This paper presents a novel energy-aware Internet of Things (IoT) and edge computing architecture designed to support decentralized infrastructure in low-resource environments. The proposed framework combines virtual MIMO-enabled wireless sensor networks, lightweight edge AI inference models, and nanomaterial-based photovoltaic systems to autonomously manage public utility systems including waste, water, and energy. The system enables localized decision-making, reduces dependency on cloud services, and optimizes energy usage for off-grid deployment. A prototype implementation in a simulated rural setting demonstrated a 28% reduction in energy consumption compared to conventional IoT architectures, with average decision latency reduced to 800 milliseconds and uptime reaching 97.5% over a 30-day period. These results validate the feasibility of deploying scalable, autonomous infrastructure systems in environments with limited connectivity and power availability.
2 June 2024 in Global Mainstream Journal of Innovation, Engineering & Emerging Technology
Designing a state-of-the-art PLC-based extrusion machine with a user-friendly HMI ensures seamless operation, enhancing Overall Equipment Effectiveness (OEE). This project focuses on automating an extrusion system with advanced technologies for optimized functionality and reliability. The architecture includes sophisticated components to boost productivity and product quality. Key aspects involve orderly control and synchronization of the extruder motor, feeder motor, lubrication pump, and vacuum pump for consistent performance with precise temperate profile. A significant innovation is the centralized blower system for machine temperature profile analysis and control, replacing individual controllers to enhance thermal management efficiency and ensure uniform temperature distribution. A high-low temperature alarm system alerts operators to deviations, maintaining process stability. Real-time data on current (Amps) and frequency (Hz) is displayed on the HMI from the inverter for monitoring and diagnostics. The system also features machine downline controlling capabilities for efficient management of downstream processes. Collectively, these innovations create a robust, efficient, and user-friendly extrusion machine that enhances OEE and product quality.
The convergence of artificial intelligence (AI) with industrial automation offers a transformative opportunity to enhance operational efficiency, fault tolerance, and cybersecurity in smart manufacturing systems. This paper proposes an AI-augmented framework integrating Programmable Logic Controllers (PLCs), Supervisory Control and Data Acquisition (SCADA) systems, and deep learning models for real-time fault diagnosis and anomaly detection. The framework incorporates cybersecurity features compliant with NIST standards to protect against unauthorized access and system failures. Using AI-enhanced SCADA modules and predictive maintenance models, the system can reduce equipment downtime, optimize operational throughput, and detect cyber-physical threats in real time. The framework is validated through experiments in a simulated smart manufacturing environment, demonstrating improved fault detection accuracy and system resilience. This research supports U.S. national priorities under the CHIPS Act and Executive Order 14028 by promoting secure, intelligent, and future-ready industrial ecosystems.