Estimate first order transfer function from Phase at Frequency

Question

John 2022-3-22

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1678089-estimate-first-order-transfer-function-from-phase-at-frequency

评论： Paul 2022-3-23

What command would I use to estimate a first order transfer function (TF), from a known Phase at Frequency?

The TF is known order (first-order), known gain of 1 at DC, and monotically decreasing phase (ie only passes through a phase value once).

Let's say the given data point is a phase P at freq F (say, -45 deg at 100 Hz).

How would I programatically find the TF that includes this datapoint?

The TF should be unique, once the TF order, DC gain, monotically decreasing phase is assumed, and P&F are defined.

I looked at tfest using iddata, and other similar commands, but non are exactly what I'm looking for...

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Paul 2022-3-23

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1678089-estimate-first-order-transfer-function-from-phase-at-frequency#answer_924584

@John

Based on the problem statement can you write the symbolic form of a first order transfer function, H(s)? How many free parameters does H(s) have? Given that angle(H(1j*(2*pi*100)) = -45*pi/180 (example data from the Question), can you solve for the free parameter(s)? Don't make the problem more complicated than it needs to be.

2 个评论
显示无隐藏无

John 2022-3-23

Apologies, I'll clarify: this is for arbitrary N-order TF, since the 1st-O example is simple (as you pointed out).

I'm looking for a more general programmatic way to find what TF(s) could fit, with a defined phase/freq point.

Paul 2022-3-23

Oh, well that's a bit different. Specifying the phase at a single frequency is not enough information to uniquely specify the parameters of an arbitrary n-th order TF, even with the stated constraints. The problem will require additional constraints or some cost that to be optimized among all TFs that satisfy the stated constraints. Additionally, I think it could be very tricky to define the constraints on the TF to enforce the monotonically decreasing phase for a general TF.

请先登录，再进行评论。