Comparing date/time values in Matlab/ calling out date time values to use in code

Hi,

So I have to sets of data one represents wind speed, and the other represents radar rain data.

My wind data is imported from excel and consists of a column of dates and the the remaining colums are values of wind speeds for different stations at these date times.

The radar data is a hdf file that has been converetd into rain intensity in another function.

Each set of data is in time increments of 5 minutes beginning September 2020.

E.g. first 4 sets of data :

• '01-Sep-2020 00:00:00'
• '01-Sep-2020 00:05:00'
• '01-Sep-2020 00:10:00'
• '01-Sep-2020 00:15:00'

I have looking to add a line that will compare the date/time values of the wind data to the date/time values of the radar data for each time point.

When the code finds a match, it can then be used in two other functinos I have.

A suggested start is

for i = 1:length(WindDatetime)
if RadarDatetime == WindDatetime(i)
then the file of interest == i 
else 
i =i +1;
end 
end 

However, this is not working for me.

I will insert the last part of my code that turns the date time vectors into a table of numbers like so :

2020 9 1 0 0 0

2020 9 1 0 5 0

2020 9 1 0 10 0

windtime = windfive.date;   % extracting the date column from excel sheet 
 
idx = ismember(windtime,timestampDT);  % show which data is missing 
windtime(~idx);                         
 
Winddata = datevec(windtime);            %turn wind time data into a vector
Radardata = datevec(timestampDT);         % turn radar time data into a vector 
Radardata(any(isnan(Radardata),2),:) =[];  % delete any 'NaN' values

Another very useful thing to me would be if anyone knows a simple way of calling out time data so that I can use it.

For example if I would like to view the data between

'08-Sep-2020 23:55:00' and '17-Sep-2020 17:45:00'

is there a way of doing this?

Thanks:)

Answer 1

[dpb]

Convert to a timetable and all kinds of tools are available; see <Overview Timetables> for all of the above kinds of operations

[Steven Lord]

In particular, using a timerange to index into the timetable rows sounds like what you want.

[dpb]

+1 Steven! I couldn't bring the function name to mind at the time...I recommend you don't get old; memory goes away!

[drb17135]

Hi, thank you for your help.

I have managed to turn one of the tables into a timerange.

However, I am having trouble converting the other. The one I am having trouble with is in the form of a 8640x1 datetime.

When I try convert this into a time range I am given an error.

Do you know how I would go about converting this?

Thanks

[Steven Lord]

You don't turn a table or timetable into a timerange. You build a timerange from a pair of datetime or duration scalars. See the first example in the help text for timerange:

rng default % Make sure I get the same data every time I run this Answer
tt = array2timetable(randn(10,3),'RowTimes',datetime(2016,4,randi(15,10,1)))

tt = 10x3 timetable
       Time          Var1        Var2         Var3  
    ___________    ________    _________    ________

    11-Apr-2016     0.53767      -1.3499      0.6715
    01-Apr-2016      1.8339       3.0349     -1.2075
    05-Apr-2016     -2.2588       0.7254     0.71724
    01-Apr-2016     0.86217    -0.063055      1.6302
    02-Apr-2016     0.31877      0.71474     0.48889
    13-Apr-2016     -1.3077     -0.20497      1.0347
    11-Apr-2016    -0.43359     -0.12414     0.72689
    05-Apr-2016     0.34262       1.4897    -0.30344
    15-Apr-2016      3.5784        1.409     0.29387
    01-Apr-2016      2.7694       1.4172    -0.78728

tr = timerange(datetime(2016,4,3),datetime(2016,4,12),'closed')

tr = 
	timetable timerange subscript:

		Select timetable rows with times in the closed interval:
		[03-Apr-2016 00:00:00, 12-Apr-2016 00:00:00]

	See Select Timetable Data by Row Time and Variable Type.

tt4to12 = tt(tr,:)

tt4to12 = 4x3 timetable
       Time          Var1        Var2        Var3  
    ___________    ________    ________    ________

    11-Apr-2016     0.53767     -1.3499      0.6715
    05-Apr-2016     -2.2588      0.7254     0.71724
    11-Apr-2016    -0.43359    -0.12414     0.72689
    05-Apr-2016     0.34262      1.4897    -0.30344

In this code April 11th is between April 3rd and April 12th inclusive, so tt4to12 includes the first row of tt. April 1st however is not in that range, so tt4to12 does not include that row of tt.

If instead you wanted to ask "which rows of tt have Time in a specific subset" you'd want to use ismember.

rowsWith5Or11 = ismember(tt.Time, datetime(2016, 4, [5 11]))

rowsWith5Or11 = 10x1 logical array
   1
   0
   1
   0
   0
   0
   1
   1
   0
   0

tt5Or11 = tt(rowsWith5Or11, :)

tt5Or11 = 4x3 timetable
       Time          Var1        Var2        Var3  
    ___________    ________    ________    ________

    11-Apr-2016     0.53767     -1.3499      0.6715
    05-Apr-2016     -2.2588      0.7254     0.71724
    11-Apr-2016    -0.43359    -0.12414     0.72689
    05-Apr-2016     0.34262      1.4897    -0.30344