Hi all,

I have got a theoretical question concerning the coherence function as defined within the TurbSim user’s manual.

Preamble: by using Matlab, I am trying to code my own wind file generator, in accordance with Sandia theory and Kaimal spectrum. Now, assuming that N = Ny*Nz, I generally calculate a distance matrix with size b[/b]; the frequency array is, instead, a NumFreq*1 array.

Therefore, when applying the coherence formula, there is a clear mismatch between frequency vector size and distance matrix size: I expect the coherence matrix to be sized as the distance matrix, that means a N*N coherence matrix.

In which way may I overcome this size issue? Would you mind to shed a light? My code, up to now, looks like this:

[code]clc, clear all, close all

%% INITIALIZATION

tic

HubHt = 110; % Hub Height

GridWidth = 150; % Grid length along Y axis

GridHeight = 150; % Grid length along Z axis

RotorDiameter = min(GridWidth,GridHeight); % Turbine Diameter

Ny = 7;

Nz = 7;

N = Ny*Nz;

AnalysisTime = 600;

UsableTime = 40; % Time as output

TimeStep = 0.05;

Uhub = 8; % Average wind speed at hub height

TI = ‘A’; % Turbulent intensity

%set delta1 according to guidelines (chap.6)

if HubHt < 60

lambda1 = 0.7*HubHt;
else
lambda1 = 42;
end
L = 0.8*lambda1;%IEC, (B.12)

%IEC, Table 1, p.22

%IEC, sect. 6.3.1.3

b=5.6;

if isequal(TI,‘A’)

Iref = 0.16;

sigma1 = Iref*(0.75

*Uhub + b);*

elseif isequal(TI,‘B’)

Iref = 0.14;

sigma1 = Iref(0.75

elseif isequal(TI,‘B’)

Iref = 0.14;

sigma1 = Iref

*Uhub + b);*

elseif isequal(TI,‘C’)

Iref = 0.12;

sigma1 = Iref(0.75*Uhub + b);

elseif isequal(TI,‘C’)

Iref = 0.12;

sigma1 = Iref

else

sigma1 = str2num(TI)*Uhub/100;

end

```
% sigma1=Iref*(0.75*Uhub+b);
sigma2=0.8*sigma1;
sigma3=0.5*sigma1;
%derived constants
l=0.8*lambda1; %IEC, (B.12)
xLu = 8.1*l; % Integral length scale along X axis
xLv = 2.7*l; % Integral length scale along Y axis
xLw = 0.66*l; % Integral length scale along Z axis
```

% Frequency defition

NumOutSteps =ceil((UsableTime + GridWidth/Uhub)/TimeStep);

NumSteps =max(ceil(AnalysisTime/TimeStep), NumOutSteps);

INumSteps =1.0/NumSteps;

NumFreq =NumSteps/2;

DelF =1.0/(NumSteps*TimeStep);
DelF5 =DelF*0.5;

f= ((1:NumFreq)*DelF);

%% GRID DEFINITION

dy = GridWidth/(Ny-1);

dz = GridHeight/(Nz-1);

if isequal(mod(Ny,2),0)

iky = [(-Ny/2:-1) (1:Ny/2)];

else

iky = -floor(Ny/2):ceil(Ny/2-1);

end

if isequal(mod(Nz,2),0)

ikz = [(-Nz/2:-1) (1:Nz/2)];

else

ikz = -floor(Nz/2):ceil(Nz/2-1);

end

[Y Z] = ndgrid(iky*dy,(ikz*dz + HubHt));

% define distance matrix

dist = pdist2([Y(:), Z(:)], [Y(:), Z(:)]);[/code]