[seastar]

最近在文献中看到TLP会发生ring与spring

请问二者有什么区别？
谢谢

[shaopizi]

引用框（seastar ＠ 2009-09-20）

最近在文献中看到TLP会发生ring与spring

请问二者有什么区别？
谢谢

springing is due to the second order sum-frequency effect, it is a steady-state phenomenon, while ringing is a transient effect which may probably be due to the third order and even higher order wave effect. Making analogy to ship hydrodynamic, ringing is similar to the so-called whipping problem.

[zzz]

springing of TLP is a combination of hydrodynamic and motion dynamic problems, while ringing is purely a nonlinear wave diffraction problem.

引用框（shaopizi ＠ 2009-09-21）

springing is due to the second order sum-frequency effect, it is a steady-state phenomenon, while ringing is a transient effect which may probably be due to the third order and even higher order wave effect. Making analogy to ship hydrodynamic, ringing is similar to the so-called whipping problem.

[shaopizi]

引用框（zzz ＠ 2009-09-21）

springing of TLP is a combination of hydrodynamic and motion dynamic problems, while ringing is purely a nonlinear wave diffraction problem.

I was talking about the exication sources of these two problem.

Springing and ringing are of importance, because the structure response is of high frequency(wi+wj+...).

springing and ringing are all defined with respect to the structure response. Nonlinear springing is nothing but a second order problem , because the second order theory has been well established, so it is straight forward to include the coupling effect of motion dynamic. That is why in the literature, when people do springing analysis, couping effect of structure responses are also included.

However, the third-order theory is still not ready to include the coupling effect of structure response. Thus people have to study from the excitation point of view, that is the nonlinear wave diffraction as you said. In the literature, Malenica and Molin,1995 and Faltinsen, Newman and Vinje,1995 have studied the third order diffraction problem. Some others have aslo solved the third problem by numerical approach. The reason people don't study the ringing problem including hyrodynamic and motion dynamics is very simple: too difficult!

Prof. Kim C.H. in TAMU has been working on this topic for many years. He has a very nice book published in 2008, Nonlinear waves and offshore structures, where detailed description could be found...

[seastar]

谢谢

那位前辈有相关的资料参考一下

[zzz]

引用框（shaopizi ＠ 2009-09-21）

I was talking about the exication sources of these two problem.

Springing and ringing are of importance, because the structure response is of high frequency(wi+wj+...).

springing and ringing are all defined with respect to the structure response. Nonlinear springing is nothing but a second order problem , because the second order theory has been well established, so it is straight forward to include the coupling effect of motion dynamic. That is why in the literature, when people do springing analysis, couping effect of structure responses are also included.

However, the third-order theory is still not ready to include the coupling effect of structure response. Thus people have to study from the excitation point of view, that is the nonlinear wave diffraction as you said. In the literature, Malenica and Molin,1995 and Faltinsen, Newman and Vinje,1995 have studied the third order diffraction problem. Some others have aslo solved the third problem by numerical approach. The reason people don't study the ringing problem including hyrodynamic and motion dynamics is very simple: too difficult!

Prof. Kim C.H. in TAMU has been working on this topic for many years. He has a very nice book published in 2008, Nonlinear waves and offshore structures, where detailed description could be found...

1. springing is well studied. But it's not based on the "coupling" effects.

2. Could you explain ring is a resonance from motion dynamics? Or can you provide some materials to verify that?

[shaopizi]

引用框（zzz ＠ 2009-09-22）

1. springing is well studied. But it's not based on the "coupling" effects.

2. Could you explain ring is a resonance from motion dynamics? Or can you provide some materials to verify that?

1. that is corrorect. springing is a resonent condition, where wi+wj equals to the natural frquency of vertical motion of TLP. In this case, the excitation and damping may be very important. the literature i enjoyed most was the phd thesis by prof. M.H.Kim, and also some studies by C.H.KIM.

2. i didn't say ringing is a resonant prolem. :>. According to my understanding, it is more like a transient effect. a strong transient may involke all the frequencies for structure dynmamic. So ringing may be more relavent to extreme response. However, if ringing happens very often, and the damping of is small, the response doen't have time to decay, it becomes difficult to separate the rining and sprining effect. This has been observed in our experiments.
Theoretically speaking, one should analyse the rininging effect just like as we do in springing analysis(do it up to third order). But the theory is still not mature when you want to include the couping effect of structure response. So one has to do some assumptions when analysing the response from ringing. The most usful one would be that, we calculate nonlinear diffraction as exciation, and consider the system as a linear system.

I have many materials on spriging and rining(also whipping for ships). I will upload some of them soon.

