Improved PSO Continuous Power Control of Aggregated Air Conditioning Loads in Demand Response

The rising penetration of renewable energy sources and the widening peak-valley difference challenges the peak shaving capacity of power generation systems. We propose a continuous power control method for aggregated air conditioning loads. It leverages demand-side flexibility to meet smart grid peak shaving demands via an improved particle swarm optimization algorithm. Specifically, we introduce PSO-AUDBTF, which enhances global optimization performance through deterministic boundary initialization and asynchronous update strategies. Additionally, we design a temperature feedback-based dynamic weight adjustment strategy to ensure real-time thermal comfort for user cluster. Simulations in typical demand response scenarios validate our approach, showing that it outperforms benchmark algorithms by achieving continuous power regulation while maintaining user cluster thermal comfort.