Case-study of heavy-duty engine for Indian Bharat stage IV and VI

Transport of material within the country is very crucial for the economy and maintaining the inflation in check. The increase in the cost of fuel has a direct impact on the material transported such as farm products, raw material for construction, mining industry etc. Efficient prime movers, less maintenance, no downtimes are some of the main requirements of the fleet owners.

Achieving upcoming BS-VI emission norms and satisfying customer demands is a very challenging task for the OEM's. Which strategy will work? High engine out emissions to maintain simplified engine configurations or reduced loading on Aftertreatment devices using advanced combustion strategies?

PROBLEM STATEMENT: To develop a commercial application on-road application engine with good fuel efficiency

PRELIMINARY DATA:

Engine parameters	Value
Engine Type	Diesel Engine
Engine Displacement [cc]	5 L
No. of Cylinders	4
Maximum Power	210HP @ 2200 RPM
Maximum Torque	825 NM @ 1400-1800 RPM
BSFC	230 g/KWhr
Fuel system	DI - Diesel System ; CRDI
Turbocharger	Single stage
Emission Norms	BS IV
AFT system	SCR with Urea Injection System

TARGET DATA:

Engine parameters	Value
Engine Type	Diesel Engine
Engine Displacement [cc]	5 L
Maximum Power	210HP @ 2200 RPM
Maximum Torque	825 NM @ 1400-1800 RPM
BSFC	210 g/KWhr
Emission Norms	BS VI
AFT system	DOC-DPF-SCR

METHODOLOGY:

1. Simulation & Analysis team : Various baseline test simulations are done according to the BS IV stage. Come up with suitable geometrical specifications such as Bore, Stroke, Compression Ratio, Expansion Ratio, Specific power etc..
2. Combustion development team : In cylinder strategies to help minimize emissions and control bsfc
3. Exhaust Aftertreatment development team : Further control of Emissions for a BS VI compliant design
4. Alternate Fuel team : Exploration of alternative fuels and latest technologies to help aid the cause. [Report only]
5. Advanced technology team : New technologies that are being researched that can be implemented [Report only]

ROAD MAP: Phase 1 - BS IV Performance ; Phase 2 - BS IV AFT ; Phase 3 - BS VI Performance ; Phase 4 - BS VI AFT

1. Start engine modeling using GT Power for Max Power and Max Torque on baseline configurations (Geometric) for Phase 1 using the FRM builder approach. Filter them out on the performance targets.
2. Use Phase 1 results to further fine tune the model and proceed for combustion development by calibrating the necessary parameters for Phase 2 [Refer HINT given below]
3. Comment on calibrations on models used such as : Combustion model ; Heat Transfer Model ; Friction Model ; Emission Model to finally arrive at Phase 3
4. List of pollutants and strategy to control emissions for Phase 4

The number of operating cars in India is about 253 million, of which 10 million are trucks, which represent the heart of the Indian economy. It's always harder to respect the latest pollutant regulations that are pushing the vehicle industries adopting new solutions, to not receive a fine that could be reflected in the economic world as an increase in the price of materials. For this reason, this case-study is based on the modeling and simulation of a heavy-duty engine, starting from the Bharat stage IV to stage VI, trying to achieve the pollutant targets imposed by the legislation. The following study is divided into 4 phases, explaining and showing step-by-step the modeling of the engine and the tuning of the most sensitive parameters.

PHASE I

Phase I consists of defining all the main parameters that characterize the engine. The preliminary data tab (shown below) gives different information about the engine:

The engine proposed is a CI engine, SOHC, with 5 L of total displacement, 4-cylinders, single-stage, and a twin-scroll turbocharger, with common-rail direct injection, maximum power of 156.5 kW @ 2200 rpm, and maximum torque of 825 Nm @ 1400-1800 rpm, with BSFC of 230 g/kWh. So, a first analysis consists of understanding what is the best configuration for the engine in terms of bore, stroke, stroke-bore ratio, and compression ratio, to reach the required performance (max power and torque) using at the beginning a simple fast running model.

To calculate bore and stroke the single displacement equation and stroke-bore ratio are used:

V=π⋅C⋅D24">V=π⋅C⋅D24V=π⋅C⋅D24

CD=x">CD=xCD=x

Knowing the single displacement of the engine:

V=Vt4=5000cm34=1250cm3">V=Vt4=5000cm34=1250cm3V=Vt4=5000cm34=1250cm3

the parameter that is necessary to define is the stroke/bore ratio. This one represents an important characteristic of the engine, that can give some important information on the type of the engine. Basing on this parameter, engines can be classified into three categories:

- Oversquare engine: C/D < 1

This type of engine is generally used for sportive vehicles. It guarantees different advantages and disadvantages: larger valves in the head of the cylinder (increasing volumetric efficiency), higher possible rpm due to the smaller stroke, a minor piston and head surface area that increases heat loss and a minor thermal efficiency compared to other engine geometries. The large size/width of the combustion chamber at ignition can cause increased inhomogeneity in the air/fuel mixture during combustion, resulting in higher emissions.

- Square engine: C/D = 1

- Undersquare engine: C/D >1

This type of engine is preferred when high thermal efficiency and low formation of pollutants is required. It guarantees peak torque at lower rpm than an oversquare engine but on the other side, a longer stroke increases engine friction, and the smaller bore also reduces the area available for valves in the cylinder head, requiring them to be smaller or fewer in number. Generally, city cars are equipped with this engine but it is even possible to find modern sports cars that are moved by an under square engine, such as Lamborghini Huracan with C/D=1.09.

