Assignment 7
FRM model: Fast Running Engine Models are dynamic, fully-physical engine models that are designed specifically to run fast.FRMs can achieve such fast run times by two means.
1) increasing the simulation time step size.
2) decreasing the number of calculations per time step.
These two means are often accomplished at the same time by lumping various flow volumes together which will reduce the total number of subvolumes and also allow for a larger time step size by increasing the effective subvolume length.
The FRM models can be created in two ways.
- FRM Converter: In this method, the detailed model is converted into the FRM model by combining subvolumes and other methods to maintain the courant number that will affect the time step size.
- FRM Builder: The model is created by using the set of libraries that were created in the GT-power.
1)Explore tutorial number 9 and write a detailed report
The main aim of this project is to convert the detailed model into the FRM model and to compare the results of both models.
The model we are using is a 4-cylinder 2.0-liter turbocharged Gasoline Direct Injection (GDI) engine. The turbine wastegate is being dynamically controlled to target BMEP. A semi-predictive charge air cooler is included such that the outlet temperature is predicted as a function of inlet gas temperature and mass flow rate, coolant temperature, and charge air cooler effectiveness.
For converting a detailed model into the FRM model first we should overview the parts that limit the time step size this is done by using the RLT counter maps. The lower the value, the more likely the part will limit the timestep. The RLT Contour map can also be used to check the number of subvolumes for each pipe model. The more subvolumes are used the slower the execution time will be. By doing this it has tagged the different similar parts that can be combined. The FRM converter is a tool that will track the FRM Conversion process, organize all model files automatically within subdirectories, and provide result comparisons of key RLTs throughout the conversion process. The detailed model is simulated and the results are used to maintain the accuracy. There is an accuracy checking dialog box to track the accuracy and to create the tolerance of the accuracy.
Step 1: Exhaust Manifold: The reason for choosing exhaust manifold in the first step because it is the place where the highest gas velocities occur, which in turn restricts the time step. It is most efficient to start by simplifying the subsystem that is currently restricting the time step. Before we apply simplifications to the exhaust manifold, the Maximum Time Step limit of 1 crank angle degree should be changed to a 720 crank degree to support the maximum time step. Lanch the FRM controller and tag the exhaust manifold system. Here there are two options if accuracy is of higher importance than run-time, choose the "Simplify for Accuracy" option. This will combine the Exhaust Ports/Runners, and left separate from the manifold volumes. If a real-time capable model is required choose "Simplify for Speed". In this case, the Ports, Runners, and Manifold volumes will be combined at once. In this simulation, we are using simplify for Accuracy model to maintain the accuracy of the results.
The Simplify Manually option will allow us to manually simplify geometry for parts that were not tagged or where a different combination of volumes is desired. The discretization length in the pipe part is automatically created and HTM_Exhaust_Manifold_1_Step_1 will be created to calibrate heat transfer in the combined volume to assure the Turbine Inlet Temperature is maintained. These parameters can be varied in the case setup toolbar. Once the model is simplified it is preprocessed before setting up the calibration.
The case set up for step1
Step 2: Exhaust Pipes: In the second step of model simplification, the exhaust pipes downstream of the turbine will be simplified. In the FRM converter select the Exhaust pipe-1 and click on simplify for accuracy it will delete the catalyst volume and flow split parts. Now select combinedly pipe, drop and orifice parts right click on the selection and launch the Combine Flow Volumes Wizard and name the part to CAT. In the CAT 1 part change the wireless signal part to "tav0: CAT-1”. Open Case Setup and set the DIA_Exhaust_Pipes_1_Step_2 to 40 mm. This is only an initial guess to calibrate the Turbine Outlet Pressure. From the FRM converter change the response RTL to the Turbine Massflow-Averaged Outlet Pressure (trb-poa: Turbine), and set the lower and upper values of the range of DIA_Exhaust_Pipes_1_Step_2 at 35 and 45 mm, respectively. After running the calibration rerun the program for satisfactory results.
