Optimizing space heating efficiency in sustainable building design a multi criteria decision making approach with...

Area of Science:

• Building Science
• Sustainable Energy Systems
• Artificial Intelligence in HVAC

Background:

• Efficient space heating is crucial for sustainable buildings, impacting energy consumption, costs, and occupant comfort.
• Nearly-zero energy buildings (nZEBs) require optimized heating strategies, especially in cold climates like Oslo, Norway.
• Balancing heating costs and thermal comfort is a key challenge in nZEB design.

Purpose of the Study:

• To investigate advanced control strategies for optimizing space heating in an nZEB in Oslo.
• To explore how a novel Model Predictive Control (MPC) framework can efficiently balance heating costs and thermal comfort.
• To assess the potential of integrating predictive control for real-time building management.

Main Methods:

• Development of a novel Model Predictive Control (MPC) framework integrating Long Short-Term Memory (LSTM) neural networks for prediction and the Ant Nesting Algorithm (ANA) for optimization.
• Utilizing EnergyPlus simulations and real weather data for dynamic indoor temperature and heating requirement predictions.
• Incorporating ASHRAE Standard 55 adaptive comfort constraints and performing Pareto set analysis for optimization outcomes.

Main Results:

• The proposed MPC framework demonstrated the ability to minimize heating costs (HC) and comfort penalties (CP) simultaneously.
• Up to 17% daily heating cost savings were achieved compared to conventional methods.
• Comparable thermal comfort levels were maintained throughout the optimization process.

Conclusions:

• The novel MPC framework offers a promising approach for optimizing space heating in nZEBs under cold climatic conditions.
• The framework has the potential for integration into automated systems for real-time predictive control in buildings.
• This research contributes to improving energy efficiency and occupant comfort in sustainable building design.