Adaptive Penalty-Based MOPSO for Multi-Objective Economic Dispatch in Virtual Power Plants

Virtual Power Plants (VPPs) are becoming essential tools for integrating conventional and renewable energy sources to meet the growing demands of sustainable power systems. California's ambitious clean energy targets, however, require effective coordination of diverse resources - something current dispatch strategies struggle to achieve, particularly when balancing operational cost and CO₂ emissions. This paper presents a novel Multi-Objective Particle Swarm Optimization (MOPSO) framework for a VPP comprising solar, wind, battery storage, and Combined Heat and Power (CHP) units. The model uses real-world grid data from California ISO (CAISO) to capture operational variability and applies Pareto-based optimization to explore cost-emission trade-offs. Key contributions include adaptive penalty scaling, demand-aware battery control, and costefficient CHP scheduling. Comparative analysis against static and mutation-less MOPSO variants demonstrates that the proposed adaptive method achieves a 2-4× increase in repository diversity and discovers dispatch strategies with up to 3.5% lower emissions at similar cost. The framework successfully generates distinct cost-, emission-, and balance-optimal strategies, offering practical tools for real-world VPP coordination and low-carbon energy planning.