Simplistic PV array circuit with constant resistance keeps oscillating

I built an extremely simple circuit to play with the PV array block. The PV array is connected to a resistor. The simulation mode is discrete with step size 5 ms. The voltage, current of the PV array are measured and the correponding power is calculated.

Let's test four values of resistance.

1. R = 1

As we see above, everything works well when R = 1.

2. R = 29 / 7.35 = 3.9456

Why is the above value chosen? Because it leads to the maximum power point (MPP) according to the PV parameters.

Rerun the simulation with R = 3.9456, we get the following result:

The output of the PV simply keeps oscillating acutely.

After more trials, I find that once the resistance approaches the above MPP one, the PV output starts oscillation. The PV output seems rather sensitive around this point. Please refer to the below test for details.

3. R = 3.8

As shown in the above section, R = 3.9456 results in severe oscillations. Now let's reduce the resistance a little to R = 3.8. Despite this slight change, the PV array returns to the normal working state (see figure below).

We also notice that the power 212.6 is indeed close to the MPP one (29 * 7.35 = 213.15).

Other resistance values like one greater than 4.0 also leads to similar oscillations. You may play with the attached SIMULINK model to check it.

4. R = 4.5

Once the resistance is larger than the MPP one to a certain extent, the simulation even fails. The output of the PV is NaN. For instance, if we set R = 4.5, the scope output is shown below (i.e., on valid output)

Overall, it seems the PV array output is extremely sensitive to the load (here simply resistance) that is close to or greater than the MPP one.

Did I missing anything in the simulation? Why does a resistance close to the MPP one lead to strong oscillations? How to fix this issue? Any suggestion is greatly appreciated. Thank you.

Answer 1

[Joel Van Sickel]

Hello Shuhua,

the specialized power system solar panel block is fairly sensitive when used with a discrete simulation. There are two paths forward to get your results to not oscillate for these larger resistances. One would be to change your simulaiton model to continuous and use a variable step solver like ode23tb.

If you do not want to use a variable step solver, instead, you can go to the preferences tab in the powergui options, and at the bottom, you should see an option for solver details for nonlinear elements. Here you can select the check box for continue simulaiton if maximum number of iterations is reached. Then go to the solar panel model, and on the advanced tab, select to use the robust method (you will not be able to vary temperature during simulation) and you will also get correct performance.

Accurately capturing the types of nonlinearities in this particular model is challenging for the specialized power systems implementation. If you want to avoid these issues, you can change to simscape technology: https://www.mathworks.com/help/physmod/sps/ref/solarcell.html;jsessionid=b57f65051e2cb436dc5cba6033fb. However, the specialized power systems solar panel model is much more user friendly as it lets you select a number of preexisting solar panels, so for that reason, you may want to stick with specialized power systems.

Regards,

Joel

[Shuhua Gao]

Hi, Joel,

Thank you very much for your quick response. Got it. As you suggested, I will change to the continuous-time mode.

Shuhua

[Guerrida]

HI Shuhua

What did you do with your simulank issue ,pleas ,did you solve it ??/?