The proposed modeling workflow was applied to two field cases in San Juan Fairway with long production history of over 20 years, which assembled adequate forecast to field data. Besides, sensitivity analysis suggested that simulations neglecting matrix flow and solely updating fracture flow were prone to elevated prediction errors for long-term production when diffusive mass flux took the predominant role in overall gas transport. The multi-scale CBM model is convenient for elucidating complex gas transport physics in coal. The outcome of this work implies that a proper diffusion enhancement method can improve overall production rate and elongate total productive lifetime of CBM fields.

Automated summary

The exploitation of coalbed methane plays a crucial role in meeting natural gas demand and reducing greenhouse gas emissions, but many resources remain undeveloped. Accurate production forecasting is essential to increase and sustain CBM recovery. Current computational models underestimate long-term production performance, while a multi-scale CBM model can improve production rates and extend productive lifetimes.