Swarm intelligence for energy-efficient heating, ventilation, and air conditioning (HVAC) systems: A case study in smart buildings

Swarm intelligence for energy-efficient heating, ventilation, and air conditioning (HVAC) systems: A case study in smart buildings

This research utilizes swarm intelligence algorithms—Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), and hybrid PSO-ACO-to optimize energy efficiency and thermal comfort in smart building HVAC systems. A thorough experimental analysis was done in a 500-kW cooling capacity smart bui...

Full description

Saved in:
Bibliographic Details
Main Authors: Vinoth Kanna I, Raja Subramani, Maher Ali Rusho, Shubham Sharma, Ramachandran T, Abinash Mahapatro, Deepak Gupta, Jasmina Lozanovic
Format: Article
Language:English
Published: Elsevier 2025-10-01
Series:Case Studies in Thermal Engineering
Subjects:
Swarm intelligence
Particle swarm optimization
Ant colony optimization
HVAC optimization
Energy efficiency
Smart buildings
Online Access:http://www.sciencedirect.com/science/article/pii/S2214157X25010834
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This research utilizes swarm intelligence algorithms—Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), and hybrid PSO-ACO-to optimize energy efficiency and thermal comfort in smart building HVAC systems. A thorough experimental analysis was done in a 500-kW cooling capacity smart building in 20 monitored temperature zones for 12 months. The hybrid PSO-ACO model performed the best energy savings of 28.9 %, better than PSO (23.7 %) and ACO (20.5 %), and also saving 29.2 % peak load demand. Thermal comfort analysis through Predicted Mean Vote (PMV) and Predicted Percentage of Dissatisfied (PPD) metrics showed better indoor conditions, with the hybrid model keeping room temperatures within ±0.8 °C of the setpoint and bringing the PPD index down to 8.5 %. Statistical validation through ANOVA and t-tests supported the energy and comfort gains with p-values always less than 0.05. The hybrid model also exhibited faster convergence, completing the optimization in 120 iterations, 20 % faster than ACO. Economic analysis estimated annual cost savings of $12,000 for a 10,000 m2 building with a return on investment within 2.5 years, while saving 35.2 metric tons of CO2 emissions per year. The results illustrate the outstanding performance of hybrid swarm intelligence algorithms to improve HVAC efficiency and occupant comfort, a scalable and inexpensive solution for smart building management. Future research will utilize machine learning combined with swarm intelligence for adaptive and predictive HVAC control.
ISSN:2214-157X

Similar Items