Hi All

I am attaching three figures.

1. Expected coherence results along the span length of a cylinder.

2. Obtained from my simulation

3. Equation for Coherence.

I have tried 'mscohere' to get coherence^2 directly. Also 'cpsd' in matlab to get coherence as an alternative. Also i tried to get correlation value (Rpp) from these signals. In all casess, values were near equal to one and not reducing as expected.

I believe my simulation is right based on the

(1) mean Cp distribution around the cylinder (compared with experiment )

(2) Strouhal number is matching with experiment

(3)mean Cd, Cl rms are matching.

%%%%%%%%%%%%%%%%%%%%%

statp_Gey_1 - pressure fluctuation at 1 point

statp_Gey_2 - pressure fluctuation at 2nd point

mscohere- gives magnitude square coherence

[cxy,f]=mscohere(statp_Gey_1,statp_Gey_2,hamming(window),noverlap,window,Fs);

cxy=sqrt(cxy);

f_stat=f*D/v;

%calculation using cpsd (cross power spectral density)

[Pxy,f_man]=cpsd(statp_Gey_1,statp_Gey_2,hamming(window),noverlap,window,Fs);

[Pxx,f_man]=cpsd(statp_Gey_1,statp_Gey_1,hamming(window),noverlap,window,Fs);

[Pyy,f_man]=cpsd(statp_Gey_2,statp_Gey_2,hamming(window),noverlap,window,Fs);

Pxy_abs=abs(Pxy);

f_man=f_man*D/v;

cc_num=abs(Pxy);

cc_denom=(sqrt((abs(Pxx)).*(abs(Pyy))));

cc=cc_num./cc_denom;

cc_sqr=sqrt(cc);

%Correlation value calculation (Rpp)

num(gg) = mean(statp_Gey_1.*statp_Gey_2);

denom(gg)= sqrt(mean(statp_Gey_1.^2))*sqrt(mean(statp_Gey_2.^2));

Rpp (gg) = num(gg)/denom(gg);

Pressure monitored is 'Static pressure' in Ansys Fluent.

Various points are monitored along the span of cylinder. 'Area weighted average' of the static pressure was taken. I doubt whether phase information is lost in taking Area weighted average of pressure.