A high granularity state-space method for contaminant detection and isolation in intelligent buildings

Abstract

Conservation of the Indoor Air Quality is essential in modern, energy efficient, intelligent buildings. However, accidents or malicious acts that often result in various airborne contaminant releases can endanger the occupants' wellbeing. In this work, an indoor air quality monitoring methodology is presented for detecting and localizing a contaminant source in the 3D indoor environment. Specifically, a state-space model is used to describe the contaminant dispersion in a zone which is discretised into multiple cuboid cells. The airflow exchange between the cells is computed based on a 3D discretized coarse-grid CFD analysis. A contaminant detection and localization methodology that considers modelling uncertainty and measurement noise, is adapted and applied to the CFD-based state-space model for detecting the existence of a possible contaminant source and estimating its location within the zone. The performance of the approach is illustrated through simulation examples.