Sensor networks for routine indoor air quality monitoring in buildings: Impacts of placement, accuracy, and number of sensors

Abstract

High-performance buildings will increasingly use sensors to control ventilation and manage indoor air quality, but there is little knowledge regarding placement and accuracy limitations. The authors simulated and assessed two approaches for locating from 1 up to 10 hypothetical sensors for measuring carbon dioxide and the sum of volatile organic compounds in typical offices. The first network approach was intended to best measure spatial variability; the second simply sited each additional sensor in the mechanical system's return duct and averaged readings. Each approach was compared at three levels of sensor accuracy in four different office cases, within a Monte Carlo analysis that varied hundreds of relevant parameters. Results indicated that concentrations in areas served by a common mechanical system, given similar per-area supply flows and at least some air recirculation, were reasonably homogeneous. The standard deviation of concentrations among locations was almost always <20% of the mean concentration, and variability was generally overwhelmed by sensor error when building grade sensors were modeled. One implication was that capturing spatial distribution is likely to be less important than improving sensor accuracy, and without better quality sensors, attempting to do so can actually degrade network accuracy.