Grid connected cascaded PV ivnerters
I'm trying to implement a grid connected cascaded inverter similar to the photo below. I am currently testing it first with a single system. I have tested my voltage and current controls with a constant DC source and it looks good. However, when I change it to a pv with boost converter, the DC link voltage just keeps on increasing.
System topology
voltage and current loops
Answers
-
do you have MPPT for the boost interface dc-dc to the panel or some other supervisory control to ensure you are operating at a proper power point?
Do you have a supervisory control to ensure that the two grid forming converters are sharing the load properly and within the bounds of the power available from the PV source?
For your PR controllers, do you have sensible limits for the integration?
0 -
Albert Dunford_20896 said:
do you have MPPT for the boost interface dc-dc to the panel or some other supervisory control to ensure you are operating at a proper power point?
Do you have a supervisory control to ensure that the two grid forming converters are sharing the load properly and within the bounds of the power available from the PV source?
For your PR controllers, do you have sensible limits for the integration?
Hi!
Yes. I am implementing MPPT in the DC-DC through a C block. I have tested it separately with an RL load without being connected to tbe inverter and it works okay.
I have not yet implemented cascaded topology. I am still doing it in a single pv-boost-inverter-grid system as I am trying to do the overall control structure first. The topology is communicationless so each DG has the controller illustrated above.
I have not used limits in the integration. all of it use -inf and inf limits.
0 -
Hi!
Yes. I am implementing MPPT in the DC-DC through a C block. I have tested it separately with an RL load without being connected to tbe inverter and it works okay.
I have not yet implemented cascaded topology. I am still doing it in a single pv-boost-inverter-grid system as I am trying to do the overall control structure first. The topology is communicationless so each DG has the controller illustrated above.
I have not used limits in the integration. all of it use -inf and inf limits.
ah i understand it is just a single system so far. your DC-DC is working properly with MPPT control? you have changed the inputs to the PV panel sunlight and panel temp and it is finding the right point?
For the inverter does it connect to grid properly? how are you synchronizing? You can try with the simulation trick of running synchronized reference vs grid instead of using a PLL to remove this complexity.
Are you using PR as you hope to remove harmonics eventually? does your system work with more conventional PI controller for inner current outer voltage?
How did you do the control design? smartctrl?
0 -
Albert Dunford_20896 said:
ah i understand it is just a single system so far. your DC-DC is working properly with MPPT control? you have changed the inputs to the PV panel sunlight and panel temp and it is finding the right point?
For the inverter does it connect to grid properly? how are you synchronizing? You can try with the simulation trick of running synchronized reference vs grid instead of using a PLL to remove this complexity.
Are you using PR as you hope to remove harmonics eventually? does your system work with more conventional PI controller for inner current outer voltage?
How did you do the control design? smartctrl?
I have kept the temperature constant but i have tested with varrying insolation level and its able to follow the output PV power. I have also changed the MPPT control to the one used in the P&O sample of PSIM.
For the inverter, I first used it with a constant DC source and it synchronizes with the grid properly. Yes, PR was used hoping to remove harmonics and I was able to achieve it. I have not tried testing it with conventional PI control. The gain values were mostly first manually calculated then fine tuned using parameter sweep and oscilloscopes.
0 -
I have kept the temperature constant but i have tested with varrying insolation level and its able to follow the output PV power. I have also changed the MPPT control to the one used in the P&O sample of PSIM.
For the inverter, I first used it with a constant DC source and it synchronizes with the grid properly. Yes, PR was used hoping to remove harmonics and I was able to achieve it. I have not tried testing it with conventional PI control. The gain values were mostly first manually calculated then fine tuned using parameter sweep and oscilloscopes.
Just because the converter runs with in ideal source does not mean it will run with a panel model which is very non-linear.
