Percentage error of HRR

Posted 7 months ago by fireowen

Post a topic
Answered
f
fireowen

What is a reasonable way to show the percentage difference between the FDS calculated HRR value and a t^2 design fire value? Or is this a reasonable value to show at all? I am using it as a sanity check to show that the HRR generated by FDS is reasonable when compared to the design fire. The graph shows the FDS and design fire HRR values and the % error between them (noting that multiplied the error % by -1 so it is easier to see on the graph). 


My concern is the large error % at small time steps. While this is approx 75% error the absolute difference is 0.03kW. In comparison at around 600 seconds there is a time stamp with 8% error and an absolute difference of 1200kW. Does this mean anything on am I grasping at straws?


Interested to hear anyones thoughts.

0 Votes

Bryan Klein

Bryan Klein posted 7 months ago Admin Best Answer

A design fire is a time-averaged approximation of some dynamic, transient, and chaotic phenomena.  I would not expect a 1:1 correlation between a simulated or a real fire and a design fire.

I would think that if you did some kind of rolling time average of the HRR output from FDS, you would begin to match the design fire curve more closely.

0 Votes


1 Comments

Bryan Klein

Bryan Klein posted 7 months ago Admin Answer

A design fire is a time-averaged approximation of some dynamic, transient, and chaotic phenomena.  I would not expect a 1:1 correlation between a simulated or a real fire and a design fire.

I would think that if you did some kind of rolling time average of the HRR output from FDS, you would begin to match the design fire curve more closely.

0 Votes

Login or Sign up to post a comment