[shaopizi]

here comes some papers

附加文件

•  Matsuielal1993_secondordersumfrequencyoscillationsoftlp_predictionandmeasurement.pdf (774.25K)
  下载次数: 72
•  ChenXBetal_numericalevaluationofspringingloadsontensionlegplatforms.pdf (1.14MB)
  下载次数: 45
•  3rdorderwaveloadsusingahigherorderpanelmethod(Zhu,1996).pdf (261.59K)
  下载次数: 33
•  Anexpermentalinvestigationofhigher-harmonicwaveforcesonaverticalcylinder(Grue,2000).pdf (915.9K)
  下载次数: 44
•  Grue1994_Nonlinearloads_ringing.pdf (259.51K)
  下载次数: 33
•  TimedomainsimulationofRingingofHeidrunTLP(Kim,C.H,2002).pdf (307.2K)
  下载次数: 61
•  Ringingloadsonaslenderverticalcylinderofgeneralcrosssection(Faltinsen,1999).pdf (140K)
  下载次数: 38

[seastar]

谢谢

[zzz]

引用框（shaopizi ＠ 2009-09-22）

1. that is corrorect. springing is a resonent condition, where wi+wj equals to the natural frquency of vertical motion of TLP. In this case, the excitation and damping may be very important. the literature i enjoyed most was the phd thesis by prof. M.H.Kim, and also some studies by C.H.KIM.

2. i didn't say ringing is a resonant prolem. :>. According to my understanding, it is more like a transient effect. a strong transient may involke all the frequencies for structure dynmamic. So ringing may be more relavent to extreme response. However, if ringing happens very often, and the damping of is small, the response doen't have time to decay, it becomes difficult to separate the rining and sprining effect. This has been observed in our experiments.
Theoretically speaking, one should analyse the rininging effect just like as we do in springing analysis(do it up to third order). But the theory is still not mature when you want to include the couping effect of structure response. So one has to do some assumptions when analysing the response from ringing. The most usful one would be that, we calculate nonlinear diffraction as exciation, and consider the system as a linear system.

I have many materials on spriging and rining(also whipping for ships). I will upload some of them soon.

Thank you very much for the references!


2. I am curious to know why "people don't study the ringing problem including hyrodynamic and motion dynamics". What's the real difficulty?

[shaopizi]

引用框（zzz ＠ 2009-09-25）

Thank you very much for the references!


2. I am curious to know why "people don't study the ringing problem including hyrodynamic and motion dynamics". What's the real difficulty?

Last year, I have been involved in developing a numerical code to solve third-order diffraction problem.
1. Frequency domain solution:
If it is only a diffraction problem, the difficulty is related to the low covergence rate of a free surface integral in the formulation. This difficulty has been investigated by many people and different treatments have been proposed.
However, when the body is freely floating, we have third-order derivatives on the body(through the body boundary condition). That is similar to but more difficult than the socalled mj-terms problem.
On the other hand, the computational cost increase dramatically in third-order problem.
2. Time domain solution:
It doesn't have the problem of low convergence of free surface integral as we have in frequecy domain solution. The difficulty related to higher derivatives in body boundary condition is still there. Also the secularity occurs in 3rd order free surface condition and the resulting solution. In 2d, Molin has proposed a method to kill the secularity, together with my supersivor I also proposed a two time scale method to treat this. However , both of them do not apply in 3D. The 3d 3rd order solution in time domain is very time consuming. I use my 8-cores workstation to run simulations.
3. Actually, after recognize the difficulties in solving the 3rd order problem, people have turned to solve a fully nonlinear problem.

[zzz]

引用框（shaopizi ＠ 2009-09-25）

Last year, I have been involved in developing a numerical code to solve third-order diffraction problem.
1. Frequency domain solution:
If it is only a diffraction problem, the difficulty is related to the low covergence rate of a free surface integral in the formulation. This difficulty has been investigated by many people and different treatments have been proposed.
However, when the body is freely floating, we have third-order derivatives on the body(through the body boundary condition). That is similar to but more difficult than the socalled mj-terms problem.
On the other hand, the computational cost increase dramatically in third-order problem.
2. Time domain solution:
It doesn't have the problem of low convergence of free surface integral as we have in frequecy domain solution. The difficulty related to higher derivatives in body boundary condition is still there. Also the secularity occurs in 3rd order free surface condition and the resulting solution. In 2d, Molin has proposed a method to kill the secularity, together with my supersivor I also proposed a two time scale method to treat this. However , both of them do not apply in 3D. The 3d 3rd order solution in time domain is very time consuming. I use my 8-cores workstation to run simulations.
3. Actually, after recognize the difficulties in solving the 3rd order problem, people have turned to solve a fully nonlinear problem.

1. Frequency dmain: Do you use flat panel ot high-order panels?, the compuating cost should be less than in time domain since you may just invert matrix "one time"?

3. Fully nonlinear, you must meet wave breaking, how can you deal with that?

--------------------------
All the above seem hydrodynamic issue, without motion response?

[shaopizi]

引用框（zzz ＠ 2009-09-25）

1. Frequency dmain: Do you use flat panel ot high-order panels?, the compuating cost should be less than in time domain since you may just invert matrix "one time"?

