Sequential Decisions under Uncertainty in Coupled Multi-Energy Systems

Smart energy systems provide a foundation for a sustainable andB efficient future society. By considering the integrated operation of multiple energy carriers, these systems can enhance their efficiency and ability to adjust to uncertainties. However, making informed decisions in energy systems with complex operational interdependencies can be challenging. This paper considers a coupled multi-energy system, including electricity, heat, gas, and hydrogen, where operational decisions must be made sequentially (close to real-time) and under uncertainty. The solution concept takes the form of a policy, i.e., a method for making operational decisions in the system as new information about the uncertainty becomes available. To approximate optimal and accountable decisions, the paper presents a stochastic look-ahead policy where the system’s uncertainty is simulated through a probabilistic forecast and captured and clustered into a finite set of scenarios that can be used in a multi-stage stochastic program. The policy’s performance is evaluated empirically on a realistic case-study system.