(669d) A Continuous-Time MINLP Model for Optimal Source-Sink Matching in Carbon Capture and Storage Systems Under Uncertainty

He, Y., Shanghai Jiao-Tong University
Sahinidis, N., Carnegie Mellon University

Optimal design of source-sink matching in carbon capture and storage systems (CCS) often faces long-term uncertainties and is critically needed to explicitly implement uncertain programming for ensuring constraints are met. In this work, a general continuous-time mixed-integer nonlinear programming (MINLP) with time, injection rate and capacity constraints is developed by using global events and disjunction strategies. In this model, an exact linearization method is adopted to eliminate the products of binary variables and continuous variables, while variable upper bounding constraints are introduced in order to model disjunctions in terms of linear constraints. Then, a robust two-stage stochastic programming (rTSSP) model with conditional value-at-risk as the risk measure is proposed, in which optimization variables are partitioned into first-stage design variables and second-stage operating variables. Finally, a generalized benders decomposition algorithm is developed to efficiently solve the rTSSP problem. Two illustrative case studies are solved to demonstrate the effectiveness of proposed models for planning CCS deployment under uncertainty.