A computer code developed by a trio of Lawrence Livermore National Laboratory (LLNL) researchers has significantly advanced predictive computer science for designing next-generation car and truck engines.
The code, dubbed Zero-order Reaction Kinetics (Zero-RK) and known as an advanced solver, is an innovative computational method. In comparison to methods traditionally used for internal combustion engine research, it speeds up simulations of chemical systems by 1,000-fold. While calculating these complex combustion simulations with tens of thousands of chemical reactions once would have required years of computer run time, scientists can now do them in days with Zero-RK.
This advanced solver, developed by lab computational scientists Dan Flowers, Matthew McNenly and Russell Whitesides, with funding from the Department of Energy’s Vehicle Technologies Office, is now moving from the national laboratory into private industry. Convergent Science Inc. in Madison, Wisconsin has licensed the solver - for use in their Converge Computational Fluid Dynamics (CFD) software.
Through Convergent Science, a major diesel engine company and an automobile manufacturer are exploring how to best design engines using CFD with the latest version of the LLNL solver. One aim of the research is to develop engines that are high efficiency but produce little pollution. Instead of counting on high-temperature combustion control methods currently used, researchers designing next-generation engines are seeking to manage the ignition process through other means that will allow lower combustion temperatures and reduce exhaust emissions. As studying and developing these advanced technologies require analyzing tens of thousands of chemical reactions occurring during combustion, this new advanced solver should significantly speed up these partners’ research and development.