A Nowcasting Central Controller with Continuous RTP for Residential Device Scheduling using Swap-Based BFPSO

Area of Science:

• Energy Systems Engineering
• Artificial Intelligence in Energy
• Consumer Behavior in Energy Markets

Background:

• Residential energy management is crucial for consumers to adapt to electricity market volatility.
• Traditional forecasting-based scheduling models struggle with inherent uncertainties, limiting their effectiveness.
• Optimizing energy consumption requires adaptable models that can handle real-time price fluctuations.

Purpose of the Study:

• To propose a novel scheduling model, the Nowcasting Central Controller (NCC), for residential devices.
• To optimize device scheduling for current and future time slots using real-time data.
• To evaluate the NCC model's performance under different electricity pricing schemes.

Main Methods:

• Implementation of a Nowcasting Central Controller (NCC) for residential energy management.
• Utilization of four variants of Particle Swarm Optimization (PSO) with swapping operations for problem-solving.
• Development of a normalized objective function incorporating two cost metrics for optimization.

Main Results:

• The Best First Particle Swarm Optimization (BFPSO) variant demonstrated significant speed and cost reduction at each time slot.
• A comparative analysis confirmed the superiority of Critical Real-Time Pricing (CRTP) over Day-Ahead Pricing (DAP) and Time-of-Use (TOD) schemes.
• The NCC model proved highly adaptable and robust against abrupt changes in pricing structures.

Conclusions:

• The proposed NCC model, particularly with BFPSO and CRTP, offers an effective solution for residential energy management.
• The model's reliance on current data enhances its implementability and robustness in dynamic energy markets.
• This research highlights the potential of real-time control for optimizing energy consumption and reducing costs for consumers.