Multi-objective optimization of a novel control algorithm and scheduling procedure for optimal use of energy storage systems

Abstract

Peak shaving with energy storage systems (ESSs) is a promising approach to optimize energy use, reduce costs, and ensure a more reliable power grid.

This paper aims to improve the performance of a novel control algorithm for efficient peak shaving by using sensitivity analysis and multi-objective optimization techniques. Regarding this, 2 hypothesis load demand profiles as well as 5 different scenarios with diverse objective functions, including ESS capacity, standby days of the ESS, etc., were considered for multi-objective optimization, and the control algorithm was applied to them. These scenarios were designed to explore the algorithm's adaptability to different operating conditions and to evaluate its effectiveness across varying system constraints. The Pareto front of each was extracted and the results of each were detailed. Among the most important obtained results, it can be mentioned that the decrease of 58.29% and 51.32% of ESS standby days in load profiles A and B, respectively, compared to basic conditions. Also, it has been possible to reduce peak demands to 16.29% and 19.66%, respectively, compared to the maximum value of profiles A and B.

To enhance the efficiency and gain more precise control over the energy storage system's charging and discharging rates, incorporating upper limits for charging and lower limits for discharging were proposed. This modification requires minimal changes to the existing algorithm. Future research could concentrate on integrating direct regulation of the charging and discharging rates of the ESS by establishing upper and lower bounds for these rates as decision variables within the optimization framework. This approach would more accurately represent the operational constraints of the ESS, thereby improving the model’s applicability and scalability for real-world implementations.