Step 3: Intake Manifold: In the FRM converter select the Intake Manifold 1 and click on simplify for accuracy. After the parts are deleted minor modifications are needed for Characteristic Length and Expansion Diameter attribute values for the combined volume Open the new IM1_Plenum1-1 part and define the Boundary Data folder and give a characteristic length of 70mm. The expansion diameter defines the maximum diameter that the flow may expand to after entering a boundary and it will be equal to the plenum diameter multiplied by 1.4. In the Calibration page, select Case 1 for calibration. Under "Configure calibration…", set the Response RLT to the Compressor Massflow-Averaged Outlet Pressure To maintain the engine airflow. Run the calibration and after that run the model for satisfactory results.
Step 4: Compressor Outlet Pipes: In the FRM converter select the Compressor Outlet Pipes and click on simplify for accuracy. Since the intercooler core has been replaced with a FlowSplit and HeatExchangerConn and this part will be used to assure the intercooler outlet temperature is maintained and set the wall temperature to 300K. In the Calibration page, select Case 1 for calibration. Under "Configure calibration…", leave the Response RLT as the Compressor Massflow-Averaged Outlet Pressure, the Intercooler Diameter will be adjusted to maintain the compressor outlet pressure. Run the calibration click Run Model to rerun the full set of Cases for satisfactory results.
Step 5: Intake Pipes: In the FRM converter select the Intake Pipes 1 and click on simplify for accuracy and set the diameter of the intake pipe to 45mm. Every subsystem has now been simplified, we will assure that there are no remaining pipes with short discretization lengths and delete the unused objects. After doing this we can say that every subsystem is been simplified and there are no small pipes present in the system and finally the pipes can be organized by varying the Length column. In the Calibration page, select Case 1 for calibration. Under "Configure calibration…", change the Response RLT to the Compressor Massflow-Averaged Inlet Pressure. Run the calibration click Run Model to rerun the full set of Cases for satisfactory results.
Step 6: Additional Changes: In the FRM converterSelect Exhaust Manifold 1 from the Tagged Subsystems list. Click on "Simplify for Speed". Go to the Calibration page, leave all cases enabled for calibration. This may take some additional time to calibrate vs. only calibrating for Case 1, but it will result in better accuracy across all operating conditions. Under "Configure calibration…", change the Response RLT to the Turbine Massflow-Averaged Inlet Temperature. Run the calibration click Run Model to rerun the full set of Cases for satisfactory results.
Step 7: Cylinder Slaving: In the FRM controller simply specify the name as cylinder slaving and collapse the tab. Right-click on any of the EngCylinder parts on the map and choose "Edit Parent Object". Go to the "Advanced" tab and for "Cylinder Slaving Option", select "slave-RT-full-v2017", then click OK. Now double-click on Cylinder-01, go to the "Advanced" tab, and set the Cylinder-01 override value of "Cylinder Slaving Option" to "master". For the full cylinder slaving to function, each Valve must be specified as either an Intake or Exhaust valve and this can be done in the valve camcon template there for every intake and exhaust spreadsheet we should specify at the output folder wether it is intake or exhaust.
Step 8: Real-Time Solver: In the FRM controller simply specify the name as Real-Time Solver and collapse the tab. In the real-time solver, certain templates are not allowed so they must first be removed from the model.
- In the Main assembly delete the Turbo_Speed-2 monitor part coming off of the turbocharger shaft.
- Evaporation Object in the engine cylinder part should be ignored this is done by right-clicking and choose "Edit Parent Object". Go to the Advanced folder and set "Evaporation Object" to "ign" and click OK.
- The "Steady State Error Tolerance should be ignored in the Wastegate controller.
The "FRM Accelerator" dialog that automatically shows up we will choose "Continue with Detailed Settings" at the bottom of this tool to leave the detailed settings. The RLTs need to be recreated for the RLT references to work. The "trb-tia: Turbine" and "pav: IntakeManifold-1" RLTs will not be used in the RT solver these columns can be deleted in the RLTCreatorRT object in the model tree to the left. Calibration will not be performed for this step, so simply click Next to get to the Results page without performing calibration, then click "Run Model" to run the full operating range with the Real-Time Solver.
Results: The results are compared in two ways one is the speed of the simulation is compared with the real-time model and accuracy is compared with the Detailed model.