- try and slow your control down,
- Check that your converter works across a range of operating voltages, try overlaying a small AC signal to your DC bus to see how it reacts. perhaps you should do a full loop gain simulation to see what the control response looks like help->tutorial videos->power supply design
- "cheat" and remove grid synchronization by using a forced angle that is already synch'd with the grid
- are you able to zip your simulation and post it?
0 -
Albert Dunford_20896 said:
Just because the converter runs with in ideal source does not mean it will run with a panel model which is very non-linear.
- try and slow your control down,
- Check that your converter works across a range of operating voltages, try overlaying a small AC signal to your DC bus to see how it reacts. perhaps you should do a full loop gain simulation to see what the control response looks like help->tutorial videos->power supply design
- "cheat" and remove grid synchronization by using a forced angle that is already synch'd with the grid
- are you able to zip your simulation and post it?
Hi.
By slowing the controller, do you mean adding delay blocks?
With an ideal DC source, the inverter works between 150V and 60V, I designed the inverter based on a 60V supply and my ac source is at 60V. However, when connected with a PV-Boost, the inverter seems to work fine even if the DC voltage goes up to 250V.
Yes, I can upload the zip file and add the papers I'm basing from as well. I have made some changes to the PR controlls already since my last upload.
Thank you very much for your responses and help Albert!
0 -
Hi.
By slowing the controller, do you mean adding delay blocks?
With an ideal DC source, the inverter works between 150V and 60V, I designed the inverter based on a 60V supply and my ac source is at 60V. However, when connected with a PV-Boost, the inverter seems to work fine even if the DC voltage goes up to 250V.
Yes, I can upload the zip file and add the papers I'm basing from as well. I have made some changes to the PR controlls already since my last upload.
Thank you very much for your responses and help Albert!
Hi Jayvie,
By slowing the controller i mean reducing the gain of the P value. Don't add additional delay!
0 -
Hi, Jayvie,
Your booster controller is not maintaining the DC bus voltage stable. This instability affects the inverter operation, too.
Please check the details and the parameters of your schematic. Maybe adding a little resistance in series with the inductors would provide some damping.
Copying other people's successful design may or may not work in your PSIM schematic. There are many details you have to get correct.
Sincerely,
Rhonda
0 -
Albert Dunford_20896 said:
Hi Jayvie,
By slowing the controller i mean reducing the gain of the P value. Don't add additional delay!
Hi @Albert Dunford ,Sorry I have not been able to update you recently. I've been working on adjusting the P values and have settled on the ones in the attached PSIM file as of now. As @Rhonda mentioned, I also added resistors in series with the inductors to provide damping. However, I've noticed an issue: the DC voltage drops to about 80V (the peak AC voltage), the modulation signal increases significantly, and the control system cannot recover. I believe this is because the error and reference current continuously increase, even though the actual reference angular frequency fluctuates around the rated value.
I managed to achieve a stable DC link voltage by placing limits on my voltage PR controller integrator blocks. However, I'm unsure if this is the correct solution, as limiting the blocks might indicate that the problem lies elsewhere.
Best regards,
Jayvie
0 -
Rhonda_20369 said:
Hi, Jayvie,
Your booster controller is not maintaining the DC bus voltage stable. This instability affects the inverter operation, too.
Please check the details and the parameters of your schematic. Maybe adding a little resistance in series with the inductors would provide some damping.
Copying other people's successful design may or may not work in your PSIM schematic. There are many details you have to get correct.
Sincerely,
Rhonda
Hi @Rhonda
I have added resistors in series with the inductors to provide damping. However, I've noticed an issue: the DC voltage drops to about 80V (the peak AC voltage), the modulation signal increases significantly, and the control system cannot recover. I believe this is because the error and reference current continuously increase, even though the actual reference angular frequency fluctuates around the rated value.
I managed to achieve a stable DC link voltage by placing limits on my voltage PR controller integrator blocks. However, I'm unsure if this is the correct solution, as limiting the blocks might indicate that the problem lies elsewhere.
Best regards,
Jayvie
0
Altair Employee