Managing optimal power flow becomes increasingly difficult as renewable energy sources proliferate throughout modern electrical grids. Solar and hydroelectric generators introduce uncertainty and variability that complicate traditional optimization approaches, creating demand for more sophisticated algorithms capable of handling both the technical and economic dimensions of grid operations.
Researchers have developed a novel hybrid optimization technique combining osprey and salp swarm algorithms to address this challenge. The osprey optimization component provides strong exploration capabilities, while the salp swarm algorithm contributes adaptive dynamics that mimic natural follower-leader behavior. Together, these metaheuristic methods balance exploration and exploitation to find near-optimal solutions efficiently.
The proposed approach incorporates probabilistic modeling of renewable generation. Solar irradiance variability follows a lognormal distribution, while water availability for hydroelectric plants uses a Gumbel distribution. These representations enable the algorithm to account for real-world uncertainty rather than assuming idealized conditions.
Testing occurred across IEEE standard test systems ranging from 30 to 118 buses, evaluating five optimization objectives: economic cost minimization, emission reduction, combined economic-environmental optimization, voltage stability, and renewable uncertainty management. Results showed the hybrid algorithm delivered 0.4% to 17% cost improvements compared to classical and competing hybrid methods, with advantages expanding as system size increased. Voltage deviation minimization improved by 3% to 23%, while uncertainty penalties decreased by 2% to 8%.
The algorithm converged rapidly, typically within 20 to 50 iterations, and effectively avoided becoming trapped in local optima—a persistent problem for traditional optimization methods. Performance remained robust across both single-objective and multiobjective scenarios.
These results suggest significant potential for deployment in real-time grid control centers managing high renewable penetration. As utilities worldwide accelerate renewable integration, optimization algorithms capable of handling complexity while maintaining computational speed become essential infrastructure components for reliable, economical grid operations.
