Computational cost of adjoint-based pollutant source characterization in OpenFOAM
Fugitive emissions from unintended and unmonitored locations, e.g., leaking valves or fittings, are important sources of greenhouse gas. An alternative approach to IR cameras to locate and quantify these leaks is a continuous concentration sensor network. The OpenFOAM toolkit was used to create numerical solvers for gas transport model and source definition. OpenFoam is capable of reading in both steady-state and transient wind fields, and diffusivity. The volumetric emission rate was also defined for each cell and taken as the emission characterization parameters. Different source types and locations can be considered as cell combinations, thus, both the scalar transport modeling time and the number of emission parameters increase as the mesh is refined. Total runtime to reach convergence for emissions characterization is affected by mesh discretization, even when scalar transport run time increase is considered. This is an abstract of a paper presented at the 109th AWMA Annual Conference and Exhibition (New Orleans, LA 6/20-23/2016).
|Keywords||Dispersion modeling, Fugitive emissions, Inverse dispersion modeling, Pollutant characterization|
|Conference||109th Air and Waste Management Association Annual Conference and Exhibition: Unmasking the Industrial Renaissance, ACE 2016|
Brereton, C.A. (Carol A.), Johnson, M.R, & Campbell, L.J. (Lucy J.). (2016). Computational cost of adjoint-based pollutant source characterization in OpenFOAM. In Proceedings of the Air and Waste Management Association's Annual Conference and Exhibition, AWMA (pp. 1821–1825).