So in this panorama, it is hard to choose immediately which one is the best configuration for the engine. For this reason, different values are considered, choosing 11 C/D ratio and so obtaining different bore and stroke values, as described into the tab below:

C/D ratio	0.9	0.93	0.95	0.98	0.99	1	1.03	1.05	1.08	1.09	1.1
Stroke [mm]	108.835	111.240	112.829	115.192	115.975	116.754	119.078	120.615	122.901	123.659	124.414
Bore [mm]	120.928	119.613	118.768	117.543	117.146	116.754	115.610	114.871	113.797	113.448	113.103

To calculate rapidly all these values a simple MatLab script has been implemented:

The other fundamental parameter that must be defined is the compression ratio. For a CI engine, the CR values are placed in the range [14 - 22]. Even in this case is necessary to run different simulations, for every couple of bore and stroke values, and finding the best for this application.

The total number of simulations is: 11 (all the stroke-bore data) x 9 (all the CR values considered) = 99 simulations.

Let's create the fast-running model:

In the picture above is shown the FRM created in GTP. It consists of:

1) intake system:

• "EndEnvironment" template;
• IntakePipe1;
• AirBox;
• IntakePipe2.

2) Turbocharger system:

• Compressor coupled on the same shaft with a turbine;
• Heat exchanger.

3) Engine:

• Intake manifold;
• Valves;
• Cylinders;
• Injectors;
• Exhaust manifold;
• Engine.

4) EGR circuit;

5) Exhaust system;

and now they are analyzed to understand what are the preliminary data automatically generated by GTP.

1) Intake system

The intake system is composed of few elements:

A) "EndEnvironment" template;

B) IntakePipe1;

C) AirBox;

D) IntakePipe2.

Let's analyze every single element.

A) The "EndEnvironment" template is used to define the conditions of the air in the ambient. Only the "main" tab is shown because the others are default:

the composition of the air is defined in "air-NASA", considering only a fraction of N2 and O2:

B) The "intakePipe1" defines a single pipe that allows the air entering into the next element:

As visible, the value of discretization length is very high, due to the fact that this is an FRM, so the speed of the simulation is preferred over higher accuracy results. The "initial state name" is referred to the initial condition of the air at the inlet, as defined into "EndEnvironment" template, and the wall temperature is set at 300 K.

C) After that, the flow enters into a big volume, called "Airbox". It is a "flow-split" template with one entry and one exit. Its volume and surface area are defined as shown below:

Inot "thermal" tab is defined: the wall temperature, set on 300 K, the "heat transfer correlation", set on Colburn, and the "flow split HTC calculation method".

The heat transfer from fluids inside of pipes and flow split to their walls is calculated using a heat transfer coefficient. The heat transfer coefficient is calculated at every timestep from the fluid velocity, the thermo-physical properties, and the wall surface roughness. The Colburn analogy is the one shown below:

The HTC calculation method can be chosen in the same tab. The "Method #2" of calculating the heat transfer coefficient in a flow split uses the boundary velocities to calculate a heat transfer coefficient for each port, which can be based on the expansion diameter or a custom dimension. This method produces a more accurate prediction for the heat transfer coefficient and is recommended especially in applications where the heat transfer in the flow split is a significant part of the overall circuit heat transfer rate. Then a "characteristic dimension port for Port HTC" is set: this attribute will only be used when Method #2 is selected. This attribute selects which value will be used as the diameter in the heat transfer coefficient calculation, and the "expansion diameter" has been chosen.

After that, is important to focus on the "Boundary data" tab:

in this tab is possible to define the spatial position of the ports of the flow split. But is very important to focus on the following two parameters: "Characteristic length" and "Expansion diameter".

The expansion diameter defines the maximum diameter that the flow may expand to after entering a boundary (or contract from when leaving a boundary). It is a function of the shape of the flow split but is completely independent of the attached pipe or orifice. Generally, this parameter is used to determinate the following things:

- The pressure loss due to contraction and expansion

- The kinetic energy loss due to expansion for determining pressure recovery

- The change in fluid properties due to changes in velocity resulting from a change in area, such as temperature, pressure, density, and speed of sound. This is important when modeling a compressible fluid and/or wave dynamics.

In GTP manuals there are some tips to how to choose this value, but is necessary to understand that this parameter must be calibrated, and so is necessary to start with appropriates default parameters and then when calibration data are available, the expansion diameter can be adjusted to calibrate pressure drop in the flow system.

The characteristic length defines the distance the flow entering the boundary will travel before crossing another boundary or impacting a surface. The characteristic length is typically defined as the user-chosen length of the flow split, ideally the discretization length.

In the picture below these two parameters are shown on the airbox:

D) The last element of the intake system is another pipe round, with a smaller diameter than the first one:

2) Turbocharger system

The turbocharger system allows to compress the air before entering into cylinders, using exhaust gases, increasing the volumetric efficiency of the engine and so its power. The system shown in the picture above is made up of the following elements:

A) Compressor coupled on the same shaft with a turbine;

B) heat exchanger.

Let's analyze every single element and understanding how they are modeled.

A) Compressor and turbine

A compressor is used to compress the air before entering into cylinders. The most important aspect of the turbocharger is that this turbomachinery is moved by a turbine, and the entire system doesn't increase engine losses because the turbine is moved by exhaust gases and is coupled on the same shaft with the compressor. Because of the irreversibility of the turbomachinery, the air at the exit of the compressor has a higher temperature than the external conditions, decreasing its density and increasing the specific volume, causing a less air flow rate entering into cylinders. For this reason, a heat exchanger, called intercooler, is used before the intake manifolds. As known, a heat exchanger allows an exchange of heat between two fluids, and ideally is possible to consider this transformation isobaric.

Compressor

This template allows modeling a compressor basing on a map object called "CompressorMap". This map template defines the compressor performances basing on a single operating point and generates a fully extrapolated map using a set of physically-based parametric equations. This single operating point is assumed to be the peak efficiency point on the compressor map. In phase III, during tuning of the engine, all the compressor map parameters will be discussed.

Turbine

The modeled turbine is a twin-scroll, single-stage turbine with a wastegate controller. The map object used is called "TurbineMapParam" that works in the same way as the previous compressor map object. In phase III all the turbine map parameters will be discussed.

B) Heat exchanger

The heat exchanger allows cooling the air before entering into cylinders. It has been modeled controlling the imposed fluid temperature and the orifice diameter.

