Details:KLI-File

The *.KLI Format for Climate Data

As already noted in the introduction to the 'climate' topic, WUFI needs the following climate data for each time step:

• the rain load vertically incident on the exterior surface in [Ltr/m²h]. For the determination of this rain load, the inclination and orientation of the surface must be taken into account.
  
• the solar radiation vertically incident on the exterior surface in [W/m²]. For the determination of the amount of radiation, the inclination and orientation of the surface must be taken into account.
  
• the temperature of the exterior air [°C]
  
• the relative humidity of the exterior air [-] (0..1)
  
• the temperature of the interior air [°C]
  
• the relative humidity of the interior air [-] (0..1)
  
• the barometric pressure [hPa]. Since the barometric pressure has only a minor

effect on the calculation, specification of a mean value may be sufficient.
 

If direct measurements are not available, the rain load can be determined from the normal rain and the wind velocity and direction. The radiation load can be determined from the global (or direct) and diffuse radiation incident on a horizontal surface.
When measured weather data are used, which do not directly contain the required rain and radiation data, but rather the normal rain and the solar radiation incident on a horizontal surface, WUFI can do this conversion automatically during the calculation, so you only need to specify the appropriate climate file for use in the calculation and the orientation and inclination of the building component.

On the other hand, you may wish to provide WUFI directly with the required data, for example because you have measured them or because you want to employ different conversion algorithms than those used by WUFI. In this case, write these data to a WUFI climate file *.KLI and specify this file for use in the calculation.

This is an example for the first few lines of a WUFI climate file:

The climate file essentially consists of a table, with each line containing all required climate data (rain, radiation, exterior temperature, exterior relative humidity, interior temperature, interior relative humidity) for a certain interval of time.

The format of the data is free. The only requirement is that a PASCAL program (like WUFI) must be able to read the numbers. All data are read as real numbers. The entries in each line are separated by blanks.

The three header lines are further discussed below.
 

The time structure of the climate data is defined by a relative hour count (given by the first entry in each line). This count is relative to a starting date (given in the first header line).
Imagine the climate file has been created by a fictitious data logger which has started the measurement at the starting date. At the end of each measuring interval (the intervals need not be equidistant) it has recorded the mean values over that interval and has added the total duration of the measurement so far as a time stamp.

The starting date of a calculation must be in the time interval covered by the climate file (i.e., it must be equal to or later than the starting date of the climate file and earlier than its ending date).
The calculation may go beyond the end of the climate file (i.e., the ending date of the calculation may be later than the ending date of the climate file); the climate file is then read again from the beginning (as often as necessary).

The format of the starting date is:
 
  day.month.year_hour.fraction,    e.g. 1.1.2000_14.5.

Fractions of hours are decimal fractions: 1.1.2001_0.5 means half an hour past midnight. The format allows two decimal places for the fraction.

The choice of the time step for the climate file should be suitably coordinated with the choice of the time step for the calculation. The simplest choice is to use equidistant time intervals in the climate file and to set the time step for the calculation to the same size.
If you want to use a different calculation time step, you should choose it so that it divides the climate time intervals exactly. Otherwise, WUFI adjusts each time step so that it does divide the current (*) climate interval exactly.
 
(* since the climate file need not have equidistant time intervals.)

Examples for time step widths:

	climate data  interval		time step  entered by user		time step  used by WUFI			because
a)	1	h	1	h		1	h	1 · 1 = 1
b)	40	h	6	h		5.714	h	7 · 5.714 = 40
c)	42	h	6	h		6	h	7 · 6 = 42
d)	44	h	6	h		6.286	h	7 · 6.286 = 44
d)	6	h	44	h		6	h	1 · 6 = 6

 

Examples for climate files

Example 1:

Start: 3.4.1994, 0.0

These climate data mean that the measurement was started on April 3, 1994, at midnight, the mean exterior temperature was 5.9°C between 0:00 and 1:00, 5.2°C between 1:00 and 2:00, and 11.1°C in the 22-hour-interval between 2:00 and 24:00.
 

Example 2:

Start: 1.1.1991, 0.0

This climate file contains monthly mean values. The first line gives the mean values for the interval from January 1 1991 0:00 to January 31 1991 24:00, etc.
 

Example 3:

Start: 31.1.1997, 10.93

This one-line climate file might be used to simulate an imbibition experiment in the laboratory.
Starting on January 1 1997 10h56m (=10.93 decimal), the laboratory climate stays constant for 500 hours. Since WUFI can read a climate file repeatedly, the time could also be entered as 1 hour. The same line would then be read over and over again. No time step larger than 1 hour would then be possible, however (see above; any different time step would be adjusted down to 1 h).

(The imbibing surface is treated here as the 'exterior' surface, the imbibed water is offered as 'rain' in a quantity that is sure to surpass the quantity of water imbibed per hour. Also see the relevant discussion in the Questions & Answers section.)
 

Example 4:

Start: 23.1.1998, 10.50

This climate file might be used to simulate an imbibition experiment with subsequent drying in the laboratory.
The experiment started on January 23, 1998 at 10:30. The specimen was imbibing for 1215 hours and was then left to dry for (3410 - 1215 =) 2195 hours.
 

How to create your own KLI files

If you want to use your own weather data for a WUFI calculation, you can simply convert them to either the WET or the TRY or the DAT format and use them directly in this form for the calculation.

If your rain and radiation data have already been determined for the surface in question, a directional conversion is not necessary, and you can write these data to a KLI file and use them in this form for the calculation.
The KLI format also comes handy if you want to use other conversion methods, if you want to use measured interior climate data etc.

In such a case, you just need to write an ASCII file which conforms to WUFI's

• .KLI format: three header lines first, then one line of climate data for each

time step.
For this, you may use any editor program that allows you to write the data in a pure ASCII format, for example a text editor such as the Windows notepad, or a self-written program which computes or converts the data in the desired manner and then writes them to a file.
If you are using Excel to create a KLI file, save the data using one of the ASCII formats

• .TXT or *.PRN, since WUFI cannot read the proprietary Excel format *.XLS; then rename

the file to the extension *.KLI.

Please note that the second header line of the KLI format has to contain data at certain positions so that WUFI can read them (it is much longer than can be displayed in most ASCII editors, which is not immediately obvious because of the large blank spaces between the data). The simplest way to obtain such a line is to copy to your own file one from a KLI file supplied with WUFI. The information contained in that line is then not correct for your file, but this does not affect the calculation (it is information about the parameter settings used for generating the file and is not used during the calculation).

The tool KLI_Create.exe in the Tools folder may be useful when you want to create a *.KLI file with monthly average values for temperature and humidity, no rain and no radiation.