Optimization Model for an Individualized IoT Ambient Monitoring and Control System

Abstract

The urban population has increased in many parts of the world, concentrating mainly in large cities, inside buildings. Thus, it is important to optimize these buildings' environments, whether in terms of its users' comfort, or in terms of energy resources. This article presents an optimization model with the goal of guaranteeing individualized comfort parameters. It is based in a flexible HVAC IoT system, previously developed using the fog computing paradigm. In order to test the model's performance, a set of simulations was performed, using real data from our IoT laboratory. The comfort values were obtained by training a Naïve Bayes model with data found in the literature to represent hot-natured and cold-natured profiles. The simulation's result shows that the system adequately reacts to internal and external changes in the environment, keeping the indoor temperature inside the comfort range most of the time, while still using few HVAC resources.