Hello, A yearly allocation for the LCRC cluster has been requested with the following updated information: Submitter/PI: Sibendu Som Project Name: CAT-engine-modeling Division: ES, MCS Project title: Simulation of Internal Combustion Engines with High-Performance Computing Tools Associated funding: DOE Office of vehicle technologies Other Systems: CATERPILLAR Cluster, CONVERGENT Science Cluster Science: The primary goal of this project is to study scaling issues using the commercial code CONVERGE with detailed chemistry. As a part of this exercise, complete 3-D simulation of a single cylinder diesel engine will be conducted. A series of chemistry models with increasing complexity (species and reactions) will be included. The objective is to find the optimum reaction mechanism and cores that will provide useful design information not obtained from 2-D and reduced chemistry models. Project description: The main objectives of this collaboration will be to identify and highlight the fundamental benefits of massive parallel computing capability on engine simulation. This will be accomplished by a very detailed and systematic study on the impact of both numerical parameters (such as grid resolution, complex geometries including inlet/exhaust valve motion, full 3-D simulations) and physical models (turbulence, combustion chemistry etc.). The impact of these numerical parameters and physical models on performance (scalability and wall-clock time) will also be investigated. This thorough study will enable the Argonne-CAT-CSI team to develop a set of modeling guidelines and best-practices for conducting comprehensive engine simulations. The impact of the choice of models and numerical parameters on the overall accuracy and required computational resources will be evaluated. Attempts will be made to quantify the advantages of using HPC over current modeling app roaches. In addition, efforts will also focus on correlating the insights gained by these advanced simulations to important design considerations such as fuel consumption, heat release rate, and emissions (NOx and soot). This team has already completed an initial collaborative study (phase 1) in which they demonstrated the value of using HPC in engine product design. The study was with a closed system, single cylinder engine geometry. The results led to establishing new “best engineering practices” at each of the partner’s institutions and has been accepted for publication with the Society of Automotive Engineers World Congress 2013, with all of the team members as co-authors. The key achievements of phase 1 include: (1) identification of a cell size which produced grid convergent engine results for many parameters of interest, (2) realization that using detailed chemistry of combustion is beneficial compared to using simplified global-step based combustion models, and (3) HPC provides insights into the combustion process which was not feasible earlier. After the completion of this first phase, the team is now ready to move into a second phase in which the approach will be to expand the control volume of the engine to include multiple cylinders, the intake air system, and the exhaust system. Due to the addition of this hardware, the wall-clock time and consequently the computational time is expected to increase significantly. The “best practices” identified in phase 1 will be retained for phase 2. The main objectives of phase 2 will be: (1) to show cylinder-to-cylinder performance variation (if any), (2) does phase 2 approach of open-cycle simulations provide invaluable insights compared to phase 1 approach of performing closed cycle simulations in terms of performance and emission characteristics?, (3) are the wall-clock times acceptable, (4) quantify cycle-to-cycle variations, and (5) identify the bottlenecks and gaps for engine simulations on HPC. An outline of the parameters that will be explored follows. All runs will use the Caterpillar C15 configuration and geometry of the full combustion chamber (360° cases) with intake and exhaust manifolds. 1) Finite volume schemes 2) LES and RANS turbulence models 3) Detailed chemical kinetic models 4) CONVERGE software is being used - ~ 2500 licenses available 5) Scales well up-to 256 processors 6) Detailed spray models used for different spray processes 7) Full cycle engine simulations Project URL: http://www.transportation.anl.gov/engines/multi_dim_model_home.html Current FY Hours Used: undetermined amount New FY Requested allocation: 499000 Q1: 125000 Q2: 125000 Q3: 125000 Q4: 124000 Justification: Thank You, The LCRC Accounts System