The imposed fluid temperature is displayed green because it is a parameter controlled by another element: IC_EFF. Below are shown the various parameters needed to calculate the imposed fluid temperature:

the math equation for the calculation is defined into the template called: "TOutCalc-1":

the parameters that it needs to calculate the imposed fluid temperature are:

Thi: the temperature of the air after the compressor;

Tci: the temperature oft he coolant fluid;

Eff: the efficiency of the heat exchanger.

3) Engine

The engine sub-assembly is shown in the picture below:

This system is composed of the following elements:

A) Intake manifold;

B) Valves;

C) Cylinders;

D) Injectors;

E) Exhaust manifold;

F) Engine.

Let's analyze every single element and understanding how they are modeled.

A) Intake manifold

Upstream the intake manifold there is a throttle. Generally, for a CI engine, the power is directly controlled regulating the fuel pump, and so the amount of fuel injected for every cycle, while the inlet air is a function of the rpm of the engine, but for a CI turbocharged engine, to have better control on the power, is possible to find a throttle. The intake manifold has two entries and four exits: the first entry is directly linked with the heat exchanger (intercooler) and so allows the compressed air to enter in this manifold; the other one entry is directly linked with the EGR, and so allows the recirculation of exhaust gases.

The intake manifold is modeled using a flow split template, and so it is a big volume, with an imposed wall temperature set on 330 K (due to the fact that it is nearer the engine, and so the temperature is higher).

B) Valves

The engine modeled is a SOHC so it has only one intake valve and one exhaust valve. This configuration could penalize volumetric efficiency, so during the tuning phase, a DOHC configuration and all its parameters will be discussed.

C) Cylinders

The cylinder template allows defining a lot of parameters about the combustion model, heat transfer, and more.

In the main tab shown above, is possible to define:

the initial state object, so the object that defines the conditions of the Intel air into the intake system;

the wall temperature is defined by a reference object called "TWall", that allows defining accurately the temperature of different part inside cylinder:

The heat transfer object used in this application is the ‘EngCylHeatTr’ that allows defining which heat transfer model to use to calculate the in-cylinder heat transfer coefficient.

The heat transfer model used for this engine is the WoschniGT: it indicates that the in-cylinder heat transfer is calculated by a formula that closely emulates the classical Woschni correlation without swirl, given below. The most important difference lies in the treatment of heat transfer coefficients during the period when the valves are open, where the heat transfer is increased by inflow velocities through the intake valves and also by backflow through the exhaust valves. This option is recommended when measured swirl data is not available.

The convective heat transfer coefficient for the Woschni models is defined as follows:

The flow object defines flow characteristics in this cylinder. This reference object must be used by cylinders that use the predictive Diesel combustion models, such as: ' EngCylCombDIPulse'. In this preliminary analysis, the combustion model used is non-predictive. A non-predictive combustion model simply imposes a burn rate as a function of crank angle. This prescribed burn rate will be followed regardless of the conditions in the cylinder, assuming that there is sufficient fuel available in the cylinder to support the burn rate. During the tuning of the model semi-predictive and predictive models will be discussed. The other tabs at the moment are default, so they are not shown.

D) Injectors

The injectors used in this preliminary performances analysis are modeled with the template called "InjDieselSimpleConn":

The injected mass is controlled directly from the "ControllerDinInj", which injects the right quantity of fuel to reach a target. In the picture below is possible to see that the target defined is the brake torque (its value will be discussed later, after the analysis of the FRM):

and the only limitation defined is about the minimum A/F ratio:

E) Exhaust manifold

The exhaust manifold is directly linked to the exhaust valves, to the turbine and to the EGR circuit. As visible in the picture above of the entire engine system, the exhaust gases are sent into two different pipes before entering into the turbine, to decrease losses due to the disturbs that can propagate into the ducts, because of the cylinders are out of phase.

F) Engine

This part of the model allows describing the geometric characteristics of the cylinder. The values of bore, stroke, rod length, and compression ratio are generic at the moment because it is necessary to complete the 99 simulations (described before) to understand the best configuration for this application. In the main tab of this template the friction object is defined and even the moment when the cycle starts:

4) EGR circuit

The EGR circuit allows recirculating and cooling part of the exhaust gases, to decrease the maximum temperature inside the combustion chamber and so to decrease the formation of NOx. This circuit is composed of two main pipes, an EGRVolume, and a heat exchanger. The most important part of this circuit is the heat exchanger: it has been modeled using the same template for the intercooler, but now there are different values of the parameters:

The diameter is imposed on default because it is directly actuated by an external actuator, while the imposed fluid temperature is controlled by EGR_Eff, in the same way as the IC_Eff for the intercooler. During the tuning of the engine different parameters will be discussed and their impact on the pollutant formations and combustion efficiency.

5) Exhaust system

The last part of the model is composed only of "EndEnvironment", which describes the external condition of the air at the exit of the entire engine, and a volume created by flowsplit template, shown below:

Now that the entire model is described, it is possible to start the simulations to understand what is the best configuration in terms of bore, stroke, and compression ratio for this engine. At first is necessary to evaluate what are the performance targets that must be reached. There are two different cases: maximum torque and maximum power:

MAX TORQUE CONDITION (@1600 rpm)

P=Mmax⋅ω=Mmax⋅2⋅π⋅n60=138.23kW">P=Mmax⋅ω=Mmax⋅2⋅π⋅n60=138.23kWP=Mmax⋅ω=Mmax⋅2⋅π⋅n60=138.23kW

Mmax=825Nm">Mmax=825NmMmax=825Nm

BSFC=230gkWh">BSFC=230gkWhBSFC=230gkWh

mgcyl=165.613">mgcyl=165.613mgcyl=165.613

MAX POWER CONDITION (@2200 rpm)

Pmax=156.6kW">Pmax=156.6kWPmax=156.6kW

M=Pmax2⋅π⋅n⋅60=679.3Nm">M=Pmax2⋅π⋅n⋅60=679.3NmM=Pmax2⋅π⋅n⋅60=679.3Nm

BSFC=230gkWh">BSFC=230gkWhBSFC=230gkWh

mgcyl=136.369">mgcyl=136.369mgcyl=136.369

As visible from the calculations, the same BSFC has been chosen for both conditions, and knowing from the preliminary data tab that maximum torque is in the range of 1400 - 1800 rpm, the max torque condition has been set at 1600 rpm.

