PFI Engine Simulation- Full Hydro
Introduction:
By definition, internal combustion engine (ICE) burn a mixture of fuel and air (oxygen), in an appropriate ratio, in order to produce mechanical power. In a four-stroke internal combustion engine, the combustion process occurs after the mixture of fuel-air has been induced into the cylinder, properly compressed and a spark generated (in case of a gasoline/petrol fuel).
The start of combustion process is linked to the position of the piston in the cylinder and usually happens before the piston hits the Top Dead Centre (TDC). As a result of the combustion process, the pressure generated in the cylinder rises and its transmitted as a torque to the crankshaft through the piston and connecting rod.
Combustion can be defined as a rapid oxidation process of a fuel, which generates heat and light. With other words, combustion converts the energy stored in the chemical bonds of the fuel into heat. In an engine, combustion occurs as a flame that propagates in the cylinder. A flame (fire) occurs if three components are present: fuel, oxidizing agent (oxigen) and heat. Depending on the type of the fuel, the heat for the occurrence of the flame comes from: a spark (gasoline/petrol engines) or heated (compressed) air (diesel engine).
The main product of an internal combustion engine is mechanical power, this is why we use them as propulsion systems. Together with mechanical power, as a by-product, we also get exhaust gases. Ideally, if the combustion process is complete, the exhaust gases should only be carbon dioxide (CO2) and water vapour (H2O). In reality, mainly due to incomplete combustion, the exhaust gases also contain pollutant emissions: oxides of nitrogen (NOx), unburnt hydrocarbons (HC), carbon monoxide (CO), soots/particles (PM), polyaromatics, aldehydes ketones and nitro-olefins.
Consider an exmaple of of Methane as Fuel:
The above chemical equation is a complete combustion of methane in presence of oxygen, giving water and carbon dioxide as by-products.
Surface Preparation:
First step was to import the geometry and look for errors present in the STL file.
The next step involves flagging the bundaries of the geometry. This step is important as it allows the user to provide different boundary and initial conditions to the flagged boundaries.
Now, since the file had a lot of intersection errors, it was difficult to flag boundaries seperately. A tool called Fence was used to create a seperation/ fence between teh boundaries so that they can be flagged as different entity.
Another tool which was handy to remove most of the intersection errors was Translate. While observing the errrors in the geometry, it was found that the intake valve was placed much higher, causing a lot of interaction with the cylinder head. With the help of Translate tool, the intake valve was translated down leading to disappearance of all the intersection errors. Following diagram provides the direction to use Translate tool.
Now, the time was to target other errors present in the geometry: non-manifold and open edges.
Following figure provides an understanding of open edge present in the geometry. Since this is an internal problem, it is necessary to make sure all the edges are filled so as to create a closed volume. The brown area present on the piston surface indicates present of open edges, ehich are filled using patch tool.
Once all the errors were removed it was time for teh case setup.
Case Setup:
IC engine ( crank angle based) solver was used for the simulation. Following are the physical parameters of teh engine.
Using the mech.dat and therm.dat, all the species and their thermo-physical properties were imported into the simulation.
Following figure provides the runtime for the simulation.
It is important to create different regions for the entire engine, which allows us to input, perform various operations on those regions seperately. Also, since the type and amount of species are different in these regions, it makes sense to create them for smooth functioning of the simulation.
Since, intake and exhaust valve operate only for a particualr time during the combustion cycle, it is necessary to create events for valves, so that they can operate for a particualr time during the cycle. File like intakelift.in and exhaustlift,in are imported into the setup to provide the required movement for the valve.
Now, it's time to decide the base grid for the simulation and use grid control techniques like: adaptive mesh refinement and fixed embedding to generate fine mesh.
Adaptive mesh refinement, as the name suggests, adapts according to the current value of the prescribed variables. The mesh refines itself if the gradient of provided variable changes beyond the prescribed limit.
