Rservoir (TL)

Hello everyone I am trying to model a thermal liquid using simscape. I changed the thermal proprieties in a way that it gives warning when the limits of temperature or pressure are reached. I did a closed loop model and when I run the model without a reservoir the model has a lot of errors mainly regarding the pressure. But when I add a reservoir all the warnings go away.

Sorry for this long introduction my question is the following: what is the real utility of a reservoir ? And does it effect the fact that my model is in a closed loop?

If you need any additionnal information please don't hesistate to ask !

Thank you in advance!

Have a wonderful day!

[Yifeng Tang]

My suspicion is that as a source block (flow source, pressure source, pump, etc.) is adding energy to the flow network, there isn't a way for the energy to get out from the closed network. So the temperature and pressure will keep build up. Adding a reservoir (infinite size) forces the pressure at a certain point in the network, thus stablize the system pressure, but also break the close loop nature of the model.

In some examples in the Documentation, a gas charged accumulator is used to model a water/coolant tank. The compression of the gas allow the liquid to expand a little bit and you won't see the pressure rise without bound. It's also realistic as most closed loop coolant system (1) has a tank and (2) has a bit air trapped inside.

If you could share the model here, it'll help pinpoint your issue.

[Karim Darwich]

Thank you so much for your answer.

I invite you to find the model in the attached file (PS: it is necessary to run the matlab code PROPRIETES before running the simulation).

I am working with the 2023a version of MATLAB.

[Yifeng Tang]

missing "Demande.xlsx"

[Karim Darwich]

@Yifeng Tang I am sorry I edited the previous comment in order to add it thank you so much!

Answer 1

[Yifeng Tang]

Hi Karim,

I did a few things that feel like best practices to me:

(1) add a Tank (TL) block to where the reservoir was. Please review the size of it making sure it's not too small. Your system looks like it's huge. Also when using this block, be careful with the height difference between the ports. If there is a non-zero elevation difference and that difference isn't canceled by elsewhere in the loop, it's a FREE source of energy.

This tank block assumes a fixed pressure, so it acts as a stablizer for the pressure for the overall system. Using a gas charged accumulator can have a similar effect. Or use Tank (G-TL).

(2) replace the two "heat exchangers" with pipes. The way you are adding and removing heat making the heat exchanger unnecessary. The prescribed heat flow can be taken out directly using the heat flow source you already have and a pipe or volume.

(3) turn back on the compressibility option where there is a volume, mostly pipes.

A couple of other things I didn't find necessary here but may help in general:

(1) if your liquid density doesn't change that much, use a mass flow source instead of a volume flow source. I found the former a bit easier on the system of equations and solver.

(2) Add a Transfer Fcn or PS Transfer Function to the input signals so the system can start from rest, instead of getting to a large flow and heat flux at time zero. Starting from rest tends to help with initialization.

Hope this helps. Revised model attached.

[Karim Darwich]

I am extremely grateful for your help. Thank you very much, sir!

(1) Did you replace the reservoir with the tank to utilize the "free" source of energy, or is there another reason?

(2) I completely agree with you, but this is not the final model. That's why I retained the heat exchangers; they will be useful later in the project.

I sincerely thank you so much for your time.

[Yifeng Tang]

(1) No, we DON'T want that free source of gravitational energy. I made the elevations the same so it won't happen. The tank serves as a point of constant pressure. The total mass inside may increase or decrease a bit depending on the overall density of the system, thus stablizing the overall pressure in the loop, while conserving the total mass.