A Data-Driven Robust Optimization Approach to Operational Optimization of Industrial Steam Systems under Uncertainty

This paper proposed a data-driven adaptive robust optimization approach to deal with operational optimization problem of industrial steam systems under uncertainty. Uncertain parameters of the proposed steam turbine model are derived from the semi-empirical model and historical process data. A robust kernel density estimation method is employed to construct the uncertainty sets for modeling these uncertain parameters. The data-driven uncertainty sets are incorporated into a two-stage adaptive robust mixed-integer linear programming (MILP) framework for operational optimization of steam systems. By applying the affine decision rule, the proposed multi-level optimization model is transformed into its robust counterpart, which is a single-level MILP problem. To demonstrate the applicability of the proposed method, the case study of an industrial steam system from a real-world ethylene plant is presented.