If you’ve worked with building energy or any other IOT, then you’ve probably seen how important yet difficult tagging all the data is. A commercial office building could be streaming over 10,000 data points at 15 minute or shorter intervals. Which one is a temperature? Which one is a set point? Which one corresponds to all the different equipment in building? Getting them tagged correctly is the first step to doing anything else with the building. Getting them tagged incorrectly could lead to serious problems.
In the past, tagging data has required a lot of time and frustration. That’s why data tagging and acquisition could account to 50% or more of a building energy commissioning project’s total cost. Recently, there’s been a lot of promise with using cloud computing and machine learning to automate data tagging. But what’s really required to streamline data tagging, whether it’s done with people, rules, or machine learning?
Based on our recent work with opentaps for a major municipal government’s buildings portfolio, I wrote an article about this, and it was accepted and published in the Fall 2019 issue of Haystack Connect, Project Haystack’s magazine:
Whether you work with building energy, IOT, or machine learning, I hope this will give you some ideas and food for thought. As always, please let me know if you have any thoughts or suggestions.