Novel Integration of Ant Colony Optimisation and Deep Neural Networks in AI Agents for Predictive Maintenance of Sustainable Energy Systems, Mostafa Mohamad
Mostafa Mohamad
College of Interdisciplinary Studies
Zayed University
Abu Dhabi
UAE
ORCID: 0000-0001-7172-1110
Amit Kohli
University Canada West
Canada
ORCID: 0009-0003-5604-1389
Ravi Srivel
Adhiparasakthi Engineering College
Chennai
India
ORCID: 0000-0002-1685-3521
Anas Najdawi
Abu Dhabi University
Dubai
UAE
ORCID: 0000-0002-2485-2465
Paper Type: Research
Received: 29 November 2025 /Revised: 10 February 2026 /Accepted: 17 February 2026 / Published: 30 May 2026
DOI:10.47556/J.WJSTSD.21.5.2026.1
Purpose: This study investigates how renewable energy firms utilise Artificial Intelligence (AI)-powered solutions to balance ecological, economic, and operational value in sustainable infrastructure.
Design/Methodology/Approach: A hybrid framework was developed, combining Deep Neural Networks (DNNs) and Ant Colony Optimisation (ACO) to train autonomous AI agents. The model was validated using multimodal sensor data from wind turbines, photovoltaic panels, and smart grids.
Findings: The framework significantly improved fault detection and maintenance optimisation. Organisationally, swarm intelligence enabled cost-effective resource allocation. Environmentally, the system reduced energy waste and carbon emissions while increasing grid reliability.
Originality/Value: This research uniquely integrates swarm intelligence with deep learning, reconceptualising AI as a foundational agent for autonomous energy management and sustainability.
Practical Implications: The framework provides a roadmap for energy executives to optimise maintenance, achieve decarbonisation goals, and ensure stability through AI-driven resource management.
Keywords: AI Agent; Sustainable Energy; Deep Neural Networks; Predictive Maintenance; Ant Colony Optimisation; Swarm Intelligence; Smart Grids.
Citation: Mohamad, M., Kohli, A., Srivel, R. and Najdawi, A. (2026): Novel Integration of Ant Colony Optimisation and Deep Neural Networks in AI Agents for Predictive Maintenance of Sustainable Energy Systems. World Journal of Science, Technology and Sustainable Development (WJSTSD), Vol. 21, No. 5, pp. xx-xx.