Optimal energy management for a grid connected pumped hydro-battery hybrid storage system supplied by renewable energy

Abstract

South Africa is experiencing serious challenges with limited electricity supply and frequent blackouts, leading to a national crisis. In response, there has been growing interest in renewable energy sources like wind and solar power to help meet the increasing demand for electricity. However, the inconsistent nature of these renewable sources leads to a mismatch between the energy available and what is needed. To solve this issue, it is crucial to integrate different energy storage technologies. Therefore, this study fills a significant gap in the existing literature on hybrid energy storage systems (HESS) as it applies to renewable energy-based microgrids, especially in residential, commercial, and industrial contexts. While much of the current research focuses on electric vehicles, the use of HESS in other areas has not been fully explored. Given South Africa's ongoing electricity crisis, marked by frequent load shedding, aging infrastructure, and rising electricity costs, there is an urgent need for innovative solutions that enhance energy security and reduce expenses. To address this, this research develops an optimal energy management model designed to lower electricity costs for a commercial farm connected to a grid-supported microgrid using a hybrid system of batteries, supercapacitors, and pumped hydro storage. The study includes a thorough review of the current literature, highlighting that while hybrid storage solutions have been well-studied for electric vehicles, their potential in other applications requires further investigation. A control model is developed using a Fuzzy Logic Controller (FLC) to ensure a reliable energy supply and efficient energy exchange within the HESS. This model is validated through simulations in MATLAB Simulink, showing that the FLC can optimize energy flow based on changing load demands and pricing conditions.The key findings reveal that the proposed energy management strategies significantly improve the resilience and reliability of the system, particularly during variable energy generation and high demand periods. The FLC effectively balances the different energy storage components, optimizing their charging and discharging cycles to meet energy needs while keeping costs low. Furthermore, the system's operational data is analyzed using both statistical and machine learning methods through real-time validation on the Opal-RT simulator. This analysis identifies patterns among the various energy sources and demonstrates how weather and demand fluctuations impact system performance. Machine learning techniques were employed to enhance these insights; specifically, anomaly detection captured unusual load patterns, while Principal Component Analysis (PCA) and K-means clustering facilitated the simplification and grouping of system states. Additionally, regression models provided strong predictions of load demand. The economic observations indicate potential savings through reduced dependence on grid electricity, which is imperative amid rising electricity costs in South Africa. Ultimately, this research enhances the understanding of HESS by providing a mathematical framework that promotes the efficient integration of renewable energy systems across different sectors. By tackling the specific challenges faced in South Africa, this work aims to support the transition to a more sustainable and resilient energy future, contributing to broader goals of energy security and economic stability in the country.