Hybrid Algorithm Optimizes Renewable Distributed Generation Under Uncertainty in Power Networks

A novel optimization planning model improves the placement and sizing of distributed generation units within power distribution networks experiencing renewable generation uncertainty, according to a study by researchers Ayat Ali Saleh and Gaber Magdy. The investigators developed the Improved Genetic Moth Swarm Algorithm, a hybrid framework that merges the Genetic Algorithm with an enhanced Moth Swarm Algorithm to address complex planning problems in distribution infrastructure.

The optimization framework is designed to reduce total investment costs, decrease line losses, and enhance voltage stability across radial distribution networks. To address the inherent variability of renewable energy sources, the model accounts for photovoltaic panels, wind turbines, and combinations paired with fuel cells, evaluating renewable-generation uncertainties using the Point Estimate Method.

The research team simulated and optimized three specific hybrid configurations: a combined photovoltaic, wind turbine, and fuel cell arrangement; a photovoltaic and fuel cell system; and a wind turbine and fuel cell configuration. These configurations were tested and evaluated on both the standard IEEE 33-bus and the IEEE 119-bus distribution systems to assess algorithm behavior across different network scales.

The algorithm yielded competitive Pareto-optimal solutions by improving the balance between exploration and exploitation, successfully reducing active power losses, optimizing network operating costs, improving voltage profiles, and facilitating the cost-effective integration of hybrid renewable sources.

With the expanding adoption of distributed generation units based on renewable energy, distribution networks face operational challenges that necessitate advanced planning models to guarantee smooth integration and enhance overall system performance. The primary objective of the proposed planning model is to solve the complex placement and sizing problem associated with distributed generation under conditions of renewable-generation uncertainty. By formulating an optimization approach that accounts for variable renewable energy sources alongside conventional distributed generation, the framework specifically targets the simultaneous reduction of initial investment costs, the mitigation of line losses, and the enhancement of voltage stability across radial distribution networks.

To capture the fluctuations inherent in renewable generation, the methodology applies the Point Estimate Method to model operational uncertainties across variable renewable energy sources. The modeled generation assets include photovoltaic panels, wind turbines, and hybrid pairings incorporating fuel cells. The simulation protocol specifically evaluates and optimizes three designated configurations: photovoltaic, wind turbine, and fuel cell units; photovoltaic panels combined with fuel cells; and wind turbines integrated with fuel cells. To evaluate performance across different distribution network scales, these three hybrid arrangements are simulated and optimized on both the IEEE 33-bus and the IEEE 119-bus distribution systems.

At the computational core of the planning model, the Improved Genetic Moth Swarm Algorithm merges the Genetic Algorithm with an improved version of the Moth Swarm Algorithm to address the complex placement problem. This hybrid formulation achieves an improved balance between exploration and exploitation during the search process. In comparative evaluations against the standard Genetic Moth Swarm Algorithm and alternative optimization algorithms, the proposed method yields competitive Pareto-optimal solutions. These findings demonstrate that the proposed approach can effectively reduce power losses, optimize operating costs, improve voltage profiles, and support the reliable, cost-effective integration of hybrid renewable resources into radial distribution networks.

Subject of Research: Optimization of hybrid renewable distributed generation placement and sizing under uncertainty

Article Title: Uncertainty-aware planning of hybrid renewable and conventional distributed generation using a novel optimization framework

Article References: Saleh, A. A., & Magdy, G. (2026). Uncertainty-aware planning of hybrid renewable and conventional distributed generation using a novel optimization framework. Scientific Reports. https://doi.org/10.1038/s41598-026-74723-w

Image Credits: AI Generated

DOI: 10.1038/s41598-026-74723-w

Keywords: distributed generation, power distribution networks, renewable energy sources, photovoltaic panels, wind turbines, fuel cells, Point Estimate Method, IEEE 33-bus system, IEEE 119-bus system, Improved Genetic Moth Swarm Algorithm, Genetic Algorithm, Moth Swarm Algorithm, voltage stability, line loss reduction, Pareto-optimal solutions, renewable generation uncertainty, radial distribution networks, optimal DG sizing