Memetic Salp Swarm Algorithm for economic load dispatch problems

Area of Science:

• Electrical Engineering
• Computational Intelligence
• Optimization Algorithms

Background:

• Economic Load Dispatch (ELD) is a critical optimization problem in electrical engineering.
• Traditional methods struggle with the non-convex, multi-modal, and severely constrained nature of ELD.
• Existing algorithms often fall short in efficiently solving complex ELD scenarios.

Purpose of the Study:

• To present a hybrid Salp Swarm Algorithm (SSA) integrated with Adaptive β-hill climbing optimizer (AβHCO) for solving Economic Load Dispatch (ELD) problems.
• To introduce a novel memetic algorithm (MSSA) that combines global search (SSA) and local refinement (AβHCO) for enhanced optimization.
• To evaluate the performance of the proposed MSSA algorithm on various ELD problem instances with different constraint sets.

Main Methods:

• Hybridization of Salp Swarm Algorithm (SSA) with Adaptive β-hill climbing optimizer (AβHCO) to create a memetic algorithm (MSSA).
• SSA acts as a general refinement agent (gene encoding), while AβHCO provides local refinement (meme).
• Testing MSSA on diverse ELD problems, including those with load balance, output, restricted operating zones, and ramp rate limit constraints, across multiple generator unit configurations (3UG-850 MW to 80UG-21000 MW).

Main Results:

• The proposed MSSA algorithm demonstrated significant feasibility and usefulness in solving complex ELD problems.
• Comparative results indicate that MSSA outperforms existing algorithms in handling severe constraints.
• MSSA effectively addressed various ELD scenarios, from small to large-scale generator systems.

Conclusions:

• The hybrid MSSA algorithm offers a robust and efficient solution for Economic Load Dispatch problems.
• The memetic approach combining global and local search effectively tackles the challenges of non-convexity and severe constraints in ELD.
• MSSA shows promise for practical application in power system optimization.