Researchers Simon Halvdansson, Lucas Ferreira Bernardino, and Brage Rugstad Knudsen from the Technical University of Denmark have developed a new framework to optimize the sizing of isolated hybrid renewable energy systems. Their work, published in the journal Applied Energy, addresses the challenge of integrating intermittent renewable energy sources like solar and wind into off-grid systems, which often rely on fossil fuel-based power sources for stability.
The researchers’ framework aims to balance the need for renewable energy integration with the practicalities of energy storage and dispatchable power sources. They use a technique called imitation learning to create a stochastic neural model predictive control (MPC) system. This approach allows them to account for the uncertainty in renewable energy generation and the optimal control of the energy system over a finite time horizon.
The framework enables decision-makers to evaluate the trade-offs between investment costs and emissions reduction for different configurations of wind and battery energy storage system (BESS) capacities. The researchers applied their framework to a case study of an offshore energy system comprising a gas turbine, a wind farm, and a BESS. They found a complex, non-linear relationship between the investment costs and the reduction in gas usage relative to the capacities of the wind farm and the BESS.
This research has practical applications for the energy industry, particularly in the design and operation of isolated or off-grid energy systems. By providing a flexible and computationally efficient sizing framework, the researchers’ work can help energy providers make informed decisions about the integration of renewable energy sources, the sizing of energy storage systems, and the reduction of greenhouse gas emissions. The study underscores the importance of accounting for optimal control in the design of isolated energy systems to maximize the benefits of renewable energy integration.
Source: Halvdansson, S., Bernardino, L. F., & Knudsen, B. R. (2023). Accounting for Optimal Control in the Sizing of Isolated Hybrid Renewable Energy Systems Using Imitation Learning. Applied Energy, 335, 120840.
This article is based on research available at arXiv.