Design Considerations | Design and Test a Grid-Tied Solar Inverter Controller, Part 1

My name is Jonathan LeSage and I'm an application engineer with MathWorks. Our goal today is going to be to implement a solar inverter control system that can adhere to grid codes and test this so that we have some confidence in our control strategies that we've designed.

Typically, if we were to implement this and actually deploy this to the field, the ultimate target would be potentially a combination of some kind of DSP type device, but also potentially a PLC as well. And these devices are going to have to integrate in with the actual hardware, in this case solar panels, three phase inverters, harmonic filters, and then the grid itself. and so this is going to require software communication.

Now techniques like hardware in the loop can actually help us improve this process. What hardware in the loop allows us to do is rather than having this three phase inverter, the solar panels out in the field, we can replace all of that actual hardware with a simulation equivalent of it that runs in real time. And this would be a virtualized simulation that emulates as if it is a real three phase inverter and solar array system.

Now the nice thing here is the controller, when it's interacting with the system, it only sees it as if it's a real-time system. Now hardware in the loop gives you a number of advantages. We're not having to interface with that expensive hardware so early on in our design process. This lets us test a lot of different things very quickly.

Now for certain things like fault scenarios, these are very hard to necessarily replicate on an actual hardware. With a virtual simulation of the inverter system, we can actually replicate all sorts of different fault scenarios and test voltage ride through situations with our utility scale solar array control system.

Again, because we're working with a digital simulation, we're not looking necessarily at the high voltage and high current power electronics hardware. Although there are avenues to kind of do what we call power hardware in the loop as well.

And then finally, because we're doing a lot of this design up front and we have this virtual simulation that we can do a lot of tests with, we can do a lot of this testing much earlier in the design process.

Now I've talked about fault ride through. And fault ride through is a very important concept, especially now that we're starting to see more and more utility sales, solar and wind generation connecting to the grid. And essentially what this is all about is if there's a voltage or frequency grid event that occurs, everybody might start to trip at the same time once they see that big voltage dip. And of course, this is only going to cascade the problem. We're going to see additional generation down the line also start to trip as well, further causing the voltage to drop.

And so what everyone has started to agree on are different grid codes. And these are kind of different across the world. But many regions of the world have different grid codes. And these essentially meet-- these essentially force us to look at if a grid event does occur, under what scenarios does my utility scale solar system need to stay connected to the grid? So under certain grid events, we're going to have to stay connected. We can't just disconnect and ride this out on our own.