fminsearch ''index exceeds matrix dimensions''

I am trying to find an optimal fit to my impedance spectroscopy measurement data using fminsearch

When I am running the code I am getting the following error

??? Index exceeds matrix dimensions.

Error in ==> @(x)myfun1(x,omega_s,Z_60(start:stop),Rinit)

Error in ==> fminsearch at 205 fv(:,1) = funfcn(x,varargin{:});

Error in ==> SOH31_Script_V4 at 152 [Y ERR] = fminsearch(@(x) myfun1(x,omega_s,Z_60(start:stop),Rinit),X0,options)

I think the error is due to the start and stop value I am using. I have the measurement of impedances (real, imaginary and the frequencies) for a large range of frequencies (10 mHz to 50kHz)

myfun1 is the error function I am minimizing

I am trying to fit this data to a model. I don't want the optimization problem to run for all frequencies so i want to restrict the start stop to a mid frequency range.

How can I select start and stop?

All the real, imaginary and frequencies are of size 68X1 double

Thanks!

Answer 1

[Matt J]

"stop" must be selected to be no greater than length(Z_60)

[Matt J]

It is also more optimal for you to pre-assign Z_60(start:stop) when setting up your objective function

 Zdata=Z_60(start:stop);

 fminsearch(@(x) myfun1(x,omega_s,Zdata,Rinit),X0,options)

Otherwise, the data extraction operation Z_60(start:stop) will be repeated every time the anonymous function is evaluated.

[Avnish]

Actually I have 68 measurements of impedance and

Frequency vector for fitting is

f_int = logspace(log10(50000),-2,200);

then I have

%Interpolated complex impedance

Z_60 = interp1(f_60,Z_60,f_int,'linear');

Size of Z_60 is 1X200 double

[Matt J]

OK, well I'm glad the problem is solved.

[Avnish]

Matt : For Start I did like this

%Select frequency range
[tmp start]=min(real(Z_60)); %Point at minimum real part (Its a nyquist plot, Imaginary vs real)

start=max(1,start-8); %I ignored first 8 points of high frequency region in the optimization

What do I do with stop? using stop = length(Z_60) gives strange results