Plotting results

[Lavanya-Bala]

Hi,
I am trying to plot the power output of my motor with the Electrical module and it runs into the following error

Traceback (most recent call last):
File "", line 135, in
out_femm.plot_2D_Data("elec.Pem_av_ref", "time")
File "\Methods\Output\Output\plot\plot_2D_Data.py", line 96, in plot_2D_Data

fund_harm_dict = self.get_fund_harm(data)

File "\Methods\Output\Output\getter\get_fund_harm.py", line 23, in get_fund_harm
if data.symbol in ["B", "P", "Is", "Phi_{wind}"]:
AttributeError: 'numpy.float64' object has no attribute 'symbol'

Also, is it possible to plot the torque-speed/power_speed characteristics ?

[BonneelP]

Hello,

The issue is that elec.Pem_av_ref is defined as "Average Electromagnetic power" and is a float and not a Data object so you can't use this method. (computed here: https://github.com/Eomys/pyleecan/blob/master/pyleecan/Methods/Simulation/Electrical/comp_power.py)
By the way, we recently changed the way we handled plot in pyleecan cf #354 so now it would be output.elec.Is.plot_2D_Data("time") for instance (is you are using the latest version of pyleecan and SciDataTool). The tutorial were updated and the old syntax is still available.

Best regards,
Pierre

[BonneelP]

Hello,

Sorry for the long time to answer. In fact we have never used the "power column" of the OP matrix and the code wasn't available. I have added the corresponding code in PR #417 (I still needs to merge it and create a new release) and now it works if you set is_power=True and provide a 5 columns OP_matrix. However this set only the "reference power" that you input (same as for the reference torque). At the end it is available in xoutput.xoutput_dict["P_ref"].
I have added the method "OutMag.comp_power" method from my post bellow and the dedicated DataKeeper so the computed power is now also available in xoutput.xoutput_dict["P"].

Best regards,
Pierre

[faraz4714]

Dear all,
if anyone can answer me I will be grateful so much
As anywhere was written about electromagnetic torque for IPMSM motor, we know it is equal to
T = 3/2N(phimIq+(Ld - Lq)IdIq).
and when I compute Ld and Lq and phim in pyleecan for a random operating point (in my case Id = -1 , Iq = 1 and rotor speed = 3000 and T_mag = 20 and N = 5) I got this value (Ld = 18110^-6, Lq = 234*10^-6 and phim = 0.0045 ) and if anyone computes T, he/she gets T= 0.034
but if I compute the electromagnetic torque in pyleecan, I got 0.068 which is double
and even if I change iq from 1 to 20 step by step and compute all the parameters, the torque is always double.

anyway, I simulated my motor in flux motor software for the different operating points and I got more or less the same results in pyleecan and the flux motor also my colleague measured the torque and it's the same.

but I am wondering and don't know why the formula is not working well because it's the general formula in the IPMSM motor.
Pierre, if you have any time could you please explain to me how the pyleecan can compute the torque in the magnetic model because I can understand the code

Best regards,

Faraz Mirali

[SebGue]

Hello Faraz,

the formula you used to calculate the torque presumes that you use amplitued values for current and flux linkage. So I guess you calculated with RMS values. If you want to calculate with RMS values you got to remove the factor 1/2 from the formula. (The factor is derived from the amplitue to RMS converserion of current and flux linkage, i.e. 1/sqrt(2) * 1/sqrt(2).)

The torque in the MagFEMM module is calculate in FEMM internally by a weighted Maxwell Stress Tensor approch on the air gap of the machine.

Hope this helps. Best regards,
Sebastian

[faraz4714]

hello Sebastian,
thank you for your answer you are right.
I check it and now everything is correct. can you please give me the reference for this formula? I should present it.
Best regards,
Faraz Mirali

[SebGue]

Hello Faraz,
sorry, I don't have any reference. But I guess you will find it in any book related to synchronous electrical machines.
BR, Sebastian

[Lavanya-Bala]

Hello,

I am trying to plot an efficiency curve with power input/output. Is there a way to do this with Pyleecan classes? The InputElec class doesn't seem to have any Input power parameters

[BonneelP]

Hello,

For now there is nothing related to power (which is a shame).

Regarding input power, it would be something like:

apparent power = qs*U_rms*conj(I_rms)
input power = 0.5*real(apparent power)

But we would need access to the voltage.

Regarding output power, you can use the following formula:
2*pi*output.elec.N0/60*output.mag.Tem_av
The best way would be to add the method OutMag.comp_power() (with the above formula) and then call this method in a DataKeeper. Do you think that you would be able to add this method ?

Best regards,
Pierre

[Lavanya-Bala]

Hello,

Thanks for the clarification, I will look into doing this. I also want to find the losses and I see the classes LossModelBertotti etc., Is there a tutorial or sample program on how these are implemented in a script used to simulate the motor ?

[RaphaelPile]

Not yet, but you can take a look at: https://github.com/Eomys/pyleecan/blob/master/Tests/Validation/Loss/test_FEMM_Loss.py

If you have some questions, maybe @SebGue can help you a little bit since he wrote the code.