Extending the concept of hourly averages to shorter time intervals
Related post: Calculations of hourly average consumption values

Background

In process industries, consumption monitoring of resources such as steam, water, electricity, and compressed air is key to ensuring energy efficiency and sustainable operations. While hourly averages provide a solid overview, shorter intervals like 30 minutes can deliver even more granular insights—especially for fast-changing processes or detailed performance analysis.

In some cases, half-hourly averages are required for regulatory reporting, for example to comply with environmental or energy-related limits imposed by authorities. Having these values available ensures transparency and compliance.

Challenge

How can we calculate average consumption values for fixed 30-minute intervals and distinguish them from rolling values? And how can we ensure the values are visualized cleanly as stepwise changes in the trend?

️ Solution

Rolling 30-minute average

This can be done easily using the Aggregation in the Tag Builder:

• Select your consumption signal (e.g., flow or energy).

• Choose the average operator.

• Set the direction to backward.

• Define a 30-minute aggregation window.

• Save the new tag.

This tag will always reflect the average of the past 30 minutes at any given time — ideal for real-time monitoring.

Fixed 30-minute average

To represent fixed 30-minute average consumption values in a stepwise format (e.g., for regulatory reporting purposes), a custom formula is used in the Tag Builder. This ensures a new value appears precisely at the half-hour mark and remains constant for the next 30 minutes.

The logic builds upon shifted versions of the TM_hour_time zone tag to identify the exact transition points between 30-minute intervals.

🧮 Formula:

if(and(a<>c, a=b), AGGREGATION_1,
  if(and(a<>d, a=c, a<>f), AGGREGATION_2,
    if(and(a<>e, a=c, a<>g), AGGREGATION_1,
      if(and(a<>b, a=c), AGGREGATION_2, sqrt(-1)))))

Variable Mapping:

• a = TM_hour_time zone
• b = a shifted by +1s
• c = a shifted by -1s
• d = a shifted by +30m 1s
• e = a shifted by +30m
• f = a shifted by -30m
• g = a shifted by -30m 1s
• AGGREGATION_1 = aggregation above
• AGGREGATION_2 = AGGREGATION_1 shifted by -29m 59s

The sqrt(-1) trick is used to enforce interpolation between steps where no fixed value should be shown. The result is a new tag that shows a constant average value for each 30-minute interval, updating exactly at the boundaries.

Value

These new 30-minute tags—both rolling and fixed—offer a powerful way to:

• Monitor short-term fluctuations.

• Detect inefficiencies faster.

• Provide more detailed insights for shift handovers or rapid diagnostics.

• Integrate into energy dashboards or set smart alerts.

Would you like to try it out?
Share your experiences or ask questions in the comments!