DC-DC converter control problem using Reinforced Learning Toolbox

When a DC-DC converter model is connected to the Reinforced Learning Toolbox (for example, DDPG algorithm),

the simulation of the DC-DC converter failed.

Is the simulation time step is too small (Sampling time=0.0001sec and final time=1 sec)?

If there is no Reinforced Learning module, the simulation is okay.

Answer 1

[Manikanta Aditya]

Hi Kim,

As per my understanding, you are interested to know how to resolve the DC-DC converter model failure when connected to Reinforced Learning Toolbox.

As a workaround, you can change the ‘Fixed-step size’ setting in the Simulink model configuration parameters and increase the simulation time step to 0.001 or 0.01 seconds and see if that helps. If it doesn’t resolve, we need to further investigate the specific error message you’re getting to determine the root cause of the issue.

If the problem persists, please reach out to our support team at:

Contact Us - MATLAB & Simulink (mathworks.com)

Please refer to the following documentation for more info on:

• ‘Fixed-step size’: https://www.mathworks.com/help/simulink/gui/fixedstepsizefundamentalsampletime.html

I hope this resolves the issue you were facing.

Answer 2

[duoke]

Can you share your model attachments to help me see if it can solve the problem?