Efficient Dispatch in Cross-Border Balancing Platforms: Elastic Demand Through Parametric Cost Function Approximation

We propose a method for rationalizing the determination of elastic demand curves for MARI, the mFRR cross-border balancing paltform, in the integrated European balancing process. Elastic demand curve for mFRR balancing energy has been proposed due to their ability to reduce the overall balancing cost compared to single-point inelastic demand. The problem is modelled as a non-convex stochastic program, with the variables of interest being the parametrization of the demand curve used by the system operator for acquiring mFRR. The structure of the model allows us to derive its gradient with respect to the demand curve parameters. The problem is solved by applying a randomized stochastic gradient scheme. A price parametrization is shown to outperform a quantity parametrization.