I am trying to run a simulation with a simplified version of the IEEE 13 bus system. The model was adapted from the "IEEE_13_Node_With_House.glm" file as given in the tutorials. I simplified this system by replacing the regulator, transformer and switch all with regular overhead lines similar to other lines in the system. So the system has a single nominal volage of 2401.77 at the primary distribution level, and 120 at the secondary level. I wrote a MATLAB script that populates this system with homes at different buses in the system. In my first implementation, I attached 10 homes at each possible connection point (e.g., 10 homes at Bus 1 Phase A, 10 homes at Bus 1 Phase B...10 Homes at Bus 12 Phase C). Each home is attached to its own triplex meter and triplex line, to appropriate triplex nodes in the system. So, for example, the triplex node (connected to the system via a single-phase center-tapped transformer) at Bus 1-A has 10 homes attached to it. Hopefully that's enough background information.
Anyway, my model is not solving. The FBS fails to converge, while NR is having parent-child issues that I cannot seem to alleviate. I have attached my model to this message. It seems like this is a relatively simple setup and so I'm hoping my problem is a minor one.
I fixed my child-parent issue and now it's failing to converge at my swing node (bus0). I've re-attached my model. Any insight you may have is most appreciated.
The main issue were in my parent-child definitions. Some of my homes had solar objects attached to them and I guess I coded this improperly. Here's an example of my mistake:
object house {
name house_5_1_b_3;
parent meter_5_1_b_3;
groupid bus_5_1;
floor_area random.normal(2500,300);
thermal_integrity_level GOOD;
cooling_setpoint random.normal(72,1);
heating_setpoint random.normal(65,1);
object waterheater {
name WH_5_1_b_3;
groupid bus_5_1;
tank_volume random.normal(50,5);
tank_diameter random.normal(3.78,0.5);
};
// Solar Block
object triplex_meter {
name solar_meter_5_1_b_3;
groupid solar_meter_5_1;
phases AS;
nominal_voltage 120;
object inverter {
name solar_inverter_5_1_b_3;
phases AS;
generator_mode CONSTANT_PF;
generator_status ONLINE;
inverter_type PWM;
power_factor 1;
object solar {
name solar_sys_5_1_b_3;
generator_mode SUPPLY_DRIVEN;
generator_status ONLINE;
area random.normal(250,50);
};
};
};
}
Although I tried to nest the solar system under the house, it needs to utimately be parented to a meter or link object. My fix was to parent the solar triplex meter to the main home meter, so I included the following in my solar meter:
parent meter_5_1_b_3;
Now, the home meter will show the net load of home with solar system. One can attach other DG, like a battery, in the same manner.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Since the post is over 5 years old, I'd be surprised if Cody responds.
There are some feeder generators/population scripts under the GridLAB-D Git repository, or there's one part of the Transactive Energy Systems Project (specifically located here) that people have had luck with in the past (and is much newer than the scripts in the GridLAB-D repository).
-Frank
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi guys,
I am trying to run a simulation with a simplified version of the IEEE 13 bus system. The model was adapted from the "IEEE_13_Node_With_House.glm" file as given in the tutorials. I simplified this system by replacing the regulator, transformer and switch all with regular overhead lines similar to other lines in the system. So the system has a single nominal volage of 2401.77 at the primary distribution level, and 120 at the secondary level. I wrote a MATLAB script that populates this system with homes at different buses in the system. In my first implementation, I attached 10 homes at each possible connection point (e.g., 10 homes at Bus 1 Phase A, 10 homes at Bus 1 Phase B...10 Homes at Bus 12 Phase C). Each home is attached to its own triplex meter and triplex line, to appropriate triplex nodes in the system. So, for example, the triplex node (connected to the system via a single-phase center-tapped transformer) at Bus 1-A has 10 homes attached to it. Hopefully that's enough background information.
Anyway, my model is not solving. The FBS fails to converge, while NR is having parent-child issues that I cannot seem to alleviate. I have attached my model to this message. It seems like this is a relatively simple setup and so I'm hoping my problem is a minor one.
Thank you for your help,
Cody
I fixed my child-parent issue and now it's failing to converge at my swing node (bus0). I've re-attached my model. Any insight you may have is most appreciated.
Thank you,
Cody
Sorry guys - more false alarms from me. It looks like I was able to fix my issue and the simulation is running (for now!).
Thanks,
Cody
Hello Cody,
Glad you figured it out. If you have any quick insights on what you did to fix it, it may be useful to post those here for future users.
-Frank
The main issue were in my parent-child definitions. Some of my homes had solar objects attached to them and I guess I coded this improperly. Here's an example of my mistake:
object triplex_meter {
name meter_5_1_b_3;
groupid house_meter_5_1;
phases AS;
nominal_voltage 120;
}
object house {
name house_5_1_b_3;
parent meter_5_1_b_3;
groupid bus_5_1;
floor_area random.normal(2500,300);
thermal_integrity_level GOOD;
cooling_setpoint random.normal(72,1);
heating_setpoint random.normal(65,1);
object waterheater {
name WH_5_1_b_3;
groupid bus_5_1;
tank_volume random.normal(50,5);
tank_diameter random.normal(3.78,0.5);
};
// Solar Block
object triplex_meter {
name solar_meter_5_1_b_3;
groupid solar_meter_5_1;
phases AS;
nominal_voltage 120;
object inverter {
name solar_inverter_5_1_b_3;
phases AS;
generator_mode CONSTANT_PF;
generator_status ONLINE;
inverter_type PWM;
power_factor 1;
object solar {
name solar_sys_5_1_b_3;
generator_mode SUPPLY_DRIVEN;
generator_status ONLINE;
area random.normal(250,50);
};
};
};
}
Although I tried to nest the solar system under the house, it needs to utimately be parented to a meter or link object. My fix was to parent the solar triplex meter to the main home meter, so I included the following in my solar meter:
parent meter_5_1_b_3;
Now, the home meter will show the net load of home with solar system. One can attach other DG, like a battery, in the same manner.
Hi Cody,
Can you please share your matlab code that populates the model with houses? Also, do you have the correct version of the GLM file?
Hello Abdullah,
Since the post is over 5 years old, I'd be surprised if Cody responds.
There are some feeder generators/population scripts under the GridLAB-D Git repository, or there's one part of the Transactive Energy Systems Project (specifically located here) that people have had luck with in the past (and is much newer than the scripts in the GridLAB-D repository).
-Frank
Hi Dr. Tuffner,
Thank you for replying to this thread! I would definitely look into this.
-Mamun