How to model marine exhaust system using Simscape.

I want to create a simulation for marine exhaust systems for large vessels but I'm not sure how to set everything up.

I have values for temperature (from turbo), ambient temp (engine compartment), k values, h values, various exhaust pipe diameters, lengths, and thermal properties (stainless steel, aluminum, etc.).

Sections of the exhaust use a multilayer multi product insulation. I know the diameters at each layer, h and k values of the different layers, and thermal resistances of the layers.

Eventually, I want to include where the exhaust gas mixes with sea water in a mixing chamber.

My end goal is to create a model that can find the surface temperature, the internal temperature, and the internal pressure at any location of the exhaust system.

Can someone help with creating a basic model to start with?

[Yifeng Tang]

Do you have a diagram of some kind to explain what you are trying to model?

The air flow part, i.e. pressure, temperature, flow rate, heat transfer, etc., can be handled reasonably well at the system level using Simscape and Simscape Fluids. The mixing with sea water, however, can be exponentially more difficult, depending on what kind of physical effects you are trying to include.

Anyway, more clarification will help.

[Jonathan]

Thank you. Please see the picture. I can provide more information if needed. The mixing with sea water can wait. I wanted to see if it would be something that I could add to the simulation at a later time.

Answer 1

[Yifeng Tang]

Hi @Jonathan

Thanks for sharing the picture. That helps.

Whether MathWorks tools, in particular Simscape, can help model this depends on what you want to use the model for. Simscape and Simscape Fluids are designed to do 1D system-level modeling, often as the plant model of a feedback control system. For example, if you want to analyse how certain valve control strategy will impact the overall system performance, yes, you may build a model in Simscape and it'll be useful.

If you are trying to analyze the flow field inside the pipe, or maybe vary the connector design and see the impact on the pressure drop, Simscape isn't the right tool and you need a CFD package instead. You requirement on "want to know at any location ..." seems to suggest this level of detailed modeling. If true, I'm afraid the physical modeling tools from MathWorks aren't the right choice.

[Jonathan]

I understand. I believe you have answered my question.

Here is a picture of the simulation that I'm working on. I was hoping there would be Simulink blocks and/or methods to do CFD that I could include within my simulation.

I'm completely new to Simulink and Simscape. This was an attempt to essentially create a CFD from the ground up using Simulink.

[Yifeng Tang]

Some quick comments:

1. If you are new to Simulink and Simscape, the best place to start, before getting to this specific model, is Simulink Onramp and Simscape Onramp. They are free, short, and super useful introductory tutorials.
2. You'll need a "Reservoir" at the end of Elbow 7. Right now the outlet is "sealed". Pressure sensor, like a pressure gauge, won't let any air passing through it.
3. Pipe blocks include convective heat transfer equation between the inner wall surface and the gas inside. So you may not need the convective heat transfer block. It's common to use conductive heat transfer and a thermal mass block to model the pipe wall, if desired.
4. There will likely be more parametrization work and that's where the real work starts :p
5. It's a good idea to run a few 1D simulation and find out what will be a reasonable sets of boundary conditions to run the 3D CFD simulation. You won't be "replacing" CFD with Simscape, or create a CFD using Simulink and Simscape, though.

[Jonathan]

Thank you very much for this information! It is exactly what I was looking for. I'll give it a try and update you.

[Jonathan]

1. Finished Simulink Onramp and Simscape Onramp. Learned a lot, but still have many questions.
2. Included a Reservior at the end of Elbow 7
3. Removed convective heat transfer blocks. Included mass blocks
4. agreed
5. Ok

Moving forward:

I would like to include my simulation here to be critqued and suggestions made. I have a lot more questions regarding this simulation.

1. Do I need to start a new thread or just post the simulation here?
2. How do I upload the simulation for you to be able to open it (if you desire)?

[Yifeng Tang]

If you feel comfortable sharing the model here, you may upload it with your comment or edit and include in your original question.

[Jonathan]

Okay. I have attached the model here.

1. I removed the mass blocks because something was not correct about the setup.
2. I don't think I have the Engine set up correctly.
3. Feel free to make any changes necessary to make the model work correctly
4. I have known values, and I can easily calculate any missing values, but I don't know where to put all values since there seems to be multiple places to insert the same values.

Here is a list of the known values

1. mass flow of the exhaust gas
2. volumetric gas flow
3. Gas temp
4. Engine room temp (ambient)
5. Gas pressure
6. Gas density
7. Dynamic viscosity
8. Gas Thermal Conductivity

Here is a list of known information about the pipes.

1. Species of pipe (Aluminum or Stainless steel)
2. All pipes (EXCEPT THE TURBO FLANGE) are wrapped in Fiberglass insulation
3. Lengths
4. Diameter ID and OD
5. Thickness of insulation
6. Pipe Thermal Conductivity
7. Insulation Thermal Conductivity
8. Flow regime
9. Reynolds number
10. Prandtl Number
11. Nusselt Number

My biggest problem Is that I don't know where to plug in some of the values since there are multiple places to insert the same values. For now, I just want the temps and pressures at the mates of the pipes, or at the very least, the temp and pressure at the end of the last pipe.

[Yifeng Tang]

Lots of these information will be useful when modeling the pipe.

The Pipe (G) block models the gas inside, the flow resistance, and the heat transfer between the gas and the INNER WALL of the pipe. The pipe material and the insulation is not part of this block but can be added easily. Below is an example.

The diameter, thickness, and thermal conductivity of the metal and insulation will go into the conductive heat transfer blocks. The density, specific heat, and total mass of the pipe material will go into the thermal mass. The outside temperature can be prescribed using a temperature source.

I took a look at your model. Some comments:

(1) thermal reference represents 0 K. I don't think that's what you intended. Use a temperature source instead.

(2) You need to set the unit for each of the PS-Simulink converters. Leaving them as inherit will get you SI unit (Pa, K, etc.). Looks like that's NOT what you wanted.

(3) when measuring a temperatue and using units like degC or degF, make sure you check the apply affine conversion. See this doc page for more information.

(4) you probably need to use the measured pressure and temperature to compare to your reference data to understand whether you need to tune parameters like "aggregate equivalent length for local resistance" and "hydraulic diameter" to get the right amount of pressure drop and/or heat transfer.

Good luck and happy modeling.

[Jonathan]

Thank you very much! This helps tremendously. I will incorporate your suggestions and update you here.