I want to plot Linear Equations but dont know how to do it on matlab

I am fairly new to Matlab and I need to derive 2 linear equations for a) wind speed is between 13km/h - 50km/h; b) wind speed is between 50km/h - 100km/h.

The Data is already collected from 12 different spreadsheets.

source_directory = 'C:\Users\thoma\OneDrive\Desktop\Project 1';
source_file = dir(fullfile(source_directory, '*.csv'));
file_name = fullfile(source_directory,source_file(i).name);
file_data = xlsread(file_name);
    
Wind_Month = file_data(:, 7);
Wind_Year = [Wind_Year; Wind_Month];


Wind_Year = [];
Power1 = [];
Power2 = [];
Day1 = [];
Day2 = [];


for i = 1: length(Wind_Year)
    
  if  13 - 50
    
      Power = m * x + c;
      
      Power1 =[Power1; Power];
      Day1 = [Day1; i];
      
      
  elseif 50 - 100
      
      Power = m * x + c;
      
      Power2 = [Power2; Power];
      Day2 = [Day2; i];
   
  end

end

How do I calculate Y (Power) in the linear equations?

Also, how do I get the 13 -50 and 50 - 100 to something like: >=13 <50 and >=50 <100?

Answer 1

[dpb]

Since you're still new, let me make some suggestions on coding --

First, use newer features -- if you'll read the doc for xlsread (unless you're using a very old version) you'll see it has been deprecated in favor of alternatives -- for this purpose I'd suggest using a MATLAB table instead of array and so readtable...then it's also easy to do the grouping in the same table...

source_directory = 'C:\Users\thoma\OneDrive\Desktop\Project 1';  % this is good to put into variable so can change
% Q? here -- Why not read the .xlsx workbooks themselves instead of making a separate .csv file???
d=dir(fullfile(source_directory, '*.xlsx'));                     % use a generic variable for a returned dir() struct; it's not a file, exactly
% Now process the files sequentially...
for i=1:numel(d)
  file_name = fullfile(source_directory,d(i).name); % get file in turn from dir struct
  tData=readtab;e(file_name);                       % and pull the data into a table
  % Now I'm going to presume the table has headers -- if not, you can create variable names to match  
  %Wind_Month = file_data(:, 7);
  %Wind_Year = [Wind_Year; Wind_Month];    % doesn't seem to make sense to catenate these two vertically???
  % Make a new grouping variable based on desired ranges of windspeed
  edges=[0, 13, 15, inf];                  % set bin edges
  tData.WindGrp=Y = discretize(tData.WindSpeed,edges, 'categorical', ...
    {'Calm', '13-50', '50-100', 'Gale'});  % create table variable by category
  Power=rowfun(@(x,y) polyfit(x,y,1),tData,'GroupingVariables','WindGrp', ...
      'OutputVariableNames','PowerPoly1','InputVariables',{'WindSpeed','Power'});
end

Above will give you a fit for each group, not just the two; that's fixable by either only defining the two groups or just using the group categories explicitly. You'll also need to save the result as cell array inside the loop; I just overwrite the output variable each pass above.

I made up a small sample dataset here to test --it produced the following output --

>> rowfun(@(x,y) polyfit(x,y,1),tX,'GroupingVariables','G', ...
    'OutputVariableNames','PowerPoly1','InputVariables',{'X','P'})
ans =
  3×3 table
      G       GroupCount         PowerPoly1     
    ______    __________    ____________________
    Calm           2         0.77278      16.191
    13-50         10          1.7144     -3.1833
    50-100         8        -0.86311      125.91
>> 

from the following table--

>> tX
tX =
  20×3 table
    X       G         P   
    __    ______    ______
    11    Calm      24.692
    50    50-100    95.668
    71    50-100    88.859
    15    13-50     40.323
    50    50-100    49.942
    72    50-100     59.38
    91    50-100    52.602
    34    13-50     28.341
    45    13-50     100.77
    98    50-100    22.288
    78    50-100    89.286
    27    13-50     4.3433
    84    50-100    36.556
    33    13-50     76.195
    31    13-50     65.677
     1    Calm      16.964
    42    13-50     96.901
    49    13-50     56.287
    34    13-50     26.501
    32    13-50     59.137
>> 

the above was created from

X=randi(110,20,1);

followed with the discretization and then P was a multiplier plus a noise component on the X value...

Salt to suit...