1. Averaged vs Raw Data Analysis

This document shows the shortcomings of using averaged data instead of the raw data for operational analysis. Averaged data is typically fine for things like production variables, where these are used for reports on production, material and energy balances, etc. However, averaged data should not be used for monitoring the performance of a process/equipment, condition-based maintenance, event monitoring, environmental exceedance, etc.

The analysis below was conducted using 3 datasets from a customer’s process:

• Tank level
• Tank pressure
• Chill water temperature

The below trends show the raw data and 1-minute average data for a period of 4 days and 14 hours. The 1-minute average data was further split into 3 datasets:

• 1-minute time-weighted average stored at the end of the averaging period
• 1-minute time-weighted average stored at the start of the averaging period
• 1-minute event-weighted average stored at the start of the averaging period

All three 1-minute average data give similar results, so the rest of the analysis in this document focusses on the 1-minute time-weighted average stored at the start of the averaging period dataset.

Zooming into the Chiller Temperature trend to a 15-minute window, you can see the 4 different datasets. The averages generally follow the raw data reasonably well when the raw data is steady, but start significantly varying when the process starts varying.

Below are trends for the 3 process variables over time ranges where the averaged data significantly deviates from the raw data.

Below is an example for Tank Pressure, where the %error is in excess of 30%!

The next step is to perform the calculation of the %error, (averaged value from raw value), for each process variable and then assessing how many times this % error is larger than 1% during the 4-day 14-hour period. Below is a table of the %errors for the 3 process variables, sorted by largest to lowest.

The total number of %errors of 1% or more, over the 4-day 14-hour period, is shown in the table below. Note, the highest error for temperature was 3.2% because the temperature varied more slowly than the other two parameters. On the other hand, for pressure, the highest error was over 115%.

Conclusion:

This analysis clearly illustrates the total inadequacy of using averaged data to monitor process operations. Averaged data can have large error and can easily miss significant process deviations that should be flagged and Event Frames created to mark these deviations and take action on them.

If you would like more information about our services, please fill out the form below and a representative will be in touch with you shortly.