.png){kind=icon}
.png){kind=icon}
Return
Multi-objective electromagnetic wave propagation algorithm for power flow optimization with renewable energy in a power system
.png){kind=icon}
OA
AI
M
K
G
D
B
A
H
DOI：10.3389/fenrg.2026.1820144.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing demand for energy has heightened the importance of optimal power flow (OPF) in achieving robust planning and cost-effective operation of power systems. OPF primarily seeks to reduce total generation costs while adhering to system constraints. Growing environmental challenges and the declining availability and escalating costs of fossil fuels necessitate the integration of renewable energy sources (RESs) into the power grid. Classical OPF; a non-convex and non-linear optimization problem; traditionally focuses on thermal generators. The complexity increases significantly when incorporating uncertain RESs. This article proposes the electromagnetic (EM) wave propagation algorithm (EMWPA) for single-objective OPF and extends it to multi-objective OPF with RESs. Stochastic RES outputs are represented by appropriate probability density functions with associated reserve and penalty costs explicitly embedded in the objective functions. Multi-objective EMWPA (MOEMWPA) augments the base algorithm through an archive-free; in-population; non-dominated sorting and crowding distance mechanism. Validation on the IEEE modified 30-bus system across three bi-objective and one tri-objective cases demonstrates that MOEMWPA achieves the highest normalized hypervolume (HV) indicator in all multi-objective cases; with superior convergence and evenly distributed non-dominated solutions. The tri-objective case additionally shows the lowest execution time; confirming computational efficiency. Scalability assessment on the IEEE 57-bus system confirms that EMWPA achieves the lowest single-objective generation cost among all competing algorithms. In three bi-objective cases on the 57-bus system; MOEMWPA achieves the highest mean HV in all cases and demonstrates statistically significant superiority over NSGA-II across all three. Notably; multi-objective particle swarm optimization (MOPSO) exhibits severe convergence failures on the larger network; whereas MOEMWPA maintains consistent robustness. These results establish MOEMWPA as an effective; scalable; and computationally efficient solver for multi-objective OPF problems in renewable-integrated power systems.
Keywords:
optimization
metaheuristic algorithms
optimal power flow
emission reduction
power loss minimization
electromagnetic wave propagation algorithm
generation cost minimization
Journal
IF:
2.4
Papers:
923
Citations:
1.4W
Organization
Cited Papers