What Is a Time Series?
A time series is a sequence of data values, each associated with a specific point in time. As time advances, new values are recorded to build up a continuous history that can be analyzed, visualized, and forecasted. The chart below shows a simple example: temperature readings being recorded every 3 hours:
Each dot represents a single measurement value paired with a timestamp. The collection of all these measurements over time forms the time series. In PowerTSM the structure (interval, unit, function) is defined once, and the system handles storage and retrieval of every recorded value.
Definition of a PowerTSM Time Series
Section titled “Definition of a PowerTSM Time Series”A time series object in PowerTSM is defined by a set of core properties that control its identity, structure, and behavior. The diagram below shows how these properties relate to each other.
ID and Name
Section titled “ID and Name”Uniquely identifies the time series within PowerTSM. Both the ID and name must be distinct across the system.
Function and Cyclicality
Section titled “Function and Cyclicality”Specifies whether the time series is left-aligned or right-aligned and cyclic or acyclic.
See Function and Cyclicality for details.
Interval
Section titled “Interval”Defines the resolution at which data is stored. Always choose the smallest interval necessary to avoid unnecessary storage overhead.
See Intervals for details.
The interval is composed of two sub-properties:
| Sub-property | Description | Example |
|---|---|---|
IntervalType |
The fundamental unit of time | minute, hour, day |
IntervalMultiplier |
A scaling factor applied to the type | 15 |
Used for unit conversions and interval transformations across time series.
See Units and Conversions for details.
Description
Section titled “Description”Free-text field for additional context. Useful for organization and search within the system.
Formula (Optional)
Section titled “Formula (Optional)”A calculation formula that is evaluated at runtime. Leave empty if the time series stores raw data rather than derived values.
See Formulas Overview for details.