#1 2016-08-20 09:59:54

rgeng

Member

Test périodique avancé -Periodic Test with Repair

Hi,
I'm trying to build a fault tree with components (repairable) subjected to periodic test .

I've discovered a strange comportament. (Bug?)   I enclose .opsa file  and some screenshoot.

Link to the files:  https://www.dropbox.com/sh/wq5s4rw6cdm9 … UNUda?dl=0

(I've checked also with XTFTA manual and it seems to me that input data for the model are correct)

Suppose to have:
-A component subjected to periodic test every 48h, with MTTR=3 h --> mu=0,33333 , lambda dangerous detected between two test equal to 2.5 e-5
-Suppose the simplest model --> final event, or port, component     (See files at link above).

Strange things discovered (I've chosed XFTA calculation):

- if mission time > 48 h (time between two tests), the calculation seems correct BUT NOT in the cases of mission time 480, 960, ecc. (Seem correct instead in case of 48, 96, ecc.)

-in particular, see enclosed files at the link above, with mission time 479..and perhaps 481.., the probability seems correct. BUT instead with mission time 480 the probability is...1 (???) - The same with 960, etc.

-in case of mission time 8760 the red diagram is quite strange (not regular, some peaks higher/lower than others).

Bye   Roberto

Emmanuel

Administrator

Re: Test périodique avancé -Periodic Test with Repair

Hi, Roberto,

this issue is known and will be fixed soon.

Don't hesitate to come back here if you have any other question or suggestion.

Regards.