article · Energy Conversion and Management X
Off-grid electrification using islanded hybrid microgrids requires robust energy management systems capable of balancing generation costs, equipment degradation, and operational performance while maintaining strict reliability. A 24-hour dispatch problem was formulated as a constrained bi-objective optimisation framework with 72 continuous decision variables. The setup simultaneously enforced three hard operational limits: power supply reliability, battery state-of-charge cyclicity, and hydrogen storage cyclicity. Three constrained multi-objective evolutionary algorithms, BiCo, C-TSEA, and CMOSMA, were evaluated across 20 independent runs under three distinct meteorological conditions ranging from surplus renewable supply to severe weather stress. BiCo demonstrated superior performance across all scenarios, achieving the highest median hypervolume metrics and the top statistical ranking. At its best-compromise operating point, BiCo delivered an energy cost as low as 0.0598 euros per kilowatt-hour alongside a carbon dioxide intensity of 0.063 kilograms per kilowatt-hour, offering an effective framework for day-ahead scheduling.
Remote communities and isolated sites rely on off-grid microgrids that combine renewable generation, batteries, and hydrogen storage. Balancing operating costs against equipment wear while avoiding blackouts is technically difficult. Demonstrating that specialised evolutionary algorithms can solve these complex multi-objective trade-offs under varying weather conditions helps system planners deliver more affordable, reliable, and low-carbon electricity to off-grid installations.
This dispatch framework can be embedded into energy management systems by microgrid operators, commercial off-grid developers, and industrial facilities using hybrid hydrogen and battery setups. The technology is applied research tested in simulation across multiple weather profiles, placing it at an intermediate stage before market entry. Commercial adoption would require embedding the algorithm into physical microgrid controllers and validating its performance against real-time sensor and forecasting data.
AI-generated from the published abstract. Always read the original work before citing.
As microgrids are being deployed globally for off-grid electrification in islanded hybrid configurations,there is a growing need for energy management systems (EMS) that optimize energy generationcost, degradation, and performance, subject to reliability constraints, to achieve optimal performance.Although multi-objective EMS optimization of hybrid microgrids has recently attracted interest inliterature, there is still a need to address the challenging problem of simultaneously enforcing hardreliability constraints and quantifying equipment wear as an explicit optimization objective within aconstrained Pareto framework. This paper formulates a 24-hour dispatch problem for a fully islandedhybrid microgrid as a Type II constrained bi-objective optimization problem (CMOP) with 𝐷 = 72continuous decision variables and three simultaneously enforced hard constraints: loss of powersupply probability (LPSP ≤ 2%), battery state-of-charge cyclicity, and hydrogen tank level cyclicity.Three dedicated constrained multi-objective evolutionary algorithms (CMOEA), BiCo, C-TSEA,and CMOSMA, were benchmarked over 20 independent runs across three meteorological scenarios,ranging from surplus-renewable toweather-stressed conditions, using the hypervolume (HV) indicatorand Friedman–Nemenyi post-hoc analysis. BiCo achieved the highest median hypervolume in allthree scenarios (0.8723, 0.1245, and 0.8421 for SC1–SC3, respectively) and the lowest averageFriedman rank of 1.08, with statistically confirmed superiority over both competitors. At the BiCobest-compromise operating point, the cost of energy reached as low as 0.0598 e/kWh under bestconditions, with a CO2 intensity of 0.063𝑘𝑔𝐶𝑂2∕𝑘𝑊 ℎ. These findings establish explicit CMOPformulation with bidirectional coevolution as a rigorous and practically deployable framework forday-ahead energy management of islanded hybrid microgrids.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1016/j.ecmx.2026.102249
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
