A simplified multi-zone model for determining the placement of bio-defense sensors in large buildings

Abstract

The anthrax mailings of 2001 increased public and government awareness to the threat of bio-terrorism. Particularly vulnerable to a bio-terrorist event are large indoor facilities such as convention centers, office buildings, transportation centers, and sports arenas with their high population densities and limited physical security. Under heightened threat levels deploying bio-aerosol sensors inside these facilities provides added protection to the occupants. The challenge is determining the number and placement of sensors needed to guarantee the detection of a release inside a particular building. The methodology proposed here aims to simplify the analysis of contamination transport within buildings and provide first-order sensing requirements for dose dependant sensors in large facilities. A reduced-order model is developed that allows buildings to be subdivided into larger sections while maintaining a higher degree of accuracy than building analysis models with the same level of granularity. The problem is formulated as a network model with the nodes representing possible sensor locations and the path lengths equal to the reduction in dose as a contaminant travels between sensor locations. Techniques borrowed from network theory are then used to determine the minimum cost set of sensors that provides full building coverage. The reduced-order model estimates sensing requirements in hours or days for problems that would take months to analyze with fine grained multi-zone models and that are too large to be considered with computational fluid dynamics. Models of an office building, a convention center, and an airport terminal are constructed and their underlying network graph is employed to understand how the structure of the indoor environment affects the placement of sensors.

(cont.) Additionally, the equations derived to formulate the network model are used to quantify the optimal tradeoff between sensor sensitivity and cost as a function of building parameters. Future efforts will continue on this path, focusing on how easily discernible building properties such as size, HVAC layout, and air exchange rates can be used to predict the sensing requirements in large indoor spaces.