How do I plot time series for multiple machines.

Question

0 个投票

I am trying to trend and compare temperatures on multiple wind turbines. I am currently only selecting one temperature output with 10 minute values for 46 machines. I have imported the data as three columns, A is the turbine number, B is the date and time, C is the temperature value.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Follow Question

Answer 1

Peter Perkins 2017-4-1

在 MATLAB Online 中打开

1 个投票

Using timetasbles in 16b,

>> Temp = 60+ 20*randn(30,1);
>> Turbine = categorical(repelem({'A';'B';'C'},10));
>> Date = repmat(datetime(2017,4,(1:10)'),3,1);
>> tt = timetable(Date,Turbine,Temp)
tt =
30×2 timetable
      Date       Turbine     Temp 
  ___________    _______    ______
  01-Apr-2017       A       71.623
  02-Apr-2017       A       16.151
  03-Apr-2017       A       13.614
[snip]
  09-Apr-2017       C       67.999
  10-Apr-2017       C       41.401
>> tt = unstack(tt,'Temp','Turbine')
tt =
  10×3 timetable
        Date         A         B         C   
    ___________    ______    ______    ______
    01-Apr-2017    35.743    78.497    46.969
    02-Apr-2017    61.324    60.001    83.842
    03-Apr-2017    73.047    58.902    27.763
    04-Apr-2017    66.541    78.223    59.511
    05-Apr-2017    81.653    71.892    21.023
    06-Apr-2017    80.122    67.004     80.41
    07-Apr-2017    46.982    85.005    77.234
    08-Apr-2017    65.141    78.596    60.023
    09-Apr-2017    41.112    64.795    58.583
    10-Apr-2017    33.564    46.193    10.274

Then using plot (output attached)

plot(tt.Date,tt.Variables);
legend(tt.Properties.VariableNames);

or using ttplot from the File Exchange (output attached)

ttplot(tt);

Prior to 16b, you could do all that with a table (pass 'ContatntVar','Date' into unstack, and use t{:,{'A' 'B' 'C'}} instead of tt.Variables), but timetables offer advantages when you need to do more than plotting.