Smart HVAC: Anomaly Detection and Energy Prediction Using Artificial Intelligence

Continuous growth in AI ML approaches has set the pace to enhance sound predictive algorithms for the air conditioning systems. Anomaly detection is such a field where additional optimizations can be achieved in order to minimize the energy consumed manyfold. The approach presented in this research work aims at presenting a new approach at building a more reliable anomaly detection model using AI models derived from clustering models with the silhouette coefficient of 0.35. The novel approach was trained on a time series dataset of indoor and outdoor environmental. Furthermore, we chose and compared the performances of two types of recurrent neural network (RNN) for energy consumption prediction after anomaly detection and elimination: parameters. Evaluated accuracies of the selected RNN models was 0.91 and 0.96 for the current study. The results demonstrate how AI can improve the effectiveness of heating, ventilation, and air conditioning or HVAC systems.