Optimization of Explosives Sensor Placement in Airports
R2-D.2

Download Project Report (Phase 2, Year 3)

Project Description

Effective detection of explosives-related threats is very important to the country and the world. In the past, the explosives detection community has invested heavily in contact sensing for portal screening. However, they are unable to answer the question: “Great sensors, but how do you know the vapors/particles get there?”

This is because an airport is a large space, so the concentration of the vapors/particles of explosives are generally low. It is difficult to direct the low concentration of explosives agents to the limited amount of sensors. This work can provide effective methods to detect explosives vapors in airports. Optimal sensor placement in water ports or airports requires fast predictions on air distribution and explosive vapor transport.

In Phase 2, Year 2 this project investigated how explosives vapors/particles were transported at the Indianapolis International Airport terminal; identified strategies that can be used to increase explosive vapor concentrations; and demonstrated how to use these strategies to detect explosives with suitable sensors. Computational fluid dynamics (CFD) simulations were used for predictions of air distribution and contaminant transport because they are more accurate than other numerical tools, and much faster than experimental measurements. Since this application is for airports, the space is much larger than those in the previous studies. Thus, CFD was found to be very expensive when it is used to find optimal sensor locations and to identify source positions.

Therefore, our ALERT Phase 2 Year 3 project developed and validated the fast fluid dynamics (FFD), which was a faster method than CFD while maintaining the accuracy. This project implemented four FFD models in OpenFOAM and provided the validations using the experimental data from literature. It was found that the FFD solvers had similar performance on accuracy with CFD simulations in predicting both steady-state and transient flow in indoor environment. The FFD solvers could be almost 40 times faster than CFD for simulating transient flow.

Core funding for this project ends in Year 3 per the outcome of the Biennial Review process. Currently funded students will be supported via the ALERT Science and Engineering Workforce Development Program.

 

The major strength of this project is that the developed FFD solvers can be almost 40 times faster than CFD in predicting transient flow.
Phase 2, Year 3 Annual Report
Project Leader
  • Qingyan “Yan” Chen
    Project Leader
    Purdue University
    Email

Students Currently Involved in Project
  • Wei Lu
    Purdue University