Loads: bin_statistics test fails due to inconsistent handling of empty bins (NaN vs 0)

[simmsa]

In the test_bin_statistics test (currently disabled) there is a failure caused by comparisons of 0 to NaN. The underlying JSON data has NaN where bins are empty and the bin_statistics function returns zeros:

MHKiT-MATLAB/mhkit/tests/Loads_TestLoads.m

Lines 5 to 29 in 31388bb

	function test_bin_statistics(testCase)
	assumeFail(testCase, "Temporarily Disabled - Need to ask @hivanov about nan vs zero comparison")
	% create array containg wind speeds to use as bin edges
	bin_edges = 3:1:25;
	relative_file_name = '../../examples/data/loads/loads_data_dict.json';
	full_file_name = fullfile(fileparts(mfilename('fullpath')), relative_file_name);
	fid = fopen(full_file_name); % Opening the file
	raw = fread(fid,inf); % Reading the contents
	str = char(raw'); % Transformation
	fclose(fid); % Closing the file
	data = jsondecode(str); % Using the jsondecode function to parse JSON from string
	% Apply function to calculate means
	load_means_struct = [data.means.uWind_80m];
	load_means_table = struct2table(data.means);
	ta = struct2table([data.bin_means]);
	st = table2struct(ta,'ToScalar',true);
	ta_std = struct2table([data.bin_means_std]);
	st_std = table2struct(ta_std,'ToScalar',true);
	bin_against = load_means_struct;% load_means_table.uWind_80m;
	datast = bin_statistics(load_means_table,bin_against,bin_edges);
	assertEqual(testCase,st,datast.averages,'RelTol',0.001)
	assertEqual(testCase,st_std,datast.std,'RelTol',0.001)
	end

@hivanov-nrel do you have any suggestions on how to fix? Should empty bins return NaN or 0. Do you think this is a bug in the test or in the bin_statistics function?

[simmsa]

@hivanov-nrel #188 was merged into develop. Would you consider this issue resolved?