This is still the small signal model. Constructing a true large signal model would look different and we wouldn't be able to use the same control techniques.
A small signal model is a linearization around a specific operating point (or quiescent point). That is we assume we're operating with some DC operating conditions (Vo, Io, D, Vg). The small signal model is something that tells us what happens when we make small perturbations to these variables at this operating point i.e. when we don't move too far from these DC voltages and currents. The small signal model is analogous to using a derivative to estimate the change of a curve at some point. Imagine you're on a quadratic curve at the x = 1. If we move slightly in the x direction we can use the derivative as the slope of a first order model (the line y = 2x-1 approximates the curve y = x^2 at x = 1) to estimate how far we will travel in the y direction from this perturbation, but only for small distances. If we are instead at say x = 2, then the model looks different because the slope at this point is different (the line y = 4x - 4 would be a better estimate). This model can be used to estimate variations around this specific point but gets worse and worse the farther from the operating point you go. A large signal model really just describes what happens to changes in y when there are large changes in x. The DC model is in a sense a large signal model, but the DC models we've derived do not take into account the dynamics of the passive elements. A further problem is that it is often not linear and as a result it is more difficult to design systems to accurately and stably control it. For the quadratic example, you could say the large signal model is simply y = x^2. No matter what the variation in x is, you can accurately predict what the change in y will be, x^2 just happens to be nonlinear.
@@timmcrae3831 thanks for this. I am not looking for the large-signal model to design a controller, but to know how it looks like. I understood what you said and I think we go for small-signal if the large signal is non-linear. because large-signal, if it considers dynamics, not the steady state only, so it is enough .... That is why I guessed that the model at 18:08 kinda large-signal model but it consider D as constant which we looking for as a control signal ... I will keep looking to solidify my understanding
![State Space Model Example w/ MATLAB Code + Helper Files (Boost Converter) -- [ep. 02]](http://i.ytimg.com/vi/pMeJTRe_pGU/mqdefault.jpg)
This is the simplest explanation I have found on entire youtube. Thank you so much!
Great Thanks.. please please continue your lectures
Great video, please could you post small signal model of dual active bridge converter
Thanks for the Great Video!, I hope you can do a video over the small-signal model of the DAB converter in the near future.
Have you seen a video on the small signal model of the DAB yet?
@@oliviannadi5046 unfortunately not 😔
Great. Is the model you got at 18:08 called large-signal model ?
This is still the small signal model. Constructing a true large signal model would look different and we wouldn't be able to use the same control techniques.
@@timmcrae3831 Thanks and I would appreciate to explain the difference because no source made it clear between them
A small signal model is a linearization around a specific operating point (or quiescent point). That is we assume we're operating with some DC operating conditions (Vo, Io, D, Vg). The small signal model is something that tells us what happens when we make small perturbations to these variables at this operating point i.e. when we don't move too far from these DC voltages and currents.
The small signal model is analogous to using a derivative to estimate the change of a curve at some point. Imagine you're on a quadratic curve at the x = 1. If we move slightly in the x direction we can use the derivative as the slope of a first order model (the line y = 2x-1 approximates the curve y = x^2 at x = 1) to estimate how far we will travel in the y direction from this perturbation, but only for small distances. If we are instead at say x = 2, then the model looks different because the slope at this point is different (the line y = 4x - 4 would be a better estimate). This model can be used to estimate variations around this specific point but gets worse and worse the farther from the operating point you go.
A large signal model really just describes what happens to changes in y when there are large changes in x. The DC model is in a sense a large signal model, but the DC models we've derived do not take into account the dynamics of the passive elements. A further problem is that it is often not linear and as a result it is more difficult to design systems to accurately and stably control it. For the quadratic example, you could say the large signal model is simply y = x^2. No matter what the variation in x is, you can accurately predict what the change in y will be, x^2 just happens to be nonlinear.
@@timmcrae3831 thanks for this. I am not looking for the large-signal model to design a controller, but to know how it looks like. I understood what you said and I think we go for small-signal if the large signal is non-linear. because large-signal, if it considers dynamics, not the steady state only, so it is enough .... That is why I guessed that the model at 18:08 kinda large-signal model but it consider D as constant which we looking for as a control signal ... I will keep looking to solidify my understanding
Awesome 😊