compute efficiency map using macro and compose script

ramin_21612

Dear Sir/Madam,

I tied to compute efficiency map of the IPM by combining Flux CreateLookUpTableFromTMProjectDQLight macro with Efficiency_maps Compose script in Flux 2D 2021.2. In the "represent_maps.oml" script, the code required an Iron losses at maximum speed, but I could compute it at nominal speed and current. How do I calculate this? 

Another question is why are Joule losses on the magnets and windings of the stator not included in the script to calculate the efficiency with high accuracy?

Thank you for your help.

Ramin

Mohammed Elamin_22169

Hi Ramin, 

For iron losses computation you could eventually do FE simulation for the maximum speed point (if you know the speed , current and control angle) and extract the iron losses for maximum speed point, in similar way to how you computed for nominal speed and current. If you don't know the information for maximum speed, I would suggest to try some estimation based on the values you computed for nominal speed. A typical iron losses curve is proportional to the square of the speed/frequency (for a given Flux density). 

I am not sure which script you are referring to. The main script used for the computation are encrypted. can you please clarify? But according to the documentation I can see the winding losses (The DC part of it) will be computed. Probably not the magnet losses as it requires transient simulation and all the computation used by this tool are static.          

Thanks, 

ramin_21612

Mohammed Elamin_22169 said:

Hi Ramin, 

For iron losses computation you could eventually do FE simulation for the maximum speed point (if you know the speed , current and control angle) and extract the iron losses for maximum speed point, in similar way to how you computed for nominal speed and current. If you don't know the information for maximum speed, I would suggest to try some estimation based on the values you computed for nominal speed. A typical iron losses curve is proportional to the square of the speed/frequency (for a given Flux density). 

I am not sure which script you are referring to. The main script used for the computation are encrypted. can you please clarify? But according to the documentation I can see the winding losses (The DC part of it) will be computed. Probably not the magnet losses as it requires transient simulation and all the computation used by this tool are static.          

Thanks, 

Hi Mohammed,

Thank you for your response. To calculate efficiency map, I first create the Transient Magnetic project of motor. Then I ran the CreateLookUpTableFromTMProjectDQLight macro as shown below.

The macro created a file *project_name*_MS_CSV_table_res.oml containing ID, IQ, Flux_D, Flux_Q, LQ, LD, and torque data. To compute efficiency maps, I used the Compose script AFIR2_Analytic_FEMT.zip, in which I modified the Flux_D.oml and Flux_Q.oml input files and added my Flux_D and Flux_Q values obtained through the macro. Then, the script "represent_maps.oml" is run to compute the efficiency map. Since the motor produces 64 N.m at nominal speed (1500 rpm), the reported data are incorrect. where is my mistake?

Mohammed Elamin_22169

ramin_21612 said:

Hi Mohammed,

Thank you for your response. To calculate efficiency map, I first create the Transient Magnetic project of motor. Then I ran the CreateLookUpTableFromTMProjectDQLight macro as shown below.

The macro created a file *project_name*_MS_CSV_table_res.oml containing ID, IQ, Flux_D, Flux_Q, LQ, LD, and torque data. To compute efficiency maps, I used the Compose script AFIR2_Analytic_FEMT.zip, in which I modified the Flux_D.oml and Flux_Q.oml input files and added my Flux_D and Flux_Q values obtained through the macro. Then, the script "represent_maps.oml" is run to compute the efficiency map. Since the motor produces 64 N.m at nominal speed (1500 rpm), the reported data are incorrect. where is my mistake?

I am not sure what is the issue. When I run the script on the file you sent, I am getting different results than the one on the image above (peak around 100). So I am not sure  if it is for the new tables you got or the old one from the example. Can you tell where is the location in which you replaced the tables. 

This script to compute efficiency map is developed to work with the axial Flux macros, so I am not sure if it compatible with the one "CreateLookUpTableFromTMProjectDQLight". I need to check more on this. I will see if someone else can take look at your question.  

 

ramin_21612

Mohammed Elamin_22169 said:

I am not sure what is the issue. When I run the script on the file you sent, I am getting different results than the one on the image above (peak around 100). So I am not sure  if it is for the new tables you got or the old one from the example. Can you tell where is the location in which you replaced the tables. 

This script to compute efficiency map is developed to work with the axial Flux macros, so I am not sure if it compatible with the one "CreateLookUpTableFromTMProjectDQLight". I need to check more on this. I will see if someone else can take look at your question.  

 

Hi,

The zip file was for the old example that I sent. You can find the new one here.

Please let me know if you find a macro compatible with IPM machine. Because I need a solution to plot an efficiency map in a few hours.

Thanks

 

Mohammed Elamin_22169

ramin_21612 said:

Hi,

The zip file was for the old example that I sent. You can find the new one here.

Please let me know if you find a macro compatible with IPM machine. Because I need a solution to plot an efficiency map in a few hours.

Thanks

 

Ramin, 

I am not sure if you are aware of this. But we do have efficiency map computation within Flux. This should more accurate than the one depending on the tables as it is based on transient FE model. If you use CDE and parallel computation you should be able to get the results within hours. You can check the example below for how it works. 

 

ramin_21612

Mohammed Elamin_22169 said:

Ramin, 

I am not sure if you are aware of this. But we do have efficiency map computation within Flux. This should more accurate than the one depending on the tables as it is based on transient FE model. If you use CDE and parallel computation you should be able to get the results within hours. You can check the example below for how it works. 

 

Mohammed,

 

To compute efficiency map, I used the Flux e-Machine toolbox (FeMT), but it worked in Transient Magnetic conditions and it took a long time (about two days with my laptop). Furthermore, the result was reported in pdf format which made it impossible to modify the graph or plot the data in Matlab. I just trying to find a new solution within few hours.

Mohammed Elamin_22169

ramin_21612 said:

Mohammed,

 

To compute efficiency map, I used the Flux e-Machine toolbox (FeMT), but it worked in Transient Magnetic conditions and it took a long time (about two days with my laptop). Furthermore, the result was reported in pdf format which made it impossible to modify the graph or plot the data in Matlab. I just trying to find a new solution within few hours.

FYI, You can export the data of the maps on text format if needed.