1) The speed of the simulation
| speed(RPM) | No of cycles for simulation | Real-time for the simulation | Computational time | The time factor |
| 5000 | 38 | 0.912 | 24 | 26.31579 |
| 4000 | 41 | 1.23 | 4.57 | 3.715447 |
| 3000 | 31 | 1.24 | 4.29 | 3.459677 |
| 2000 | 24 | 1.44 | 4.91 | 3.409722 |
| 1000 | 66 | 7.92 | 17.12 | 2.161616 |
In the high load conditions, the time factor of the FRM model is high it approximately takes the 26 times of the real-time simulation but in the part load and low load conditions, The FRM is showing better results.
2) The accuracy of the simulation.
For comparing the accuracy of the simulation main parameters of both Detailed and FRM are compared.
These are the results of Detailed model simulation
| Engine Speed (RPM) | 5000 | 4000 | 3000 | 2000 | 1000 |
| Brake Torque(N-m) | 298.5477 | 298.5665 | 298.5583 | 230.8689 | 143.2527 |
| Brake Power(kW) | 156.3192 | 125.0632 | 93.79485 | 48.35308 | 15.00139 |
| BSFC(g/kW-hr) | 353.7887 | 321.05 | 290.8626 | 276.7585 | 283.8656 |
| BMEP(bar | 18.75966 | 18.76084 | 18.76033 | 14.50697 | 9.001484 |
| volumetric efficiency | 1.663446 | 1.57603 | 1.565458 | 1.241498 | 0.790118 |
| Air-Fuel Ratio | 10.52625 | 10.98939 | 12.04889 | 12.98679 | 12.9867 |
These are the results of FRM model simulation
| Engine Speed (RPM) | 5000 | 4000 | 3000 | 2000 | 1000 |
| Brake Torque(N-m) | 285.5299 | 298.1404 | 298.2328 | 230.7744 | 143.2419 |
| Brake Power(kW) | 149.5031 | 124.8848 | 93.69259 | 48.33327 | 15.00025 |
| BSFC(g/kW-hr) | 376.9496 | 328.5703 | 295.9898 | 279.2054 | 285.6253 |
| BMEP(bar) | 17.94167 | 18.73407 | 18.73987 | 14.50103 | 9.000803 |
| volumetric efficiency | 1.695421 | 1.611104 | 1.591508 | 1.251523 | 0.795087 |
| Air-Fuel Ratio | 10.52847 | 10.99251 | 12.05034 | 12.98225 | 12.98885 |
The comparison of both results are done and tabulated
| Engine Speed (RPM) | 5000 | 4000 | 3000 | 2000 | 1000 |
| Brake Torque(%) | 4.360362 | 0.142705 | 0.109024 | 0.040958 | 0.00756 |
| Brake Power(%) | 4.36036 | 0.1427 | 0.109021 | 0.040963 | 0.007566 |
| BSFC(%) | -6.54651 | -2.34242 | -1.76277 | -0.88413 | -0.61993 |
| BMEP(%) | 4.360366 | 0.142696 | 0.109017 | 0.040953 | 0.007565 |
| volumetric efficiency(%) | -1.92221 | -2.22548 | -1.66408 | -0.80756 | -0.62899 |
| Air-Fuel Ratio(%) | -0.02111 | -0.02845 | -0.01206 | 0.034974 | -0.01657 |
These are percentage error of the FRM model when compared to the detailed model, The percentage of error is more in high load conditions when compared to the part-load and low load conditions.
On analyzing both results we can say that the FRM model will be useful for part-load and low load-load conditions but at high-loads, it is not recommended for this engine.
FRM builder approach: It is another type of FRM model which is created by using the FRM builder tool in the tool tab. We are building watergate turbocharger Compression engine, specifications of the engine are
- Bore-102 mm
- stroke- 115 mm
- Compression Ratio- 17
- No of a cylinder- 6
- Twin Scroll Turbine
This the FRM builder model in which all the parts are divided into a separate subsystem and all these parts are connected by the controller unit.
GI-Control unit: This is the unit that maintains all the connections between the parts.
Receive signal controller unit; It receives the signal from the different parts of the engine through the wireless signal and converts the signal into accessible information for the controllers and sends the information to the respective controllers.
The signal carries this real-time information from different parts.
ControllerDInject: This is used to target various engine performance parameters at part-load operation by adjusting the injected fuel quantity. The amount of fuel quality is based on three parameters engine speed, Airflow rate, and BMEP. The targeted BMEP and Engine speed are inputs of the engine according to the Controller control Air mass flow rate and Fuel quantity.
