An intelligent energy management for an on-grid hybrid energy system cost reduction

• Published in: Journal of energy storage 2025-02, Vol.110, p.115053, Article 115053
• Main Authors: Riverón-Miranda, Israel, Gómez-González, José Francisco, Méndez-Pérez, Juan Albino
• Format: Article
• Language: English
• Subjects: Bellman–Ford Algorithm; Electric battery; Energy Management System; Energy Storage System; Fuzzy Logic Controller; Hybrid Renewable Energy System; Microgrid; Optimal Power Flow; Supervisory Control
• ISSN: 2352-152X
• DOI: 10.1016/j.est.2024.115053

This paper introduces an innovative Energy Management System (EMS) designed to reduce the cost of energy flows in a Hybrid Renewable Energy System (HRES) connected to a grid with an Energy Storage System (ESS) based on Li-ion batteries. This study considers power purchase and sale tariffs, as well as the variable costs associated with the battery wear contingent on its State of Charge (SoC). The proposed EMS is a supervised controller that combines three Fuzzy Logic Controllers (FLCs) in a cascade to regulate the SoC of the battery via its current, and a setpoint controller based on a Hysteresis-Based Calculation Algorithm (HBC) to establish charge and discharge setpoints and SoC bounds according to the battery wear cost profile. Other EMSs consider battery wear fixed costs or calculate them by counting algorithms, and arbitrary minimum SoCs to extend its Cycle Life (CL). However, these approaches are insufficient to achieve adequate battery management that, in addition to preventing deterioration and premature replacement, analyses the costs due to its usage alongside the other costs of the HRES in order to reduce them. The present work proposes a more advanced EMS that uses a battery usage cost profile based on its Full Equivalent Cycle (FEC) according to the Depth of Discharge (DoD). This approach allows a more realistic calculation of the battery wear cost. Additionally, it establishes more appropriate SoC limits aligned with the energy flow costs across the entire HRES. This perspective adds robustness to the EMS, enabling its use with cost profiles for batteries of different technologies and capacities, as well as varying energy purchase-sale tariffs and power limitations. The introduced approach was found to be more reliable against uncertainties in power-balanced forecasts, and also more computationally cost-effective and convenient for real-time applications compared to existing optimisation techniques. Accordingly, the EMS’s supervised control strategy was evaluated in several simulated scenarios, and its cost was compared to the benchmark optimal solution obtained by the Bellman–Ford Algorithm (BFA). A final innovation presented in this work is a methodology for establishing power balance profiles to simulate test cases of energy surplus and deficit in the HRES, based on battery capacity and maximum power. These test cases aim to assess the suitability of any EMS under extreme conditions and enable its validation for comparison with other strate