MATLAB Answers

0

Simscape Linear Analysis Tool, Steady state for switching circuit

Asked by Christian Olsen on 16 Mar 2019
Latest activity Answered by Arkadiy Turevskiy on 18 Jun 2019
Hi people,
seems I have a big problem here.
I am trying to analyse a simple boost converter created using SIMSCAPE components, which I find in the 'Simscape' > 'Foundation library' > 'Electrical'.
After simulating the circuit and finding the converter was working in the so called CCM condition (as you can see from 'Image1.png) I wanted to extract the steady state operating point using the 'Linear Analysis Tool' provided by Simulink, so I opened the related GUI, as pictured in 'Image2.png'.
In the drop-down menu called 'Operating Point' I selected 'Trim model', opening the window showing a table of states.
In the 'Value' column I found automatic guesses of the steady state states' values with the one related to the output capacitor voltage already set to 360 Volt.
One question is: what are those last four states standing for? I see they are coming from the 'Solver Configuration' block but I don't know if it is correct to put the 'steady state' constraint on those too. I really need an answer for this.
Anyway, let's go on.
After typing my guess for the steady state value of the inductor current, which I supposed to be 10 Ampère, (you can see this in 'Image3.png') I clicked on the 'Start trimming' button at the bottom of the window. After a while I got the results of the analysis in the 'op_trim1' structure along with the warning message you see in the picture.
Clicking on 'Details' ('Image4.png') I saw the values for the states and their derivatives which seemed insane.
Obviously the results I obtained are wrong.
I don't think Simulink has understood that I am simulating a switching converter, where all the state variables don't reach a constant value but rather they show some ripple over the mean value I am interested to.
When talking about these type of circuits what the control designer is looking for is the mean value of the state variables, like inductor currents and capacitor voltages, and the steady state point from the point of view of these mean values. In my particular case these values are 360 Volt for the capacitor and 10 Ampère for the inductor.
I am here to ask how I could use the linear analysis tool for dicsovering the steady state mean values of the reactive passive components of a switching circuit.
Every single tutorial I found in internet talk about ordinary circuits or simulink model, none of these tutorials focus on switching circuits.... this is not good...

  1 Comment

please someone help me, I feel like I'm in the middle of the Pacific ocean aboard a little raft.... :(

Sign in to comment.

1 Answer

Answer by Arkadiy Turevskiy on 18 Jun 2019

Hi Christian,
Unfortunately Linear Aanalysis Tool does not currently support specification-based computation of periodic operating points, which is what you are interested in. Your boost converter model is in steady-state in the sense that voltage and current values stay at the same average value, but actual signal derivatives are not zeros dues to switching in power semiconductors.
I want to offer two-fold response.
  1. We know periodic operating point capability is needed, and are working on it.
  2. In the mean time you can try the following: create a variant of the model using average-value blocks, like this one. You will be able to trim the average value model.Another approach is to run your model to steady state and compute a steady state operating point using simulation snapshot. Next time you run the simulation and want to start from steady-state, initialize from this snapshot.See here.

  0 Comments

Sign in to comment.