Enhanced multi-objective evolutionary mating algorithm with improved crowding distance and levy flight for optimizing comfort index and energy consumption in smart buildings

Muhammad Naim, Nordin and Mohd Herwan, Sulaiman and Nor Farizan, Zakaria and Zuriani, Mustaffa (2025) Enhanced multi-objective evolutionary mating algorithm with improved crowding distance and levy flight for optimizing comfort index and energy consumption in smart buildings. Franklin Open, 11 (100286). pp. 1-17. ISSN 2773-1871. (Published)

[img]
Preview
Pdf
1-s2.0-S2773186325000763-main.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (2MB) | Preview

Abstract

This paper introduces a novel Multi-Objective Evolutionary Mating Algorithm (MOEMA) designed to address the inherent challenges of optimizing comfort index and energy consumption in smart building systems. While current Evolutionary Mating Algorithms (EMA) primarily focus on single-objective optimization and rely on weighted functions for handling multiple objectives, such approaches prove impractical for the complex trade-offs between comfort index and energy efficiency. The proposed MOEMA enhances the original EMA framework through two key innovations: an improved crowding distance function inspired by the Non-dominated Sorting Genetic Algorithm (NSGA) to enhance solution diversity and selection pressure, and the integration of Levy flight mechanics to improve exploration efficiency by balancing local and global searches. These enhancements enable MOEMA to effectively navigate complex multi-objective landscapes, leading to more diverse and well-converged Pareto-optimal solutions. The algorithm's performance is thoroughly assessed using the chosen benchmark functions and validated through practical applications in smart building environments. It simultaneously optimizes various comfort parameters, including temperature, illuminance, and air quality, while minimizing energy consumption and maximizing the comfort index. Comparative analysis against established algorithms, like NSGA-II demonstrates MOEMA's effectiveness in achieving superior solution diversity and convergence characteristics. The results indicate that MOEMA offers a robust framework for handling the complex balance between the smart building's comfort index and energy usage where it achieves 0.03 better at comfort index and with 10.65 lower energy consumption than NSGA-II. It contributing to the broader fields of building automation and sustainable development while aligning with Industry 4.0 initiatives.

Item Type: Article
Additional Information: Indexed by Scopus
Uncontrolled Keywords: Comfort index optimization, Crowding distance, Energy efficiency, Evolutionary Mating Algorithm, Levy flight, Multi-objective optimization, Pareto optimization, Smart buildings
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
T Technology > TK Electrical engineering. Electronics Nuclear engineering
Faculty/Division: Faculty of Electrical and Electronic Engineering Technology
Institute of Postgraduate Studies
Faculty of Computing
Centre for Research in Advanced Fluid & Processes (Fluid Centre)
Depositing User: Dr. Mohd Herwan Sulaiman
Date Deposited: 09 Jul 2025 01:04
Last Modified: 09 Jul 2025 01:04
URI: http://umpir.ump.edu.my/id/eprint/44973
Download Statistic: View Download Statistics

Actions (login required)

View Item View Item