Optimizing Air Conditioning Control with Fuzzy Logic and IoT Sensor Based Automation

Abstract

This research presents the development and implementation of a fuzzy automatic air conditioning system integrated with IoT technology for real-time climate control. The system incorporates a Fuzzy Inference System (FIS) alongside IoT sensor units and a climate control system to automatically regulate temperature and airflow. The IoT sensor unit captures environmental parameters such as room temperature, humidity, air pressure, and weather conditions, transmitting these as input to the FIS. Through fuzzification, logical rule evaluation, and defuzzification, the system generates precise control signals for actuating air conditioning components, ensuring optimized climate regulation. The system architecture supports two implementations: Mamdani FIS (MFIS) and Sugeno Adaptive Neural Fuzzy Inference System (SANFIS). The Mamdani model applies fuzzy rule-based logic, while SANFIS leverages neural network training and adaptive mechanisms for enhanced performance. Extensive simulation experiments using MATLAB and Simulink in order to demonstrate the system's efficiency in dynamically adjusting the varying environmental conditions. Experimental results reveal significant improvements in system performance, energy conservation, and prediction accuracy through adaptive learning techniques.