Week-7 : Converting a detailed engine model to a FRM model

Assignment 7: Converting detailed engine model to FRM Model

Aim:

To understand the concept of FRM model through examples

Introduction:

In this assignment we would be understanding the concept of FRM (Fast Running Model) through some examples. First, we would be exploring the Tutorial 9 in detail and then would be building and running a FRM model using FRM Builder tool.

Exploring Tutorial 9:

This tutorial which describes the conversion of a very detailed 4-cylinder GDI engine to a FRM model is divided into 8 steps. These steps essentially outline the transition mentioned above. We will also follow the same order and would discuss the steps in detail.

Step 1:

Refer Fig_1 for the model. This is the model that has been given to us. It is a very detailed model of a 4-cylinder Turbocharged GDI engine. The level of detail is such that it covers all aspects of the engine right from the intake to the exhaust. The components that we can observe are:

• Engine intake and intake pipe
• Helmholtz resonator sensor
• Flow restrictions (caps)
• Pipe length, bends, diversions in the pipe
• Air Box
• Compressor
• Orifice
• Intercooler and its controller
• Throttle plate
• Intake manifold and runner
• Intake and exhaust valves for each of the 4 cylinders
• Engine Crank Train
• Gasoline injectors
• Outler pipe
• Turbine
• Catalytic converter brick
• Exhaust

Considering that these many components have been modelled, it is obvious that running this model will require a lot of time and substantial computational resources. If one has ample of the above two, this model will do the job perfectly, no change would be required. But in cases when such models have to be used in the case of Hardware in the Loop testing, it is imperative that this model respond quickly with the same dynamics as that of the system it simulates.

In such cases, the model will not be able to catch up with the actual real-world components and thus would fail to give any results.

To remedy such cases, the model is manually simplified wherever possible to convert it to a FRM, which can runner faster due to the increased simplicity. The next steps will thus outline the opportunity for simplifying the model and would show the same.

Step 2:

Refer Fig_2 for the second step. In this step we see that an opportunity to reduce the model has been born out of simple prioritization of the needs of the model. As we had seen before, this model consists of the catalyst brick which would simulate the after-treatment system dynamics.

But the main aim of this model is to simulate the engine. The results of the catalytic converter make little sense for the required results.

Thus, we see that in this model, the catalytic brick and its associated systems have straightaway been removed. Now we will proceed to step 3.

Step 3:

The next glaringly visible opportunity is the intake runner and manifold circuits on the left of the engine cylinders. Refer Fig_3 for the same.

For this component around 42 objects have been used in the previous model for a detailed representation of the same. If the intake dynamics are not so much important, this list of components can eb cut down to simple representative ones. In the step 3, using only 14 components the system has been modified. A single representation of the intake manifold is used by modelling the entire area as one. Also, instead of using 2 intake valves for each cylinder, only 1 valve is used.

Step 4:

Refer Fig_4 for the above model. Using the same approach as used in step 3, further simplifications have been made in the intake system upstream. The number of intake components and the intercooler components have been simplified.

Step 5:

Refer Fig_5 for the model. The approach used in the above steps is used again and further lessening of the components upstream of the intake system is done. The 13 objects earlier were reduced to only 4

Step 6:

Refer Fig_ 6 for the model. The focus again shifts to the output side, where further reduction I objects take place. Earlier, individual exhaust line paths were left much detailed. In the new one though, these individual paths were combined into 1.

Step 7/ 8:

In step 7 as we see, the FRM model is obtained after simplifying using the steps that have been mentioned above. Thus, in step 8, we have a manually built FRM model ready! Refer Fig_7.

Building a FRM Model using FRM Builder:

Using the FRM model builder, we have to build an engine using the given data. The following steps signify the steps followed:

• Fig_8, CI engine configuration selected
• Fig_9, Engine geometrical details like the bore, stroke, and compression ratio specified
• Fig_10, Twin scroll turbine selected
• Fig_11, random values for other operational parameters selected
• Fig_12, 6cylinder configuration selected
• Fig_13, GT Suite Controller selected.
• Fig_14, Final Model

Thus, we have built the FRM model using the FRM Builder

Running the cases:

For this model, considering the computational cost of all the given case, we would be running first 3 cases given. An example of the result is given in Fig_15

Thus, all the required questions have been completed