Merge two tables with different time stamps

KNIME Analytics Platform
1

Hi,

I just started working with knime and encountered the following problem for which I haven't found a solution in the forum. 

I have two tables with different time stamps, like this:

Time1 Temperature
0:15 25
0:30 27
0:45 27
1:00 26
Time2 Energy
0:10 40
0:20 45
0:30 44
0:40 42
0:50 41
1:00 43

How can I merge them?

Thanks a lot!

2

Hi,

what is your expected outcome?? I would guess you need the concatenate node?

3

Yes, excactly. I want to concatenate/merge my two tables on the time nodes.

Right now, the temperature is measured every 15 minutes and the energy every 10 minutes such that if I did an inner join I would only extract the time stamps 

0:00, 0:30. 1:00

but I would like to have at least four time stamps per hour in the merged table. Is there a way to do this? 

4

Okay, if you want to remain all TimePoints, use the Joiner with a Full Outer Join. In the column selection panel of the Joiner do not check the "Filter... columns" Checkboxes.

 

Using the Column Merger you can now combine the values from column Time1 and Time2. Now sort on this column. To resolve the missing values I would suggest the Time Series Missing Value node, which can interpolate based on the ordering of the series.


This is my final outcome:

Time1                                             Temperature                              Energy

0:10 25.0 40.0
0:15 25.0 42.5
0:20 26.0 45.0
0:30 27.0 44.0
0:40 27.0 42.0
0:45 27.0 41.5
0:50 26.5 41.0
1:00 26.0 43.0
5

Thanks for the fast response.

The problem is, which I haven't pointed out before, that the Energy is the consumption over 15 minutes, if it's 45 kwH over 15 minutes, I would have to modify the value if the time stamp would be only 10 minutes. Therefore, simple interpolation might not be useful.

Instead, I was wondering if a different approach is possible: 

To start with the one table 

time temperature
0 25
0:10 26
0:20 26
0:30 25
0:40 27
0:50 26
1:00 26

and quickly transform it to 

time temperature
0 25
0:15 26
0:30 25
0:45 26.5
1:00 26

Could I do that easily with Knime? 

I'm sorry, I haven't worked with Knime before which is why my question might seem really stupid.

6

It is not so easy but it does work. I attach you a workflow which is a modification of what I wrote above. But does some filtering afterwards.

7

Thank you, this is exactly what I was looking for!