This blog post explores the development of a React <DateTimeChart /> component to visualise numeric, boolean and string time series data on a single chart.
Managing and converting units is challenging, particularly as the variety of units grows. This post introduces SI base units, explains unit dimensions, and demonstrates a formula for conversions using scales and offsets. By leveraging resources like Project Haystack's unit database, it can be relatively straightforward to build a flexible and accurate unit converter.
The choice of interval impacts calculations in time series data, particularly for operations like division or multiplication. Using examples of energy consumption and production data, this post demonstrates how different intervals can yield varying averages for the same dataset.
Folding aggregates multiple data points into a single value for a given interval using functions like sum, average or maximum. Interpolation fills gaps in data using techniques such as linear, change-of-value (COV) or apportioning.
There are two key components of time series data: timestamps and values. Timestamps include the date, time and timezone, with intervals indicating the duration between readings, while values can take various forms such as numeric, boolean, string or period.
