Automatic Detection and Control of Hazardous Plumes in Wall-Bounded Flow Systems
[electronic resource].
Description
- Language(s)
-
English
- Published
-
2013.
- Summary
-
investigate the effect of various parameters on their accuracy, reliability and expediency. The results indicate that gradient-based optimization methods are successful for the boundary control phase; however the source inversion problem is complicated by issues arising from the discretization of the optimization parameters and ill-posedness. Hybrid optimization approaches offer some benefits with regards to the former problem, yet ill-posedness remains a significant challenge with respect to the source inversion process.
inversion phase in order to trace the history of the plume and determine its original properties, followed by a boundary control phase in which the recovered source information is used to predict the propagation of the plume in time and space and thereby determine a control strategy to be performed in order to effectively mitigate it. Both the source inversion and boundary control phases can be formulated in terms of numerical optimization, and hence in this work a coupled CFD-optimization model has been developed using open source software. The CFD model implemented in this research is a RANS (Reynolds-Averaged Navier-Stokes) based finite-volume model developed using the OpenFOAM CFD library. The CFD model has been linked to an external optimization software suite (DAKOTA). The resulting model is used to simulate the source inversion and boundary control components of the automatic detection and control algorithm and to
Recent advances in technology and computational power have made the once theoretical concept of a real-time detection and control system, for the purpose of reducing risk from the deliberate or unintentional release of a hazardous plume, a practical reality. Such a system utilizes strategically placed sensor arrays and actuators in order to first detect the release of a hazardous chemical, and subsequently to determine the actions required by the actuators to effectively and expediently mitigate the plume. This basic framework is applicable to a number of real-world scenarios that can be described as wall-bounded, fluid-based systems, such as airport terminals, aqueducts, tall buildings and passenger tunnels. The present research develops the theoretical and numerical framework for the automatic detection and control algorithm, which requires two main steps: a source
- Locate a Print Version
-
Find in a library service is not available from this catalog. Search Worldcat
Viewability