Sensitivity analysis of linear programming economic dispatch models

Abstract:
Linear programming (LP) is widely used for the economic dispatch of energy systems. Due to the increasing inhomogeneity of supply and storage technologies, the consideration of sector coupling and increasing temporal resolutions, dispatch models are becoming more and more complex. In order to robustly interpret the optimization results, sensitivity analyses are thus ever more important. The sensitivity of the objective function w.r.t. changes of boundary conditions can easily be calculated using dual variables of the LP model, as is, e.g., typically done for deriving locational marginal prices. In a recent work, we have proposed a novel method to the calculate sensitivities of all system variables w.r.t. changes of the constraints. Here, we apply this approach to economic dispatch and show how to present the changed dispatch decisions intuitively for multi-modal energy systems. Lastly, we show how to compute each sensitivity’s interval of validity.