In "case setup" of GTP the two cases are set:

The wastegate is controlled by the bmep target, set at 20 bar for the max condition simulation, and the boost pressure, set at 2.5 bar.

This first simulation is necessary to understand which one is the best configuration in terms of bore, stroke, and compression ratio. As seen before, there are 99 simulations, due to different combinations, that are simulated implementing a DOE. At first, case setup parameters are created, and the connecting rod length is set as two times the stroke:

In case setup, at max torque condition, all the bore and stroke values shown at the beginning of the project are set:

and into Design of Experiments tab, all the compression ratio are defined:

RESULTS

All the simulations have been completed in 26 minutes, using a Dell Alienware (2019) laptop, with 16 Gb ram and i9 processor:

Every single simulation will be analyzed below, showing different aspects of the engine for every couple of bore and stroke values at different compression ratios. The main aspects visualized are:

- max pressure inside cylinder: if two engines give the same power, it's maybe preferable choosing the engine that shows a lower peak pressure inside the cylinder. In this way is possible to suppose that the head of the piston, the connecting rod, and even the crankshaft are subjected to a lower force, and so it is possible to retain that these elements could be produced with a less expensive technique (so a less expansive engine), less heavy, and lasting longer. All these aspects have benefits on the total price of the engine, but even on the total fuel consumption, in fact, one of the main parameters that influence fuel consumption is the weight of the vehicle. So, reducing its weight is possible to appreciate a minor consumption by the engine;

- max temperature inside cylinder: as the previous considerations for the pressure, is preferable an engine with a lower peak of temperature. In this way, all the components inside the engine are subjected to a less thermal load, which gives a longer life to the component. It could be even seen from the cooling system perspective, that requires less coolant fluid and so a less sophisticated system, with less expensive component and a less total weight of the engine. But one of the most important aspects linked to the peak of the temperature inside the cylinder is the NOx formation: to limit this pollutant family is necessary to decrease the peak of the temperature inside the cylinder;

- BMEP: the brake mean effective pressure can be evaluated by the following equation:

BMEP=4⋅π⋅MV">BMEP=4⋅π⋅MVBMEP=4⋅π⋅MV.

So, this equation shows that the BMEP is directly proportional to the torque and inversely proportional to the total displacement of the engine. All the simulations conducted for the engine considered always the same total displacement, but a different combination of stroke, bore, and compression ratio. For this reason is preferable choosing the simulation with the highest BMEP, because it means that the engine can work better in that configuration because, with the same total displacement, it can give more torque;

- BSFC: the brake specific fuel consumption defines how much grams of fuel are necessary to generate one kW-h. This is a very important term, that is directly linked to the fuel consumption of the vehicle. The configuration with the lowest bsfc value should be chosen;

- air-fuel ratio: this represents the ratio between airflow rate and fuel flow rate that enter inside the cylinder. The Diesel engines are characterized by the fact that they can work with a very lean mixture. A typical value of this ratio is set in the range 18-22;

- Stroke/Bore ratio: this ration, as discussed at the beginning of this project, is very important. Having a big bore allows different important things, such as a bigger space to locate valves and the possibility to use two inlet valve and two exhaust valves, increasing the "breathing" of the engine, so its volumetric efficiency; a smaller stroke allows reaching higher rpm; oversquare engines are preferred for sports vehicles; oversquare engines can produce more HC pollutants due to the presence of more "crevices", zones where the fuel is not reached by the flame front; undersquare and square engines are preferred to reach a higher thermal efficiency;

- compression ratio: this is another fundamental parameter. It is defined as: CR=V+VoV">CR=V+VoVCR=V+VoV the ratio between the volume of the cylinder when the piston is at the BDC and the volume of the cylinder when the piston is at the TDC. Considering an ideal cycle, called Sabathè, the ideal efficiency can be evaluated with the following equation:

ηlim=1-1ρk-1">ηlim=1−1ρk−1ηlim=1-1ρk-1

so, increasing the compression ratio is possible to appreciate an increase in efficiency. For this reason, it is preferable choosing an engine with higher CR. But is necessary even to understand that engines with higher CR are more expansive because of the bigger precision needed.

After comparing all the results for every simulation, the best one has been chosen for every single couple of bore and stroke values, as shown below:

S/B	0.9	0.93	0.95	0.98	0.99	1.0	1.03	1.05	1.08	1.09	1.1
CR	19	19	18	18	18	18	19	19	19	19	19
BMEP [bar]	21.7	21.7	21.6	21.6	21.6	21.6	21.6	21.6	21.6	21.6	21.6
IMEP [bar]	23.37	23.37	23.25	23.24	23.25	23.24	23.36	23.35	23.35	23.34	23.35
BSFC [g/kWh]	207	207	207	207	207	207	207	207	207	207	207
Max Temp. [K]	1990	1990	2000	2000	2000	2000	2000	2000	2000	2000	2000
Max. Press. [bar]	147	147	138	138	138	138	147	147	147	147	147
Mean Piston Vel [m/s]	5.8	5.9	6	6.1	6.2	6.2	6.4	6.4	6.6	6.6	6.6

Basing on the previously discussed parameters, the best configuration seems to be the one highlighted.

But there are other two geometric parameters that need to be discussed:

- TDC clearance height: that value represents the cylinder clearance height from the top of the piston to the top of the cylinder wall when the piston is at TDC;

- Wrist Pin to Crank Offset: that represents the wrist pin offset relative to the crankshaft axis when the wrist-pin bearing position on the piston end is projected toward the crankshaft on a line parallel to the cylinder axis.

