Application of Reinforcement Learning in Energy Management of Microgrids and Hybrid Energy Storage Systems

Dear Colleagues,

The increasing application of renewable energy sources for alleviating energy poverty and reaching climate goals has led to massive deployment of microgrids as they offer a scalable way of integrating renewable sources and different forms of energy storage into the utility grid. Since renewable resources such as solar and wind are intermittent and weather dependent, the design and energy management of microgrids need to aim towards providing a secure and stable energy supply to its customers and achieving cost-optimal and sustainable operation. Microgrid design, energy management and grid-integration planning raise significant challenges due to the stochastic nature of resources and loads and their interaction with the grid while operating in the grid-tied mode. It is also well noted that deploying energy storage systems can significantly buffer the impacts of these uncertainties as they provide various auxiliary services to the power system, i.e., load shifting, frequency regulation, voltage support and grid stabilization. Thus, for the microgrid to guarantee a reliable supply of power and efficient utilization of the battery storage and renewable resources, an energy management system (EMS) needs to be developed.

In recent times, intelligent learning-based techniques are being increasingly applied in decision-making problems and have also proved ideal in overcoming these limitations, as they can automatically extract, monitor, and optimize generation and demand patterns. They can relax the idea of an explicit system model to ensure optimal control. This is of great benefit in energy management as this is a normally partially observable problem, i.e., hidden or unknown information always exists. The reinforcement learning (RL) approach falls under machine learning and is well known for its ability to solve problems in stochastic environments. It aims at making optimal time-sequential decisions in an uncertain environment. Reinforcement learning involves a decision maker (agent) that learns how to act (action) in a particular situation (state) through continuous interaction with the environment to maximize cumulative rewards. Hence, this can be very effective for resource optimization problems in renewable and storage-powered microgrids, where supply and demand are changing rapidly.

This Special Issue aims to present and disseminate the most recent advances related to the application of reinforcement learning in energy management of all types of microgrids (DC, AC, hybrid, static generation, rotational generation, stand-alone, grid-tied) and hybrid energy storage systems. The hybrid energy storage system may include battery, flywheel, hydrogen fuel cells, super capacitors and any other form and combination of storage.

Topics of interest for publication include, but are not limited to:

• Reinforcement learning approach in energy management and optimal operation of renewable energy systems with grid-following and grid-forming inverters.
• Application of reinforcement learning in energy management in islanded and grid-tied microgrids and hybrid energy storage systems for secure and stable energy supply, grid support and ancillary services.
• Application of reinforcement learning in design, control, protection and optimal operation of microgrids and hybrid energy storage systems.

Dr. Sunetra Chowdhury
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Energies is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.