General Information on the Time Series Methods
Access to the time series methods is provided by the menu bar and the context menus of the time series in the Project Explorer and in the Plot window (see following slider).
The functions that repeat in the various time series methods are presented below.
Selecting Time Series
To select the time series to be processed with the selected method, click on the Select time series button (see following figure).
A window opens for selection from the open time series (see following figure).
Check the checkboxes. The nodes that are higher in the hierarchy are automatically selected.
Confirm your selection with OK.
Some methods only allow the selection of one time series. The following slider shows what this looks like.
Whether a multiple selection is possible (e.g., Transfer to another year) or only one time series can be selected varies from method to method.
Create or Overwrite a New Time Series
For example, while in the Merge method the time series cannot be overwritten, in most time series methods it is necessary to choose whether the time series generated by the method are used as a replacement for the original time series or as new additional time series (e.g., Transfer to another year). A check mark can be used to specify that the existing time series are not to be overwritten, but that the new time series are to be created additionally. For some methods the check mark is set by default (e.g., Scale), for others (e.g., Convert units) it is not.
A Suffix can be defined for the names of the new time series to be created (see following figure).
Calculation Methods
The following methods are available for calculating missing values in time series:
Examples for the application of these methods are described in the articles Synchronizing Points in Time and Making Equally Spaced (preserving integral or interpolating).
Typical Periods/Typical Days
Typical periods/typical days are calculated sample periods/days with a typical load curve.
Typical periods/days are used to synthesize and compress time series.
Typical periods are generated as clusters: first the time series data are sorted by periods (e.g., days). Similar periods (e.g., days) are then combined to form a group.
The number of periods (days) combined in a group determines the proportion of the typical period (typical day) in the total period under consideration (for example, one year).
The following figure shows the capacity distribution within a year (12 months), which was divided into four typical days. The wide red lines show the boundaries between the typical days.
The extension of the typical day on the timeline (the unit mon stands for month with 730 hours each) depends on how often this typical day is represented throughout the year. The more frequently a typical day is represented in the year, the wider the bars of this typical day are and the larger the time portion of the typical day in the entire year.
The resolution of a typical day results from the number of time steps per typical day. In the figure above, each typical day has a resolution of 24 time steps, each of which is represented by a light blue bar.
For your own compressed time series you can specify the length (in hours) and the number of time points of a typical period in the simulator options menu.
Time series compression with typical days is used, for example, in Structural Optimization with Storages.
Holiday Calendar
Certain methods, such as synthesize time series by typical days, transfer to another year, and filter time intervals, take into account regional holidays. For Germany, United Kingdom, and Australia, public holidays are stored in the ETA. In the dialog windows of the methods that use holidays, the Country, the Region, and partly the City are to be selected. The selection Other stands for the countries for which no public holiday calendar is stored in the ETA.
The time zone is independent of the public holiday calendar.