These two parameters can impact on different other values such as heat transfer inside the cylinder and mechanical loss.

To evaluate the impact of the single parameter on the engine's performance, other two DOEs have been implemented.

The Wrist Pin to Crank Offset DOE is shown in the picture below:

Obtaining the following results:

The picture above shows different aspects due to the variation of the wrist pin to the crank offset. The best compromise in terms of pumping mean effective pressure, friction mean effective pressure, piston pin side forces and brake torque is with a value of 0.25. So it means that the wrist pin is located 0.25 far from the crank offset, as shown in the picture below:

After the evaluation of the previous parameter, another DOE has been set, to understand the best choice in terms of TDC clearance height:

obtaining the following results, summarized in this diagram:

The best TDC clearance height values are 0.125 mm and 0.5 mm. The difference in terms of brake torque, BMEP, IMEP, BSFC, is very small (around one point), but considering even the production costs the best choice is 0.5 mm, which allows using a less expansive production technique.

Now the geometrical characteristics of the engine are defined. The only other value that must be considered is the target boost pressure, for both cases.

Another DOE has been set:

After completing all the simulations, the entire geometry of the engine was set, showing below the geometrical aspects, the boost pressure and the bmep target chosen (controlled by the wastegate):

Stroke	119.078
Bore	115.61
Stroke/Bore ratio	1.03
Connecting rod length	238.156
TDC clearance height	0.5
Wrist pin crank offset	0.25

rpm	1600	2200
Boost pressure [bar]	2.4	2.03
BMEP target [bar]	20	18

Before ending Phase I, let's see the percentage difference between the values obtained by the simulation and the target values:

MAX TORQUE CONDITION

Brake Torque: 825.8-825825⋅100=0.09%">825.8−825825⋅100=0.09%825.8-825825⋅100=0.09%

BSFC: 207-230230⋅100=-10%">207−230230⋅100=−10%207-230230⋅100=-10%

Fuel: 28.7-31.7931.79⋅100=-9.72%">28.7−31.7931.79⋅100=−9.72%28.7-31.7931.79⋅100=-9.72%

MAX POWER CONDITION

Brake Torque: 679-679.3679.3⋅100=-0.04%">679−679.3679.3⋅100=−0.04%679-679.3679.3⋅100=-0.04%

BSFC: 213.8-230230⋅100=-7.04%">213.8−230230⋅100=−7.04%213.8-230230⋅100=-7.04%

Fuel: 33.3-36.01836.018⋅100=-7.54%">33.3−36.01836.018⋅100=−7.54%33.3-36.01836.018⋅100=-7.54%

All the percentage differences between target values and values obtained by the simulations are under 15%. So it is reasonable to retain that the target values have been reached.

PHASE II

After defining the main geometrical aspects of the engine, in phase II is necessary starting to face off the main problems of every vehicle today: pollutants and after-treatment systems.

The "Bharat stage emission standards" are emission standards defined by the Government of India, to regulate the max quantity of pollutants generated by compression and spark-ignition engines. Different and different regulations have been emanated starting from 1992 till 2020. In 2010 the "Bharat stage IV" was defined but it was applied only in certain regions of India. It regulated the max quantity of the main pollutants, as described in the picture below (referred to the heavy-duty engines):

From Bharat stage III there are two different tests which pollutants are measured:

- ESC (European stationary cycle): was introduced together with the ETC and ELR test for emission measurement from heavy-duty diesel engines. The ESC is a 13-mode, steady-state procedure that replaced the older R-49 test. The engine is tested on an engine dynamometer over a sequence of steady-state modes. The engine must be operated for the prescribed time in each mode, completing engine speed and load changes in the first 20 seconds. The specified speed shall be held to within ±50 rpm and the specified torque shall be held to within ±2% of the maximum torque at the test speed. Emissions are measured during each mode and averaged over the cycle using a set of weighting factors. The final emission results are expressed in g/kWh.

- ETC (European transient cycle): different driving conditions are represented by three parts of the ETC cycle, including urban, rural, and motorway driving. The duration of the entire cycle is 1800s. The duration of each part is 600s. Below, a brief description of every part is reported:

1) Part one represents city driving with a maximum speed of 50 km/h, frequent starts, stops and idling.

2) Part two is rural driving starting with a steep acceleration segment. The average speed is about 72 km/h

3) Part three is motorway driving with an average speed of about 88 km/h.

To compare the GTP result with the emission standards is necessary to take in consideration only the value obtained with the ESC test.

Before starting comparing the values obtained by the model and the targets that must be respected, it is necessary to understand the pollutants, where and how they can born inside a CI engine. In the picture below is represented a spray of fuel, highlighting the main areas where the pollutants born:

- CO: the monoxide-carbon is a product given by partial oxidation. During the combustion of the fuel inside the cylinder, the main reaction for the CO production is the following one: CO+OH ↔CO2+H">CO+OH ↔CO2+HCO+OH ↔CO2+H hypothetically, as visible from the equation, at the end of the combustion, all the CO should have become CO2. But this equation has a double verse arrow, so continuously CO becomes CO2 and CO2 becomes CO. At a certain point, during the expansion phase, while the piston is moving from the TDC to the BDM, the temperature decrease so fastly (around 100 K per millisecond), that generates the freezing of this reaction, stopping the transformation of the CO in CO2. The main zone where it is possible to locate the born of this pollutant is Zone D, so the last part of the fuel spray, and Zone C at partial load, so into the heart of the spray.

- HC: the hydro-carbons are particles of fuel that don't take part at the combustion phase. These pollutants can born in different circumstances: in an over-square engine, there are zones called "crevices" where the fuel, during the combustion, doesn't ignite and it is recalled outside during the exhaust valves opening; in a long-stroke engine, some fuel can be attacked on the cylinder's wall, generating these pollutants; the formation of HC is even linked to the type of injector used: a "mini-sac" injector generally favors this pollutant. The zones D and A are low fuel concentration zones and here is where HC can born.

