Optimization of well placement, CO2 injection rates, and brine cycling for geological carbon sequestration

In geological carbon sequestration, CO2 can be trapped through several mechanisms. Some forms of trapped CO2 may be more susceptible to leakage and are thus less desirable. Computational optimization can be used to provide modes of operation that maximize preferable trapping mechanisms. In this work we apply optimization procedures to minimize the mobile fraction of CO2 at the end of a 1000-year injection and storage operation in a brine aquifer. A noninvasive gradient-free direct search technique is used for the optimizations. Examples involving only CO2 injection, as well as cases that also include brine cycling, are considered. Brine cycling entails the periodic production of brine from the bottom of the aquifer along with the coincident re-injection of brine into the top of the aquifer. Optimization variables include well locations and injection rates for CO2 injection wells, and the timing and injection volumes for brine cycling. Results demonstrate that optimization can significantly reduce the mobile CO2 fraction compared with base case scenarios. We also show that mobile CO2 fraction decreases monotonically with increasing volumes of brine cycling. The resulting plot of minimum mobile CO2 fraction versus pore volume of brine injected provides the Pareto front for a bi-objective optimization involving these two variables as objectives. The impact of grid resolution on the optimization results is also considered. (c) 2012 Elsevier Ltd. All rights reserved.