How to Simply Import CSV and Excel Time Series
Time series in the form of a CSV or Excel file can be imported into the ETA via Simple time series import.
The following figure shows the button for simple time series Import in the Project ribbon of the ETA.
When selecting the time series to be imported, please refer to the notes on the file format. Once you have selected the time series, the following dialog box for importing time series appears.
The following dialog box contains important information about the time series import (see figure below).
Equidistant and Nonequidistant Time Series
The imported date and time entries displayed in the Data Processing table are replaced by equidistant time steps in the size of the specified Time increment starting with the Start date and Start time to be named if you Specify start time point and time step in the import dialog box.
In the window that opens, the name and Unit of the time series can be changed. Save the imported time series with a click on OK within the dialog.
If you want to import further time series, go back to Load file and execute the import dialog again. Click Transfer–> to go to the last section of the import dialog.
The imported time series is then displayed in the Project Explorer (top left window in the ETA) in the Temporary time series folder (see the figure opposite).
Fix Data Errors
When importing real measurement data, difficulties can arise if measurement values are not recorded correctly or not at all, if data points do not have the correct time stamp or if the duration of the measurements does not match. The following sections contain useful tips on how to correct data errors with the ETA.
Date Not Recognized
If the date and time are in an unusual format, they cannot be automatically captured by the ETA. In the first figure of the following slider, the date and time are contiguous in column A of the Data Processing table.
After the import, the preview of the Imported time series (at the right edge of the time series import window) shows that the time values have got mixed up (see second figure of the slider above).
This problem can be solved using the table operations. In the example, column A can be divided into two columns. Use the button 
Separate into two columns (see first figure of the following slider).
The date and time information is separated into two columns (see second figure of the slider above). Once the columns have been assigned correctly (see orange frame in second figure), these time series can be imported without any problems.
Error Messages about Incorrect Times
During the import process, various error messages can occur, the cause of which can be, for example, incorrectly recorded data points or a discontinuous time sequence. The error message informs the user about the type of error (see the first and second figure of the following slider). In the preview of the Imported time series, the error sources can be localized using the debugging buttons 
Navigate next invalid line (see the third figure of the following slider). You can view the relevant points in time, evaluate the cause of the error and finally agree to skip these data points. The following slider illustrates this using the example of individual data points with incorrect time stamps.
NaN
Another cause of errors in measured values is recording, e.g., when the measuring instruments fail. A corresponding error message (see the following two figures) indicates the automatic correction of the data. In this case, it is worth taking a closer look at the corrected values using the time series preview. By clicking on the debugging button 
(tooltip: Navigate next auto-corrected line) the view jumps to the corresponding position. The third figure of the following slider shows the detection of automatically corrected data points. This shows that the raw data contain measuring points at which no data value was recorded. In the ETA, NaN values are temporarily inserted for this purpose. NaN stands for Not a Number.
If you end the import with the Transfer–> button, the automatically corrected data (NaN) are referred in a dialog box before closing the import dialog (see first figure of the following slider).
To import the NaN values, click on Yes. The time series in the ETA then retains the NaN values. They are displayed in the plot as gray bars.
If you select No, the corresponding time points are deleted. This results in a time series that is nonequidistant.
In a final dialog box, decide whether the same operation should be performed for the automatically corrected data for all time series (see second figure below).
The following slider contains the two dialog boxes which appear before the import is finished.
Automated Time Series Import
To import time series in the same format more often (e.g., for daily operation optimization), the import process can be automated. The automated import with the Time Series Interface component is suitable for this. See also the article Import and Export of Time Series.