3. Fully nonlinear, you must meet wave breaking, how can you deal with that?

--------------------------
All the above seem hydrodynamic issue, without motion response?

1. i use a cubic higer-order bem. You are right. Frequency domain computation takes much less time than time domain.
3. I strongly agree that breaking wave is a big problem. we try to simulate cases without wave breaking, that is the wave condition is not too strongly nonlinear. i don't have much experience on dealing with breaking waves. Theoretically speaking, potential flow theory will not be valid when the overturning wave hits the free surface. The 'impact' of overturning wave on the free surface introduces a strong local turbulent flow. There is no rational way to introduce a new initial condition(due to this 'impact') to the potential flow model.
In the case of wave breaking, Prof. Beck in U. Mich. has introduced a local damping effect to surpress the breaking. i don't know how good it is to do so compared with what is happening in reality.

4? yes, all those are without coupling effect of motion response. Because i work on hydrodynamic, so i have focused more on that aspect. For third-order simulation, i guess nobody has been able to include coupling effect of motion response. In the fully nonlinear case, some studies including the structure response have been reported, for instance, by prof. R. Beck in U. Mich., Dr. kring in U.S., Dr. Tanizawa in Japan...

[zzz]

引用框（shaopizi ＠ 2009-09-26）

1. i use a cubic higer-order bem. You are right. Frequency domain computation takes much less time than time domain.
3. I strongly agree that breaking wave is a big problem. we try to simulate cases without wave breaking, that is the wave condition is not too strongly nonlinear. i don't have much experience on dealing with breaking waves. Theoretically speaking, potential flow theory will not be valid when the overturning wave hits the free surface. The 'impact' of overturning wave on the free surface introduces a strong local turbulent flow. There is no rational way to introduce a new initial condition(due to this 'impact') to the potential flow model.
In the case of wave breaking, Prof. Beck in U. Mich. has introduced a local damping effect to surpress the breaking. i don't know how good it is to do so compared with what is happening in reality.

4? yes, all those are without coupling effect of motion response. Because i work on hydrodynamic, so i have focused more on that aspect. For third-order simulation, i guess nobody has been able to include coupling effect of motion response. In the fully nonlinear case, some studies including the structure response have been reported, for instance, by prof. R. Beck in U. Mich., Dr. kring in U.S., Dr. Tanizawa in Japan...

I agree with you on the hydrodynamic aspects.

hehe, but you still have not explain why "nobody has been able to include coupling effect of motion response".

What's the difficulty?

[shaopizi]

i think i have answered that. the difficulty is the how to accurately calculation of high order derivatives on free surface and body surface.

The hydrodynamic formulation of problem considering body motion is much more complicated than that for a pure diffraction problem. One can try to extend the theory of Malenica and Molin,1995 to include the body motion, and then i think one would kill himself.

[windgo]

Hi,shaopizi, do you have Dr.M.H.Kim's thesis? Would you share it with me? I am kinda very interested.

引用框（shaopizi ＠ 2009-09-22）

1. that is corrorect. springing is a resonent condition, where wi+wj equals to the natural frquency of vertical motion of TLP. In this case, the excitation and damping may be very important. the literature i enjoyed most was the phd thesis by prof. M.H.Kim, and also some studies by C.H.KIM.

2. i didn't say ringing is a resonant prolem. :>. According to my understanding, it is more like a transient effect. a strong transient may involke all the frequencies for structure dynmamic. So ringing may be more relavent to extreme response. However, if ringing happens very often, and the damping of is small, the response doen't have time to decay, it becomes difficult to separate the rining and sprining effect. This has been observed in our experiments.
Theoretically speaking, one should analyse the rininging effect just like as we do in springing analysis(do it up to third order). But the theory is still not mature when you want to include the couping effect of structure response. So one has to do some assumptions when analysing the response from ringing. The most usful one would be that, we calculate nonlinear diffraction as exciation, and consider the system as a linear system.

I have many materials on spriging and rining(also whipping for ships). I will upload some of them soon.

[shaopizi]

hi, windgo, i will try to send it to you by email, may i have your email address?

[windgo]

引用框（shaopizi ＠ 2009-09-27）

hi, windgo, i will try to send it to you by email, may i have your email address?

pls check you inbox, thank you very much!

[zzz]

引用框（shaopizi ＠ 2009-09-22）

here comes some papers

Shaopizi,

Do you have copy of vinje's two papers in 1980? If you have, can you send to my email address?

Thanks a lot!

[shaopizi]

引用框（zzz ＠ 2009-09-30）

Shaopizi,

Do you have copy of vinje's two papers in 1980? If you have, can you send to my email address?

Thanks a lot!

vinje is an applied mathmatician. he worked not only on hydrodynamics but also on stochastic theories. do you mean the paper dealing with numerical modeling of breaking waves ?