Wave actions - dynamic analysis 1

[kostast88]

Hi everyone,

I design a 2-storey steel building on a jetty head which is 100m into the sea. The water depth to seabed is shallow and waves are small (Mean SWL 4m+3m storm surge, Significant wave 2.5m). The jetty is a set of unbraced steel hollow sections (Diameter 457x16) @5m centres.

I want some opinions on whether I need to design the steel building elements & connections (steel frames) for wave actions. So far I assumed the substructure and superstructure are separate, and performed static analyses for each individually.

I followed BS 6349 which uses the Morrison equation to determine the wave actions. This generates a maximum force 21 kN. It also gives me the wave profile with time.

Should I perform a Time-history analysis? My experience from earthquake design tells me that a spring mass system subject to cyclic velocities/accelerations will generate forces on the first storey due to the motion. However these forces are invisible in static analysis. So should I do time history and design beam/column connections for the wave? Is this common practice on offshore structures?

I looked many books and papers but couldn't find any reference. So I would appreciate some feedback.

Thank you

 

[1503-44]

Invisible in static analysis?

 

[kostast88]

I mean the wave force is applied on the piles. The deck level is above the wave load.

On a static analysis (single pile) this is a cantilever with a load applied around mid-height. Moments above deck are 0.

On a dynamic analysis, the wave induced displacements generate inertia forces on the masses due to the push-pull effect of the waves.

I have attached an image for clarity:

 

[1503-44]

Unfortunately your diagram raised more questions.

Static load deflection of above deck structure is not correct, unless the pins are actually in the columns. You have made the pins to look like they are at the ends of the beams, hence the columns would be continuous at floor levels, more like as you have shown in your dynamic load diagram.

In your dynamic diagram the beams are experiencing reverse bending. You have only drawn the top beam and you have shown it as a flat translation, which can only be if beam has pinned ends, although you seem to have shown moment resisting connections to the columns.

Could you please be more clear about how beams and columns are connected and why you think they change between static and dynamic cases.

 

[kostast88]

Thank you for the response, this is a quick sketch. The Deck beams are pinned. The rest of the beams are fixed against lateral loads (Type II/ Wind frame). The bottom "columns" are the piles. So they are not continuous. On top of them we have pinned columns of the building (steel frame).

I don't understand what this has to do with the time history analysis though. The question is whether you consider the dynamic case or not, regardless of the structural system formulation. Have you worked on a similar structure design or platform where this had to be taken into account?

 

[1503-44]

Perhaps better you ask your question in the Structural Engineering General Discussion forum.

 

[phamENG]

Is this wave action the result of, say, a 700 year storm? Or is it going to see it a couple times a year? If this is an extreme event load, I'd go with static (including a close evaluation of second order effects and stability analysis, of course) and be conservative. If it's going to see it frequently then a more nuanced and accurate assessment of the dynamic loading would be justified.

 

[1503-44]

Maybe wrong, but I'd think such small waves would be a relatively common occurrence.
Close coastal waves normally have a period of between 6 to 8 seconds. Would you not use static equivalent loading for such load application time periods?

 

[kostast88]

This is the significant wave therefore it should the 1 in 10 years. This is expected more than a couple of times per year. The maximum wave height which is 1 in 100 years is almost x1.5.

You are right 1503-44 this will be a common occurrence during an average storm, with a 5.5s period.

My only conservatism is the storm surge water level (+3m). I understand that most offshore structures would be designed with statics but there would be conservative assumptions. In this case we can't afford it. Unfortunately we have no budget for being generally conservative, this is a small project. So I try to get my assumptions as accurate as possible, that's why I think about the moments on first floor generated by the waves.

This is a Wind Moment Frame building (ref: Link), so the connections are designed only for the wind load moments (not gravity loads, everything is pinned for vertical loads, but fixed for lateral loads). So my question is should connections/members be designed for wind+wave loads.

 

[phamENG]

Yes, they absolutely should be designed for both moments. Here in the states we refer to those as "partially restrained moment connections." It's neat in theory, but a pain in practice - though we have alternative procedures to simplify it. Since your frame is going to try to rack with the wave action, the rigidity that does exist in that connection will "attract" some proportion of the moment in the frame. If that moment is too great and your moment connection fails, you'll be in trouble next time the wind blows.

Sorry for the confusion on returns - when I hear "jetty" I think of breakwaters at harbor entrances, not of piers, so I was a bit confused on the water reaching the structure.

I don't do offshore structures, so to me these wave heights are rather large in terms of engineering experience - I live and practice in a region with a large but very sheltered harbor. (I spent some time in the Navy, so I do understand how big waves can get.) Given the frequency with which you expect these waves to hit, I would tend toward a dynamic analysis.

 

[1503-44]

2.5m waves aren't hardly worrying at all. The storm surge currents are likely to be a far more severe loading.

BTW, I have never done, nor would ever do, nor do I know of any offshore structure designed as a "wind frame", hence my referral to the structural forum.

 

[kostast88]

Interesting.

I have seen that the current force is much bigger than the wave force (combined wave and current force = 1.4 x current force) however I am using the sea current velocity available from published metocean data.That's about 3m/s.

Are you saying the storm current could be bigger than that? I am slightly worried that the above value is not conservative as generally it isn't a big number. The site is somewhat sheltered but winds are very strong so I imagine that wind wave currents would be more prevalent than tidal currents.

 

[1503-44]

Sea currents in the data books are usually given for buoys that are quite far from the coasts. Nearshore water is shallower and there may be obstructions and focal points, which can increase local waves, swells and water current velocities. You can't use those for nearshore design works. You should do some kind of a localized hydrodynamics study. A 3m surge is also not much of an extreme event as those things go. Gulf of Mexico hurricanes tides are reaching 5-7m at what seems to be an almost yearly basis. Maybe you do have a very protected location and are fortunate not to be forced to deal with those.

We had numerous focused wave shoaling and swell current effects a couple of years ago that ran up our usual 2-3m typical storm waves to 5-8m in some areas that damaged 2nd and 3rd story balconies. And not due to hurricane. These waves were generated by the tail of a fast moving cold front in the Atlantic a thousand km farther to the north.
Skip to 3m:00s

https://m.youtube.com/watch?v=wDU5BuL361I