mADM1 – A MULTIDIMENSIONAL PARALLEL NUMERICAL SOLVER OF ADM1 MODEL FOR ANAEROBIC DIGESTERS IN OPENFOAM FRAMEWORK R. NOURI 1 , C. FERNANDES 1 , A. ILANGOVAN 1 , L.P. SOLINO 2 , M. FIGUEROA 2 , I.J. NEIRA 3 , C. COTELO 4 , A. GOMEZ 4 , J.M. NÓBREGA 1 1 Institute for Polymers and Composites/i3N, University of Minho, Campus de Azurém, 4800-058 Guimarães, Portugal 2 EnergyLab, Edificio CITEXVI – Local 1, Fonte das Abelleiras, s/n, Campus Universitario de Vigo,E- 36310 Vigo, Spain 3 Norvento Enerxía, Rúa Aller Ulloa, Ramón María Nº23, 27003 Lugo, Spain 4 Galicia Supercomputing Center (CESGA), Santiago de Compostela, Spain. 28 th of May 2018 The Fortissimo 2 project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 680481.
Outline • Anaerobic digesters • ADM1 (a bulk model) • mADM1 – idea • mADM1 – implementation in OpenFoam • Preliminary Results 2
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results A digester is a huge container where chemical reactions produce methane gas from waste materials. digester aerobic anaerobic Aerobic digester works in the presence of oxygen (fast but expensive) Anaerobic digester works in the absence of oxygen (slow but cheap) 3
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Types of anaerobic digesters: anaerobic Psychrophilic Mesophilic Thermophylic Psychrophilic (4-15 degree) => not efficient enough Mesophilic (20-40) => the most used type Thermophylic(45-70) => expensive 4
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Scheme of a digester: Methane (Gas holder) outlet- waste impeller Inlet - waste 5
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results ADM1 (Anaerobic Digester Model No. 1) by D.J. Batstone (2002) ADM1 is quite a complex model: There are 109 coefficient and 36 variables ADM1 implementation Implementations of ADM1: DAE approach DAE approach DE approach pH pH and h2 6
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results • The idea of multidimensional ADM1 Gaden [2013] • Biggest limitation of Gaden implementation Not parallelized • ADM1 considers only one single cell, a black box. • m-ADM1 is applying ADM1 for each cell, and all the cells have interactions. M-ADM1 ADM1 Black Box Gaden, D. (2013) Modelling Anaerobic Digesters in Three Dimensions: Integration of Biochemistry with Computational Fluid Dynamics. PhD Thesis, University of Manitoba,Canada 7
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results m-ADM1 Chemical Fluid part part • Chemical part solving 36 Differential equations • Fluid part solving fluid flow (the impact of advection and diffusion) • ADMOneFoam It is based on pisoFoam (a transient turbulent solver). 8
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start Initial condition + reading data from dictionaries ADM1 algebraic routines Considering reactions Update variables Next time step source terms Momentum and transport equations Gases concentrations in headspace YES NO End time? Finish 9
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start A Sample dictionary, Initial condition + Stoichiometric coefficients reading data from dictionaries ADM1 algebraic routines Considering reactions Update variables Next time step source terms Initial conditions, Momentum and reading variables transport equations Gases concentrations in headspace YES NO End time? Finish 10
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start Initial condition + reading data from dictionaries A Sample Code, ADM1 algebraic routines Considering reactions Update variables Next time step source terms Momentum and transport equations Gases concentrations in headspace YES NO End time? Finish 11
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start Initial condition + reading data from dictionaries A Sample Code, ADM1 algebraic Using a switch function routines Considering reactions Update variables Next time step source terms Momentum and transport equations Gases concentrations in headspace YES NO End time? Finish 12
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start Initial condition + reading data from dictionaries A Sample Code, ADM1 algebraic routines Considering reactions Update variables Next time step source terms Momentum and transport equations Gases concentrations in headspace YES NO End time? Finish 13
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Start Initial condition + reading data from dictionaries A Sample Code, ADM1 algebraic Using FDM routines Considering reactions Update variables Next time step source terms Momentum and transport equations Gases concentrations in headspace YES NO End time? Finish 14
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results To verify the solver, the averaged concentration of each species is considered. A simple geometry with the following conditions is used • No of cells: 16 • Considering 2D solution • No flow (to replicate ADM1 bulk model) 15
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results The evolution of two species with different delta t and comparison with Oskar [2014] 16
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Digester geometry with a baffle inside Intermittent velocity Side view • No of cells: 5504 • Zero velocity at the walls • Using a velocity table for baffle Top view 17
mADM1 – Preliminary Anaerobic digesters ADM1 mADM1 – idea implementation Results Time (s) Velocity (m/s) 18
Conclusion • mADM1 first implementation is done • Ongoing Verification and Validation 19
Thank you This work is funded by FORTISSIMO 2 European Project. The FORTISSIMO 2 project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 680481 The University of Minho authors acknowledge the funding by FEDER funds through the COMPETE 2020 Programme and National Funds through FCT - Portuguese Foundation for Science and Technology under the project UID/CTM/50025/2013
Recommend
More recommend
