A machine learning technique for optimizing load demand prediction within air conditioning systems utilizing GRU/IASO...

Area of Science:

• Engineering
• Computer Science
• Energy

Background:

• Air conditioning systems are vital for thermal comfort in hot, humid climates.
• High energy consumption by these systems necessitates efficient energy management.
• Accurate load demand forecasting is essential for optimizing air conditioning system performance.

Purpose of the Study:

• To introduce a novel machine learning model for dynamic optimal load demand forecasting in air conditioning systems.
• To enhance energy management and operational efficiency in air conditioning.

Main Methods:

• Development of a model integrating a Gated Recurrent Unit (GRU) network, a type of recurrent neural network adept at handling temporal data.
• Optimization of the GRU network using an enhanced metaheuristic algorithm, the Improved Alpine Skiing Optimizer (IASO).
• Training and validation of the proposed GRU/IASO model using real-world data from a commercial complex in a hot and humid climate.

Main Results:

• The GRU/IASO model demonstrated significant accuracy and robustness in load demand forecasting.
• Performance evaluation showed advantages over other commonly used forecasting techniques (specific techniques not detailed in the abstract).

Conclusions:

• The proposed GRU/IASO model offers a highly accurate and robust solution for dynamic optimal load demand forecasting in air conditioning systems.
• This approach contributes to better energy management and optimization for air conditioning in demanding climates.