- NOx: CI engines generally produce a lot of these pollutants, due to the thing that these types of engines can work with big excess of air. The NOx typically born because of high temperature during the combustion and the big air flow rate (in a turbocharged engine the formation of NOx is bigger, linked to the big excess of air). The main zone where the NOX born is the C, in the center of the fuel spray, where during the combustion it is reached the maximum temperature.

- particulate matter: the formation of this pollutant is not totally defined, but it is one of the most important pollutants produced by CI engine. It borns in the zones where there is a leak of oxygen and so zone C and zone D.

Now it is possible to compare the pollutant values obtained by the GTP model with the Bharat stage IV values. A DIWiebe model (a non-preductive combustion model) has been chosen for the model developed before, so the accuracy of the pollutant values is not very high, and for this reason during this PHASE II only NOX pollutants are analyzed. Later, with PHASE III, using a predictive combustion model, even the other pollutants will be analyzed.

MAX TORQUE CONDITION

The percentage difference between NO2 value of the model and the target is shown below:

NO2%=14-3.514⋅100=75%">NO2%=14−3.514⋅100=75%NO2%=14-3.514⋅100=75%

The engine modeled in PHASE I produces the 75% more of the limit of the NOX defined by BS IV. To reduce this gap is necessary to use different strategies, without losing the engine performances required. All the strategies can be divided into upstream strategies, in-cylinder strategies, downstream strategies.

- The upstream strategies concern the actions before the cylinder, such as desulphurization and elimination of the lead into the fuel; homogenous charge; cut-off of the fuel during the release of the accelerator pedal, and many others...

- The in-cylinder strategies are applied at the charge inside the cylinder. They can be summarized in:

a) giving the right turbulence at the inlet charge. Generally, for a CI engine, the swirl turbulence is preferred to control the delay ignition, and for a SI engine the tumble turbulence, to control the speed of combustion. In both cases is fundamental to create a homogeneous charge;

b) reducing "crevices", so the places inside the cylinder where the fuel can be trapped and doesn't take part in the combustion, causing the formation of HC;

c) EGR (exhaust gas recirculation), which permits to recirculate a part of the exhaust gases inside the cylinder, to decrease the maximum temperature of the combustion, reducing the formation of NOx.

- the downstream strategies consist of different elements used to reduce the pollutants before exhaust gases leave the outlet system. The main after-treatment systems (that will be discussed later) used for CI engines are:

a) DOC (diesel oxidation catalyst);

b) SCR (selective catalyst reduction);

c) DPF (diesel particulate filter).

The FRM built already contains the EGR but is necessary to calibrate at the best, and then the SCR with urea injection must be introduced.

The main problem of the EGR created by the FRM model is that it doesn't work. It is a High-pressure EGR, so the exhaust gases are directly taken at the outlet of the cylinder, before entering into the turbine. But, the pressure of these gases is less than inlet air, where the EGR is linked, as visible in the picture below, representing the "RLT contour map" of the average pressure:

In this way, the EGR doesn't recirculate the exhaust gases but the inlet air, causing an incorrect tuning of the turbocharger and so of the engine. To avoid this problem, and to use the EGR correctly, is possible to use a "low-pressure EGR", that works in the same way but the recirculation starts after the turbine (and so the exhaust gases have lower pressure and temperature) and ends into a flow split before the compressor. It's better to use a very good heat exchanger for this EGR configuration because the compressor can work better with colder air. In the picture below the "low-pressure EGR" into FRM is shown, and even how it is linked to the flow split of the intake and exhaust system:

Intake system Exhaust system

Now is necessary to understand what is the best percentage of EGR for both cases, obtaining the same performance of the engine and a reduction of the NOX, but is even necessary to visualize that at different values of EGR corresponds differents value of brake power, fixed the boost pressure. For this reason, in "case setup" is possible to visualize seven cases at max torque rpm, changing only the boost pressure, and evaluating the effect of different percentage of EGR, using a DOE:

CASE SETUP

MAX TORQUE MAX POWER

DOE SETUP

The choice of the best EGR percentage value has been based on the following parameters: BMEP, Brake Torque, BSFC, in-cylinder maximum temperature, and pressure, Indicated specific NO2.

After analyzing all the results, the best compromise is obtained with the following values:

rpm	1600	2200
Boost pressure [bar]	2.7	2.2
EGR %	0.1	0.08
BMEP [bar]	20.8	17
Brake Torque [Nm]	829	678
BSFC [g/kWh]	205	214
Maximum Temperature [K]	1880	1910
Maximum Pressure [bar]	156	131
Indicated Specific NO2	8.8	9.33

The EGR is tuned and it is clearly visible the impact that has on the performances and on the NO2 formation. Below the percentage difference between the NO2 values obtained by the simulation and the ones of the Bharat stage IV are shown:

MAX TORQUE CONDITION MAX POWER CONDITION

NO2%=8.8-3.58.8⋅100=60%">NO2%=8.8−3.58.8⋅100=60%NO2%=8.8-3.58.8⋅100=60% NO2%=9.33-3.59.33⋅100=62%">NO2%=9.33−3.59.33⋅100=62%NO2%=9.33-3.59.33⋅100=62%

After tuning the EGR the NO2 value obtained by the simulation is decreased from 72% in the first simulation, to 60% (at max torque condition).

Looking at the "preliminary data tab" at the beginning of this project, at the voice "AFT system" there is "SCR with urea injection". So to complete the PHASE II is necessary to add the SCR at the model. The SCR (selective catalytic reduction) is an after-treatment system that has the target to reduce the NOx family. It works with another substance added manually, called AD-BLUE, composed by Urea and distilled water. Thanks to the high working temperature of the exhaust gases, the Urea (that becomes NH3 reacting with exhaust gases) and noble materials deposited on the surface of the catalyst (titanium oxides, vanadium, and tungsten) is possible to transform the NOx family into nitrogen (N2) and water (H2O).