FRM_EGR_Control: This controller is used to control the EGR valve of the engine and it mainly depends on the speed and BMEP values. The inputs of the valve opening at different speeds and BMEP conditions are already specified in the input according to this the controller passes the signal to the EGR valve.
ControllerTurboWG: This template is used to target various engine performance parameters in turbocharged applications. For every step, it calculates the desired wastegate turbocharger diameter based target boost pressure it controls the mass flow rate to the engine. All the controllers are interrelated and these controllers should be tuned for best results.
SendSignal: It sends the information from controllers to different parts of the engine and it converts the signal into accessible information for the parts.
These signals are sent to their respective parts to control the engine and this process repeats for every cycle.
Diesel Engine: In the FRM builder all the parts are divided into different groups that access the information from different parts and process the information to the preceding parts and RTLs are generated.
The above figure is the FRM diesel engine which gets the Air mass flow rate from the compressor, EGR flow rate from the controller, and Fuel quantity injection rate from the controller, and all this information is processed and gives to Exhaust and RTL signals are generated.
Turbocharger: The turbocharger we are using is a Twin entry Wastegate turbocharger with an intercooler. The diameter of the wastegate turbocharger is varied by the respective controller.
EGR: The High-pressure EGR is used in this model, the exhaust gases will directly be connected to the intake manifold before entering into the Turbine.
The above EGR setup consists of an intercooler which is used to decrease the temperature of the exhaust which is controlled by a set of equation. The EGR valve diameter is controlled by the EGR controller.
Simulation: The simulation was run for 10 different cases for 5 sets each set has 2 cases with the same load target for different speeds, the loads we are selected are 100%, 80%, 60%, 40%, 20%, and results ate tabulated according to it.
Results:
The case set up for 100% load at 4000 and 3000rpm
| Engine Speed | 4000 | 3000 |
| Load % | 100 | 100 |
| Target BMEP in the bar | 17 | 22 |
| Target boost pressure | 2.8 | 2.8 |
The engine performance results are
The performance table for the turbine and compressor at 100% load.
The efficiency map of the compressor at 100% load.
The case set up for 80% load at 4000 and 3000rpm
| Engine Speed | 4000 | 3000 |
| Load % | 80 | 80 |
| Target BMEP in the bar | 13.6 | 17.6 |
| Target boost pressure | 2.6 | 2.6 |
The engine performance results are
The performance table for the turbine and compressor at 80% load.
The efficiency map of the compressor at the 80% load.
The case set up for 60% load at 4000 and 3000rpm
| Engine Speed | 4000 | 3000 |
| Load % | 60 | 60 |
| Target BMEP in the bar | 10.2 | 13.2 |
| Target boost pressure | 2.4 | 2.4 |
The engine performance results are
The performance table for the turbine and compressor at 60% load.
The efficiency map of the compressor at 60% load.
The case set up for 40% load at 4000 and 3000rpm
| Engine Speed | 4000 | 3000 |
| Load % | 40 | 40 |
| Target BMEP in the bar | 6.8 | 8.8 |
| Target boost pressure | 2.2 | 2.2 |
The engine performance results are
The performance table for the turbine and compressor at 40% load.
The efficiency map of the compressor at 40%load.
The case set up for 20% load at 3000 and 2000rpm
| Engine Speed | 3000 | 2000 |
| Load % | 20 | 20 |
| Target BMEP in the bar | 4.4 | 4.4 |
| Target boost pressure | 2 | 2 |
The engine performance results are
The performance table for the turbine and compressor at 20% load.
The efficiency map of the compressor at 20% load.
Upon studying the above results the following observations are made.
- For this engine, the maximum power and torque can be obtained at 3000rpm when compared to 4000rpm.
- At the respective load, the brake specific fuel consumption is high at 4000rpm. The BSFC values increase with decreases in load percentage.
- At high loads, the engine runs at a rich A/F ratio mixture as the load percentage decreases the A/F mixture gets leaner.
- The brake efficiency at 3000rpm is high when compared to 4000rpm at different loads.
- At 100% load, at both speeds, the compressor runs at a choking point.
- The engine running at 4000rpm and loads up to 60% the operating point at the compressor runs at a chocking point.
- At low load conditions, and at 3000rpm the operating point compressor falls under maximum efficiency zone.