Help modifying a User-defined-function

17 Ansichten (letzte 30 Tage)
Vivek
Vivek am 20 Apr. 2020
Kommentiert: Ameer Hamza am 21 Apr. 2020
I am trying to compute the mean max and mean min amplitude of time series datasets. These timeseries are arranged in each column of a matrix, so that the matrix is 351x117 . So 117 timeseries. I have found a user defined function that can do just that.The problem is that this UDF is meant for only 1 timeseries array at a time, x. How can I modify the following UDF to accomodate for a matrix,A, instead of an array x. Note that the desired output will be in temp78. I will need to compute the average of the fist two columns of temp78 in order to find the avg max and min amplitudes for every timeseries.
  2 Kommentare
Walter Roberson
Walter Roberson am 20 Apr. 2020
It is easier for us to work with code than with a picture of code.
Vivek
Vivek am 20 Apr. 2020
Here is it. Any help is much appreciated.
% OUTPUT
% temp80 gives sorted wave height and corresponding period
% temp78 gives raw upcrossing info.
% NT is the number of waves
%
% INPUT
% x is the time series of interest
% N is length of time series
% dt is 1/(sampling frequency)
%---------------------------------------------------------------------
function [temp80,temp78,NT]=zero_cross(x,N,dt)
% find zero-crossing points
temp70 = x(1:N-1).*x(2:N);
temp70(N) = 1.0;
temp72=find(temp70<0);
temp73 = zeros(N,1);
temp74 = zeros(N,1);
temp75 = find(x<0);
temp73(temp72) = 1;
temp74(temp75) = 1;
temp76 = and(temp73,temp74);
temp77 = find(temp76>0);
temp555 = temp77/24;
temp78 = zeros(N,4);
% total number of complete waves
NT = sum(temp76)-1;
% find local maxima, minima, duration and occurance
% Hmax Hmin Tdur Tocur
for i = 1:NT;
temp78(temp77(i),1) = max(x(temp77(i):temp77(i+1)));
temp78(temp77(i),2) = min(x(temp77(i):temp77(i+1)));
temp78(temp77(i),3) = temp77(i+1) - temp77(i);
temp78(temp77(i),4) = temp77(i);
end
%output
temp78=temp78(find(temp78(:,4)>0),:); %raw upcrossing data
temp79(:,1)=temp78(:,1)-temp78(:,2); %wave height
temp79(:,2)=temp78(:,3)*dt; %wave period
temp80=flipud(sortrows(temp79,1)); %sorted H and T

Melden Sie sich an, um zu kommentieren.

Akzeptierte Antwort

Ameer Hamza
Ameer Hamza am 21 Apr. 2020
try something like the following code. Save this in a single file named zero_cross.m
function [temp80,temp78,NT]=zero_cross(x,N,dt)
m = size(x,2); % num columns
temp80 = cell(1,m);
temp78 = cell(1,m);
NT = cell(1,m);
for i=1:m
[temp80{i}, temp78{i}, NT{i}] = zero_cross1(x(:,i), N, dt);
end
end
function [temp80,temp78,NT]=zero_cross1(x,N,dt)
% find zero-crossing points
temp70 = x(1:N-1).*x(2:N);
temp70(N) = 1.0;
temp72=find(temp70<0);
temp73 = zeros(N,1);
temp74 = zeros(N,1);
temp75 = find(x<0);
temp73(temp72) = 1;
temp74(temp75) = 1;
temp76 = and(temp73,temp74);
temp77 = find(temp76>0);
temp555 = temp77/24;
temp78 = zeros(N,4);
% total number of complete waves
NT = sum(temp76)-1;
% find local maxima, minima, duration and occurance
% Hmax Hmin Tdur Tocur
for i = 1:NT;
temp78(temp77(i),1) = max(x(temp77(i):temp77(i+1)));
temp78(temp77(i),2) = min(x(temp77(i):temp77(i+1)));
temp78(temp77(i),3) = temp77(i+1) - temp77(i);
temp78(temp77(i),4) = temp77(i);
end
%output
temp78=temp78(find(temp78(:,4)>0),:); %raw upcrossing data
temp79(:,1)=temp78(:,1)-temp78(:,2); %wave height
temp79(:,2)=temp78(:,3)*dt; %wave period
temp80=flipud(sortrows(temp79,1)); %sorted H and T
end
  2 Kommentare
Vivek
Vivek am 21 Apr. 2020
This works! Thanks
Ameer Hamza
Ameer Hamza am 21 Apr. 2020
Glad to be of help.

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (0)

Kategorien

Mehr zu Dates and Time finden Sie in Help Center und File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by