The SCR, already modeled, has been taken from an example model of GTP, in the following directory: "Example - Aftertreatment - Engine + Aftertreatment - Detailed Aftertreatment CI engine", obtaining the following complete model:

Before showing the results obtained using the SCR, and how it impacts on the performance of the engine, it is necessary to understand how it has been built. Let's analyze the SCR sub-assembly:

as visible, there are two different paths. The first one is solved with the explicit method, while the second is solved with a quasi-steady method, to solve the chemical reactions on the surface of the catalyst. Before the SCR there is an injector, that injects UREA, controlled by different sensors. To work properly this second path needs more information about the exhaust gases, such as the molar fraction composition. To obtain these values, different sensors are used and the values obtained are combined to determinate the composition of the fluid, as visible in the following picture of the exhaust sub-assembly:

Below the results are presented, for both cases, highlighting good and bad things of using an SCR:

rpm	1600	2200
Brake Torque [N-m]	724	649.2
Fuel Flow Rate [kg/h]	25.3	32.9
BSFC [g/kW-h]	208.9	219.7
A/F Ratio	20	20
EGR [%]	11.1	8.8
dPmx/DCA [bar/deg]	4.401	4.17
Intake Pressure [bar]	2.388	2.183
Indicated Specific NO2 [g/kW-h]	8.38	9.03
Maximum Temperature [K]	1869	1904
Maximum Pressure [bar]	137.88	130.68

Comparing these results with the one obtained in the last simulation (using only EGR), is clearly visible a reduction of NOX, thanks to the SCR. But using this after-treatment system is possible to visualize a decrease in the performance of the engine, especially at max torque condition (1600 rpm), staying far from the performance target. This is due to the wave phenomena inside exhaust pipes, that doesn't work properly, causing an aggravation of the performances.

MAX TORQUE CONDITION MAX TORQUE CONDITION

Brake Torque: 724-825724⋅100=-13.9%">724−825724⋅100=−13.9%724-825724⋅100=-13.9% Brake Torque: 649-679.3649⋅100=-4.6%">649−679.3649⋅100=−4.6%649-679.3649⋅100=-4.6% `

NO2%=8.38-3.58.38⋅100=58%">NO2%=8.38−3.58.38⋅100=58%NO2%=8.38-3.58.38⋅100=58% NO2%=(9,03-3.5)/9.03*100=61.2%`

PHASE III

Starting from the results obtained by the last model, using and tuning an EGR and an SCR with urea injection, now let's increase the accuracy of the model, acting on the combustion model, heat transfer model, friction model, and even on the injectors.

Let's start analyzing the model used to calculate the temperature of the wall inside a cylinder. For the past simulations, the reference object used to evaluate these temperatures was called "TWall":

it allows to define the temperature of some parts inside the cylinder, as shown below:

This reference object will be used even in the next simulations because a reference object with more accuracy is hard to set because of the lack of experimental data.

The heat transfer object used is the Woschni GT, as described at the beginning of this project. It will be used even in the next simulations because for this application is a quite accurate model.

The flow object used in the past simulations is called "flow" and was directly defined by GTP when creating the FRM. This flow object will be used in the next simulation because is needed when using a predictive combustion model. This object allows setting the piston cup object, defining the geometry of the piston, that's fundamental to increase the accuracy of the model:

The dimensions of the piston shown in the picture above were defined automatically by FRM. Basing on the paper "Design of piston bowl geometry for better combustion
in direct-injection compression ignition engine " (PRABHAKARA RAO GANJI, RUDRA NATH SINGH, V R K RAJU, and S SRINIVASA RAO
Department of Mechanical Engineering, National Institute of Technology, Warangal 506 004, India) in which three different designs of bowl were investigated to reduce NOx and SOOT formation, one of these geometry has been chosen to conduct the next simulations. In the picture below the first geometry is the one chosen. It represents a piston bowl with a spherical combustion chamber.

To increase the accuracy of the model, the combustion model chosen is a predictive model called "DIPulse": this combustion model predicts the combustion rate and associated emissions for direct-injection diesel engines with single and multi-pulse injection events. The burn rate predicted by the model is extremely sensitive to the injection profile and timing.

The version used for the model is "v2017" that includes overmixing and partial oxidations models for HC and CO emissions. All the other parameters are multiplier and are set automatically by FRM.

In the "emission" tab is possible to define the model used to calculate NOx and SOOT formation. The NOx model chose is the "Extended Zeldovich" (discussed in another challenge) and the SOOT model is "Hiroyasu": a two-step empirical soot model, that considers the soot formation process as involving only two reaction steps: 1) the formation step in which soot is linked directly to fuel vapor molecules; 2) the oxidation step, which describes the destruction of soot particles via the attack of molecular oxygen.

The last element modified is the injector: the InjDieselSimpleConn has been replaced by InjMultiProfileConn, which allows defining fuel flow rate, pressure profile, nozzle geometry, and even multi injections for every cycle.

This injector has different and different tabs, that give the control on every single parameter of the injection. The following one is the "Profile tab" that allows defining injection timing profile, injection mass profile, profile object, and multi injection for every cycle.

To control the combustion inside the cylinder and to imitate the formation of pollutants, two injections were defined:

the first one is called "Pre-injection", that injects a little quantity of fuel before the TDC, and is useful to control the entire combustion.

The second one is called "Main-injection" and is the biggest injection of the cycle.

