Simulink - Simscape HTC input using model parameters

Hi,
I'm trying to define a variable convective heat transfer coefficient using model parameters (i.e. temperature) in the simulink model shown below. Is there a way to define this?

I get this error when I run the simulation ' ...or the model referenced by it contains a block that updates persistent or state variables while computing outputs and is not supported in an algebraic loop. It is in an algebraic loop with the following blocks.'
So does that mean we can't define the conv_HTC as a function of temperature or is there a way to do it?

Many thanks

[Youssef Noureddine]

Hi Delika,

It seems that the simulink model attachment/image did not make it through, can you please re-attach the model or image. Additionally if you can provide the full error message this would be addionally useful for me to investigate further.

Sincerely,

Youssef Noureddine

[Delika Weragoda]

Hi Youssef,

Thank you very much for replying. I think I've made a mistake in how the loop was created. Initially I created the input signal using simulink and then converted to PS and got the error. But then I directly used the PS output from the temp sensor and gave it as an input. I'm not sure if this is the way to go. But the figure below shows what I was trying to do. i.e. basically define a custom convective HTC as a function of temperature. (connections shown in blue are the ones I had before, apologies for my childish art here :D .)

I'm not sure if this is the way to do it.

Any feedback would be much appreciated. (sorry I couldn't find a snapshot of the error I got, but this is what I basically got ...or the model referenced by it contains a block that updates persistent or state variables while computing outputs and is not supported in an algebraic loop. It is in an algebraic loop with the following blocks.')

Many thanks

Delika

[Youssef Noureddine]

Hi Delika,

Can you provide the contents of the "variable convective heat transfer" block? I would recommend continuing with this for now as it seems you are not encountering any errors. I would however suggest simplifying the model (blocks like temperature 1 and temperature 2 seem identical and may cause additional confusion).

Since I do not know what the contents of your custom block are, I cannot tell for certain if you will run into an error, but if you do, feel free to attach it here so we can understand what the exact issue simscape is running into is.

[Delika Weragoda]

Hi Youssef,

Thanks, yeah I figured there is an extra temp sensor when I uploaded. Did change it afterwards. Please see the content for the variable convective HT block below.

-----------------------------------------------------------------------------------------------

inputs

T_diff = {1, 'K'}; % S:bottom

end

parameters

k = {0.598, 'W/(m*K)'}; %Thermal conductivity of water (W/mK)

C_p = {4200, 'J/(kg*K)'}; %specific heat capacity of the liquid (J/kgK)

beta = {210e-6, '1/K'}; %Volumetric thermal expansion coefficient

mu = {8.9e-4, 'kg/(m*s)'}; %Dynamic viscosity of water (Pa.s)

rho = {998.2, 'kg/m^3'}; %Density of water

g = {9.81, 'm/s^2'}; %acceleration due to gravity

%Battery dimensions

Bat_width = {160e-3, 'm'}; %Width of the battery (m)

Bat_height = {227e-3, 'm'}; %Height of the battery (m)

Bat_thickness = {7.25e-3, 'm'}; %Thicknesss of the battery (m)

end

parameters (Access=private)

nu = mu/rho; %Kinematic viscosity

alpha = k/(rho*C_p); %Thermal diffusivity

Vol_of_cell = Bat_height*Bat_width*Bat_thickness; %Volume of the pouch cell

Surf_area_cell = 2*Bat_height*Bat_width; %Surface of cooling of the cell (only one surface considered)

L = Vol_of_cell/Surf_area_cell %Lenght scale of the cell

end

nodes

A = foundation.thermal.thermal; % A:bottom

B = foundation.thermal.thermal; % B:top

end

variables (Access=private, Conversion=relative)

T = {0, 'K'}; % Temperature difference

end

variables (Access=private)

Q = {0, 'W'}; % Heat flow rate

end

branches

Q : A.Q -> B.Q;

end

equations

T == A.T - B.T;

%Equation for heat transfer coefficient is embedded in the following equation

Q == Surf_area_cell *((k/L)*(0.68 + ((0.67*((g*beta*T_diff*L^3)/(nu*alpha))^(1/4))/(1 + ((0.492*k)/(C_p*mu))^(9/16))^(4/9))))* T;

end

end

---------------------------------------------------

Cheers

Delika