Wind load to "BS EN 1991-1-4" 3

[mte12]

In calculating wind loads to BS EN 1991-1-4, there is an aspect which I a trying to reconcile with AS 1170.2 and AS 1170.2.

In AS 1170.2 there's an Annual probability of exceedance from Table 3.3. For a Design working life of 50 years with an Importance level of 2, the API is 1/500.
AS 1170.2 states: Annual probability of exceedance is the inverse of the so-called ‘return period’ better described as the average recurrence interval.

BS EN 1991-1-4 refers to annual probability of exceedence of 0.02, which is equivalent to a mean return period of 50 years.
Refers to BS EN 1990 4.1.2 (7)P which mentions "However in some cases the character of the action and/or the selected design situation makes another fractile and/or return period more appropriate".

Firstly need to check I'm comparing apples with apples.

Then I'm wondering why a return period of more than 50 years (say 250 years or 500 years) is not explicitly mentioned, rather than mention in 4.1.2 (7)P. Or does it it come out in the wash, as BS EN seems to use 10 minute wind speeds with an exposure factor.

The other thing I noticed was a consequence factor (kf) from BS EN 1990.

See attachment for reference.

 

[Trenno]

This thread might be of use perhaps.

EC & AS codes have a fundamentally different approach to ultimate wind loading. Going off an offhand comment by a Wind Engineer years ago when working in the UK, using actual ultimate/service wind velocities to calculate pressures is a much more robust/accurate way of design. Simply factoring up the service wind load by a factor of 1.5 (as per EC) can be unconversative in some situations.

 

[HTURKAK]

I do not have any idea regarding the standard AS 1170.2 and the similarities with EC EN 1991-1-4.
Eurocode uses characteristic wind velocity values having annual probabilities of exceedence of 0,02, which is equivalent ( reciprocal of 0,02) to a mean return period of 50 years.

Binom distribution is used for the probability that a wind velocity exceeds (r) times in (n ) years; ( for example , characteristic n=50 year return period , annual risk p=0.02 and No exceedence in 50 years P=(1-0.02)^50=0.364 and AT LEAST ONE EXCEEDENCE P=1-0.364=0.636 .That is , the probability that 50 year return period wind velocity at least one time exceeded in 50 years is 63.6% )

Please look BS EN 1990 Annex ( informative ) B3 Reliability differentiation for consequences classes ,

I will suggest you to search the forum for (Eurocode wind loads ) since this subject discussed several times and one of them ,EN 1991-1-4. Probability Factor (Cprob), 1 Year Return Period ,thread507-507085






Use it up, wear it out;
Make it do, or do without.

NEW ENGLAND MAXIM

 

[mte12]

Thanks Trenno and HTURKAK.

I did see the sections for consequence classes. The equivalent in AS is an Importance level, for a 50 year design life, Annual Probability of exceedance is 1/500 and 1/1000 for an Importance Level of 2 and 3 respectively. Comparing the effect on the probability factor (c.prob) for 1/500 and 1/1000 events is less than 1.1, similar to the consequence factor in British Standard.

Trenno, I did see your post a few days ago, and will spend some more time looking into it.

 

[BJI]

It is a bit misleading regarding the terminology since the 50 year return period in the Eurocode is the design life, as opposed to the average recurrence interval, which is the time between exceedance of the wind speeds given, e.g. a 1 in 500 year storm event.

Regarding the gust wind speeds, I would recommend the following publication. There is a comparison to other international codes near the end.

https://www.acsev.org.au/images/document/47_1.pdf

 

[mte12]

Thanks BJI, will look into publication.

It is confusing since BS EN 1991-1-4 states "annual probability of exceedence of 0.02, which is equivalent to a mean return period of 50 years".
Interestingly, someone else did have the opinion that the return period was analogous to the design life.

 

[BJI]

You are correct about the design life. It is essentially the same, 1/L = 0.02 (for 50 years) multiplied by the lifetime risk, r, to give the annual probability of exceedance. Refer to AS 1170.0 Table F2, Note 2, for the risk values based on Importance Level.

 

[mte12]

Thanks BJI.

Looked at Table F2 in AS 1170.0, annual probability of exceedance is r/L where r is the risk of exceedance equal to 0.1 for importance level of 2, so annual probability of exceedance for design life of 50 years is 1/500.

Perhaps I didn't fully understand British/European to begin with, but my inference now is that the risk of being exceeded of 0.02 or 1/50, for a 50 year design life, must be because it's a 10 minute mean wind velocity.