To understand the best injection timing angle, different values were simulated for max power and max torque condition, with a fixed bowl piston geometry. The results obtained are very satisfying as visible below:

rpm	1600	2200
Pre Inj. Timing [deg]	-13	-30
Main Inj. Timing [deg]	7	-3
Brake Torque [N-m]	797	680
Fuel mass [mg]	149.6	126.4
BSFC [g/kW-h]	215	213
A/F Ratio	20	20
EGR [%]	11.1	8.7
dPmx/DCA [bar/deg]	4.979	4.196
Maximum Temperature [K]	1950	1980
Maximum Pressure [bar]	150	139
Indicated Specific NO2 [g/kW-h]	5.12	4.29
Indicated Specific Soot [g/kW-h]	0.07	0.11
Indicated Specific HC [g/kW-h]	0.26	0.16
Indicated Specific CO [g/kW-h]	0.01	0.01
Indicated Specific CO2 [g/kW-h]	630.55	611.27

Referring only to the indicated specific NO2, for both cases, the results obtained are very near to the Bharat stage IV limit, and the same is for the other pollutants. But, especially for maximum torque condition, the brake torque obtained is lower than the target. So, a compromise has been reached between the performance targets and pollutants.

The main parameters about the cylinder, injection, and combustion are tuned, now it's fundamental to tune other two elements: valves opening and closing, and the turbocharger.

The following pictures show the mass flow rate through the valve compared to the lift:

MAX TORQUE CONDITION

Intake valve Exhaust valve

MAX POWER CONDITION

Intake valve Exhaust valve

For both cases during the intake phase, a part of the air goes outside the cylinder, going up to the intake valve. At max torque condition, this air reflux happens at the end of the intake phase, while at max power condition it happens at the beginning and at the end. Totally different is what happens at the exhaust valve, where for both cases there isn't reflux of gases inside the cylinder. So, the intake air reflux could be solved using two intake valves, or even varying the timing of the valve. But, to contain the design cost, production cost, and the weight of the engine, it was chosen to not solve this problem, but to leave the engine exactly in this state, even because the performance and pollutant values are quite near to the target values.

Totally different is the situation for the operating point of the compressor:

MAX TORQUE CONDITION MAX POWER CONDITION

While at max power condition the compressor work very near the core of the map, so in the zone with the highest efficiency, at max torque condition the operating point is outside the map! So it is necessary to solve this problem, and to find a compromise between two cases.

To shift the operating point to the right of the surge line, two parameters have been modified:

the first one is the wheel diameter of the compressor, set at 65 mm

the second one is the ratio "Φp/Φc for Low Compressor Speeds", which defines the position of the surge line for low speeds of the turbomachinery. Its value must stay between 0.0 and 1, higher values shift the surge line to a higher flow rate, for low speeds. For this reason, a lower value has been chosen, to shift the surge line to the left of the map.

The results obtained for both cases are shown below:

MAT TORQUE CONDITION MAX POWER CONDITION

As visible from the pictures, at max torque condition now the compressor works with the highest efficiency, while in max power condition works with elevated efficiency, very near to the max efficiency. These changes are reflected in the performance of the engine:

MAX TORQUE CONDITION MAX POWER CONDITION

Brake Power: 794.5 Nm Brake Power: 679.5 Nm

Indicated Specific NO2: 5.11 g/kW-h Indicated Specific NO2: 4.3 g/kW-h

The main parameters of the engine are defined and tuned, the values obtained by the simulation are very near to the target defined by the BS IV, so is possible to consider PHASE III ended.

PHASE IV

This is the last part of the project. The engine has been modeled, tuned, and described in every single part, obtaining performance values very and maximum production of pollutants very near to the target, defined by BS IV. But that emission standards was defined in 2010, so ten years ago, and today India must approach to the new legislation: Bharat stage VI (2020).

The limit of every pollutant, controlled by the legislation, is very low. Much lower than BS IV. From the tub is possible to visualize that other two pollutants are controlled: CH4, particulate number, and NH3 (linked to the urea injection into SCR).

The target tab shown below suggest new limits for BSFC and AFT system:

The performance of the engine must be the same, but the limit of the BSFC is passed from 230 g/kWh to 210 g/kWh.

So, to try reaching these new targets is necessary to add at the model other two AFT systems:

a)DOC (diesel oxidation catalyst): promotes chemical oxidation of CO and HC as well as the organic fraction (OF) of diesel particulates. They also oxidize sulfur dioxide which is present in diesel exhaust from the combustion of sulfur-containing fuels. In modern diesel aftertreatment systems, an important function of the DOC is to increase the NO₂ content of the exhaust gas, to support the performance of SCR catalysts.

b) DPF (diesel particulate filter): it is a filter that captures the particulate particles.

The DOC and the DPF have been taken from the same directory of the SCR (used in PHASE II), obtaining the following sub-assembly:

The final results and a complete summary of the geometric characteristics of the engines are shown in the tab below:

Bore [mm]	115.6
Stroke [mm]	119.1
Connecting Rod Length [mm]	238.2
Piston Pin Offset [mm]	0.25
Total Displacement [L]	5.0
Number of Cylinders	4
Compression Ratio	19
Stroke/Bore ratio	1.03
SOHC

		Target		Target
rpm	1600		2200
Pre Inj. Timing [deg]	-13		-30
Main Inj. Timing [deg]	7		-3
Mean Piston Velocity [m/s]	6.4		8.7
Brake Power [kW]	129.1		155.3
Brake Torque [N-m]	797	825	674	679.3
IMEP [bar]	21.11		18.86
BMEP [bar]	19.36	20	16.93	18
Fuel mass [mg]	145.5	165.13	125.9	136.36
Air Flow Rate [kg/h]	550.8		654.0
BSFC [g/kW-h]	216.3	210	214	210
A/F Ratio	19.72		19.72
EGR [%]	11.1		8.7
dPmx/DCA [bar/deg]	4.83		4.202
Intake Pressure [bar]	2.609		2.183
Maximum Temperature [K]	1950		1982
Maximum Pressure [bar]	146.68		138.92
Indicated Specific NO2 [g/kW-h]	4.87	0.4	4.20	0.4
Indicated Specific Soot [g/kW-h]	0.07	0.01	0.03	0.01
Indicated Specific HC [g/kW-h]	0.25	0.13	0.17	0.13
Indicated Specific CO [g/kW-h]	0.01	1.5	0.01	1.5
Indicated Specific CO2 [g/kW-h]	633.08		613.63