Self-Learning Evolutionary Control Networks for Adaptive Energy Optimization in Smart IoT Cyber-Physical Systems

The explosive growth of IoT devices in cyber-physical systems (CPS) has caused tremendous energy consumption issues, requiring intelligent, scalable optimization techniques. Rule-based and static machine learning models do not learn adaptability in dynamic environments and are incapable of generalizing in diverse IoT deployments. The present study introduces a Self-Learning Evolutionary Control Network (SLECN) to obtain adaptive, real-time energy optimization in smart IoT-CPS environments. The architecture combines evolutionary algorithms for policy evolution with reinforcement learning for self-adaptation. It consists of preprocessing, feature engineering, model training, and deployment. A private IoT Energy Consumption dataset gathered from smart meters, sensors, and appliances in residential and commercial environments. The hybrid strategy ensures scalability, real-time adaptability, and less energy wastage without compromising system performance. The model attained 56.7 % energy efficiency, 145 ms response time, and <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\text{9 5. 4 \%}$</tex> resource utilization. Python was employed for full-stack implementation, which involved preprocessing, model training, and deployment.

Ҳали таржима қилинмаган