Optimal Operation of Transient Gas Transport Networks

Abstract In this paper, we describe an algorithmic framework for the optimal operation of transient gas transport networks consisting a hierarchical MILP formulation together with sequential linear programming inspired post-processing routine. Its implementation is part KOMPASS decision support system, which currently used in industrial setting. Real-world are controlled by operating complex pipeline intersection areas, comprise multiple compressor units, regulators, and valves. following, introduce concept network stations to model them. Thereby, represent technical capabilities station hand-tailored artificial arcs add them network. Furthermore, choose from predefined set flow directions each time step, determines where enters leaves station. Additionally, have select supported simple state , consists two subsets arcs: Arcs that must cannot be used. The goal determine stable control satisfying all supplies demands. intersections, represented stations, were initially built centuries ago. Subsequently, due updates, changes, extensions, they evolved into highly involved topologies. To extract their basic properties using computer-readable optimizable descriptions took several years effort. dispatchers controlling network, need compute continuously updated list recommended measures. Our motivation presented here make fast decisions on important global parameters, i.e., how route compress gas. Detailed continuous discrete measures realizing them, take hardware details account, determined subsequent step. present computational results project detailed real-world data.