Fixed embedding grid control technique is used to modify the mesh size in critical areas so as to get an accurate result.
Following image provides an example how cylindrical embedding is used for combustion chamber. It is neccesary to provide mesh refinement in the combustion chamber, since all the reaction happens in this area. Through grid control, it would be possible to correctly simulate the combustion analysis.
Simulation:
Now, Converge software is used to simulate PFI engine various parameters and efficiency of the engine. To simulate the engine properly various models have been introduced in the simualtion so as to replicate the actual combustion cycle computationaly.
Following Models were used during the simulation:
1. Spray Model:
For Liquid spray Converge uses Lagrangian solver to model discrete parcels and the Eulerian solver to model continuous fluid domain.
Heat, momentum and mass transfer occur between the discrete and the continuous phases with the help of souce term in the transport equation.
Converge introduces parcels into the domain at the injector. Each parcel represents identical drops (radius, velocity, temperature) and as a result converge solves for per-parcel basis rather than per-drop basis. These parcels statiscally represent entire spray field.
Also, spray modeling uses subgrid models to model various physics of fuel spray after it interacts with the air in the chamber.
The interaction of fuel parcels with the air results in breakup of the parcels and leads to various physical interaction provided in the above diagram. Sub grid model takes care of these interactions so as to provide accurate results closely matching the experimental data.
- Injector:
2. Combustion Modeling:
Converge contains detailed chemistry solver and simplified combustion model to provide the result for combustion. The detailed chemistry solver provides accurate way to compute combustion which can be easily compared with experimental data. Due to this the computational time taken by detailed chemistry is more as compared to simplified combustion model.
- Sage Solver:
It uses the local condition to calculate reaction rates based on principles of chemical kinetics. Each mesh cell present inside the engine has a unique value for various thermo-physical properties, which the SAGE solver uses to calculate the reaction rates. The solver uses Arrhenius rate law equation to solve the rate of change of species concentration.
3. Turbulence Modeling:
RNG K-e model is used for turbulence modeling.
Now, Solutions to the given questions in the challenge:
1. Compression Ratio:
Defined as the ratio of maximum to minimum volume in the cylinder of internal combustion engine.
Maximum Volume: 5.74125*10e-4 m^3.
Minimum Volume: 5.71859*10e-5 m^3.
Therefore, Compression Ratio: 10.04
2. Efficiency of Engine:
Efficiency= Output Energy/ Input Energy.
Input Energy= mass of fuel* Calorific Value of Fuel= 3*10e-5 (kg) * 45(MJ/kg)
Output Energy= Integrated Heat released =1240 J.
Efficiency= 91.85%
3. Power and Torque Calculation:
Work= 468.646 N*m
Duration= 240.199 deg
Crank angle speed= 3000 rpm.
Time per 240.199 deg= 0.01334 sec.
Power= 468.646/ 0.01334 =35.130 KW.
Torque= 111.8 Nm.
4. Significance of Ca10, Ca50 & Ca90:
Ca10 - crak angle degree for 10% of cumulative combustion heat release- 6.83
Ca50 - crak angle degree for 50% of cumulative combustion heat release-18.462
Ca90 - crak angle degree for 90% of cumulative combustion heat release-31.701
Graphs:
1. Pressure Plot:
2. Temperature:
Temperature profile are used to determine the emissions from, combustion of fuel. The temperature attained during combustion, dictates the type and amount of emission from the chamber. A high value of temperature leads to formation of Nox, a low value of temperature leads to failed combustion. So, this graph allows to understand the reaction process that might happen inside the chamber.
3. Integrated Heat Release:
Integrated heat release value is used to determine the combustion efficiency of the engine. This value provides the output energy of the combustion, while the input energy is determined using mass of fuel and calorific value of fuel.
4. Heat Release Rate:
5. Emissions:
Timing Map:
It provides an overview of the entire combustion process.
PFI Temperature Contour: