Dynamic Optimization of Energy Hubs with Evolutionary Algorithms Using Adaptive Time Segments and Varying Resolution

Abstract

Many Renewable Energy Sources (RESs) need to be installed and integrated into the existing grid infrastructure for further developing the energy system towards a 100% RESs supply. To provide the required flexibility for actively compensating local volatile power fluctuations caused e.g. by RES, scheduling of controllable Distributed Energy Resources (DERs) using the concept of Energy Hubs (EHs) and their integration into power networks is a promising approach. The complexity of optimized operation of such EHs while simultaneously providing the required flexibility can be faced by using Evolutionary Algorithms (EAs) for calculating the schedules of internal components of the EHs. To focus the computational effort of the used EA on more important time segments of the problem space, a dynamic optimization method with adaptive time intervals and time resolution is presented. The method improves the quality of the optimization in form of better approximation to a given target value an EH can follow. The proposed approach leads to an average reduction of 11% in the Root Mean Squared Error (RMSE) between target value and measurement by appropriate increased calculation time.