Integrating digital twins and neural networks for real-time temperature management in smart homes: An innovative approach using ZigBee networks

This paper presents a novel approach to real-time temperature management in smart homes by integrating digital twins with neural networks, leveraging the capabilities of ZigBee-enabled wireless sensor networks. The proposed system continuously monitors and analyzes environmental data collected from strategically placed Xbee S2 modules within a smart home, facilitating dynamic temperature control through an intelligent management system. A digital twin of the smart home is created to simulate and predict temperature variations based on real-time data, while a neural network model optimizes the control strategies to maintain desired thermal comfort levels. Results demonstrate that the integrated system achieves significant improvements in temperature regulation efficiency. Specifically, the proposed method reduced temperature variance by 14.12 % compared to traditional thermostat-based systems, leading to a more consistent indoor climate. Additionally, the energy consumption associated with heating and cooling was reduced by 8.05 %, highlighting the system's potential for energy savings. The neural network model, trained on historical data, achieved a prediction accuracy, enabling precise adjustments to the HVAC system in response to predicted temperature changes. Overall, the integration of digital twins and neural networks with ZigBee-based sensor networks offers a powerful and efficient solution for real-time temperature management in smart homes, demonstrating substantial advancements in both comfort and energy efficiency.