Assign width of Bin Edges within rainflow matrix c = rainflow(x)

Is it possible to specify the width of Bin Edges when calculating the rainflow cycle counts matrix c = rainflow(x) ?

After computing the cycle counts for the data, using the command: [c,hist,edges,rmm,idx] = rainflow(Y) , I plot a histogram of cycle counts as a function of stress range, using the command: histogram('BinEdges',edges','BinCounts',sum(hist,2)) , as indicated in the example of the Matlab Documentation

It seems that the 'Bin Edges' are assigned an arbitrary width value. Instead, I would like to assign a specific width value to the Bin Edges, while ploting the histogram of the cycle counts, resulting from the cycle counts matrix 'c = rainflow(x)'.

How can we specify the width of Bin Edges when calculating the rainflow cycle counts matrix 'c = rainflow(x)' ?

Thank you very much in advance for your help

[Ash Deval]

I want to know the same thing. I'm stuck on the same issue.

Answer 1

[Steven Lord]

As written, the BinWidth property of the histogram object is determined by the BinEdges that you specified when you called histogram.

x = randn(1, 1e5);
E = [-3 -2 -1.25 -0.5 0 0.5 1.25 2 3];
h = histogram(x, 'BinEdges', E);
h.BinWidth
oldEdges = h.BinEdges

To change the BinWidth before the fact, change your BinEdges. To change the BinWidth after the fact, set it. This will affect the BinEdges property.

h.BinWidth = 1;
newEdges = h.BinEdges

[breathi]

Steven, can you show us an explicit example of doing this with the results given by the rainflow function? I'm a bit stuck with this and this seems utterly complex and I may be misunderstanding something:

    [c, rm, rmr, rmm, idx] = rainflow(randn(1,1000), linspace(0,5,1000));
    cfr = arrayfun(@(x,y) repmat(x,1,y),c(:,2),c(:,1).*2,'UniformOutput',false);
    cfr = [cfr{:}];
    cfm = arrayfun(@(x,y) repmat(x,1,y),c(:,3),c(:,1).*2,'UniformOutput',false);
    cfm = [cfm{:}];

    % plot amplitude (range/2) over mean
    xEdges = floor(min(cfr)):1:ceil(max(cfr./2));
    yEdges = floor(min(cfm)):1:ceil(max(cfm));
    figure, hTest = axes; p = histogram2(cfr./2, cfm, xEdges, yEdges, 'FaceColor','flat');
    p.BinCounts = p.BinCounts./2; % divide cycle values by 2 (as half cycles have been multiplied by 2 previously)
    
    hTest.ZScale = 'log';
    cb = colorbar;
    caxis([1 max(cb.Limits)]);
    colormap jet;
    set(hTest, 'Colorscale', 'log');

Thanks.

[Steven Lord]

I've never used the rainflow function and from reviewing its documentation I'm not sure it's possible to control the bin widths in that function. My background is not in signal processing.

You might want to contact Technical Support and ask the Support staff for such an example (and if this is a common task for signal processing, ask Support to enter into the enhancement database a request for that example to be added to the rainflow function documentation.)

[Max Rose]

I changed defined specific bin width by defining edges, like Steven recommended, within the 'localhistcounts2' function. This function is a subfunction of the 'rainflow' function (lines 290 ff.).

function [rm,xedges,yedges] = localhistcounts2(xc,xm)
    % Compute two-dimensional histogram.
    isSingle = isa(xc,'single');
    n_edges=101;
    binWidth=0.01;
    for i=1:n_edges
        edgesx(1,i)=(i-1)*binWidth;
    end
    [~,xedges,binx] = histcounts(xc,edgesx); % specific edges for range
    [~,yedges,biny] = histcounts(xm,1); % only 1 bin for average

Maybe this works fine for you as well.

[Nandar Hlaing]

@Max Rose Thank you very much. It is really helpful.

I just have an additional curosity. If I don't want to have the same binwidth for all bins, it is still okay if I directly define edgesx as I prefer right?

[Max Rose]

I didn't try by myself but actually thats what is done in:

    n_edges=101;
    binWidth=0.01;
    for i=1:n_edges
        edgesx(1,i)=(i-1)*binWidth;
    end

There the edges are defined, so I think it should work. So give it a try and let us know.