HCCI Engine Optimization with CONVERGE CFD

Julian Toumey

Just this week, Mazda Motor Corporation announced plans to introduce a gasoline compression ignition engine technology deemed SKYACTIV-X into commercial production engines in 2019. Mazda claims that this will be the world’s first commercial gasoline compression ignition engine.

Gasoline compression ignition, or homogeneous charge compression ignition (HCCI), applies combustion techniques previously used in diesel engines to gasoline engines. In a diesel engine, fuel and air are compressed in the cylinder to the point at which the pressure and temperature are high enough for the charge to ignite. Advantages of this approach include higher efficiency due to a higher compression ratio and lower emissions due to lower combustion temperatures.

In a typical gasoline engine, fuel is injected into an intake manifold or directly into the cylinder. A spark plug fires to ignite the fuel-air charge.

Simulation of spark-ignited (SI) combustion inside of a cylinder.

As with diesel engines, Mazda’s SKYACTIV-X HCCI engines rely on compression of the charge for ignition but still use gasoline as the fuel. The result is an engine that is 20 to 30% more efficient than a spark-ignited gasoline direct injection engine. Also, because combustion temperatures are lower than those of typical spark-ignited gasoline engines, emissions are likewise reduced. SKYACTIV-X engines still contain a spark plug for spark ignition when conditions are not ideal for effective HCCI.

Many obstacles stand in the way of producing a commercially viable HCCI engine. For one, the fuel and air must be thoroughly mixed in the combustion chamber to ensure even combustion. Also, combustion timing must be carefully controlled to prevent undue wear on the engine.

To address these obstacles, engineers at OEMs and researchers at institutes such as Argonne National Laboratory rely on computational tools to simulate HCCI engines. These tools allow engine designers to investigate a wide array of operating conditions, designs, fuel mixtures, and more. Many of these design conditions are too expensive or time consuming to test with experimental setups.

CONVERGE, Convergent Science’s flagship computational fluid dynamics software, is a numerical tool commonly used for such investigations. CONVERGE generates the computational mesh (discretized representation of the engine) at runtime, drastically reducing the pre-processing time required for running these engine simulations. The result is that engineers can focus more on improving the engine design and less on the simulation setup.

Simulation of ignition timing optimization in GCI combustion. Video courtesy of the Virtual Engine Research Institute and Fuels Initiative (VERIFI) at Argonne National Laboratory.

An advantage of autonomous meshing is that CONVERGE determines where and when to refine the mesh and will do so on the fly via Adaptive Mesh Refinement. HCCI engine designers rely on CONVERGE to automatically capture relevant flow features during the simulation.

One significant challenge for simulating HCCI engines is accurately modeling combustion. CONVERGE includes the ability to simulate detailed chemistry and accurately capture the combustion in the cylinder. This allows engineers to ensure even combustion in their production engines.

Dr. Sibendu Som, Group Leader and Principal Computational Scientist at Argonne National Laboratory, says, “The ease of mesh generation, Adaptive Mesh Refinement, and advanced combustion models in CONVERGE, together with high-performance computing systems, enable development and optimization of new combustion concepts such as gasoline compression ignition.”

These advantages add up to the ability to rapidly optimize design parameters in HCCI engines, which, if Mazda’s announcement is any indication, might be the next step in commercial engine design.