Resist temperature profile improvements

Just discovered beestat a few weeks ago and LOVE the additional insight into my home’s thermal performance!

One feature I find particularly useful are the temperature profiles, however I would like to see a small tweak to provide even better data. I live in an area where humidity is not a huge factor for comfort, so I utilize opening the windows whenever there is a favorable temperature difference between outside and inside. This applies year round (cooling the house in summer, warming in winter).

Given that, I feel the resist temperature profile would be greatly enhanced by the ability to split and have a user-adjustable “gap” for near-normal outdoor temperatures. Since I typically have windows open when the outside is cool enough in the summer or warm enough in the winter, the resist data is likely skewed/not accurate for me in that range.

Potentially a different way to approach this would be to ignore/not calculate resist profile values when the system is set to “off”. When I have the windows open, I will manually turn the system off to prevent any equipment from running. This could even be a user checkbox setting to provide more options for those that may not want this feature.

I still care about resist data when actively heating or cooling, but those tend to only be in the ranges away from “normal”-ish outdoor temperatures.

1 Like

I couldn’t be 100% sure without really looking into the data, but when beestat generates a profile (resist, heat, or cool), it also does some basic statistic analysis to remove outlier points. When that’s all done it fits a linear trendline to the final data.

So if you have a range of temperatures where the windows are open that creates significant outlier data points, beestat will throw those away. Any that are “close enough” and not thrown away will tend to get averaged out and then cleaned up a little more with the linear trendline.

Just wanted to follow up here now that a few more months of data are in.

It appears there’s not much of an issue with my resist temperature profile even with windows open at various times. Based on the (large) instantaneous changes I can sometimes see in the indoor temperature, there must be some outlier rejection in the algorithm as my resist profile is VERY linear.

Now I just need some additional stage 1/2 heating runtime to get those slopes more accurate. Thanks @ziebelje for the additional info on this, and for Beestat in general!

1 Like