Menu
▾
▴
Re: [Anuga-user] Simulation of bores
|
From: Gareth D. <gar...@gm...> - 2019-05-14 09:15:54
|
***"Transmissive_n_momentum_zero_t_momentum_set_stage_boundary" would conserve momentum with a given stage boundary condition where inflow momentum is not known, is it correct?*** -- Instead of saying "conserve momentum", you could say it will "extrapolate momentum" from the interior solution. If I were troubleshooting this, I would double check whether the result is convergent (i.e. remains the same if you reduce the grid size). Then I would very carefully check the boundary condition specification in both models for any mistake (e.g. wrong sign of UH, incorrectly parsed, are you sure the data is measuring what you think, etc). And check all the other model results, for any sign of weirdness. Without detailed analysis, I would guess it's some boundary condition issue. However, if the solution is convergent, and you can repeat it with multiple models, and are completely confident that the boundaries are specified correctly, then the other question to ask is whether the shallow water equations are appropriate to this problem. Is the wave too short? Shallow water models will dissipate energy through a shock, even with no friction (e.g. the analytical dam-break solution does that). The non-hydrostatic solution will differ (?? but by this much ??-- you would have to dig into the theory to justify that). Good luck, Gareth. On 14/5/19 5:32 pm, Ananth Wuppukondur wrote: > Hi Stephen and Gareth, > > Thanks for the suggestions. > > I'm currently using file_boundary that specifies the stage and > x-momentum, y-momentum with time for the inflow boundary condition. > Attached is a sample inflow condition. In my case, i have both stage > as well as momentum as input boundary condition but still the wave > height dissipates. > > If my understanding is correct, the above > "Transmissive_n_momentum_zero_t_momentum_set_stage_boundary" would > conserve momentum with a given stage boundary condition where inflow > momentum is not known, is it correct? I tried using this as inflow > boundary but the waves still dissipate as much as before with > file_boundary. > > Best regards, > > Ananth. > > > / > / > > // > > *Ananth Wuppukondur* > > B.Tech, M.S (by Research) > > PhD Candidate > > School of Civil Engineering > > The University of Queensland > > Brisbane Qld 4072 Australia > > *M*+61 0431 453 828 > > *E*a.w...@uq... <mailto:a.w...@uq...> > > *w*//Personal webpage > <https://ananthwsharma.wixsite.com/ananthwuppukondur> Google scholar > <https://scholar.google.ca/citations?user=AROgxX4AAAAJ&hl=en> > ResearchGate <https://www.researchgate.net/profile/Ananth_Wuppukondur> > LinkedIn <https://in.linkedin.com/in/ananth-wuppukondur-a37a6048> > > CRICOS code: 00025B > > > The University of Queensland is embracing the Green Office philosophy. > Please consider the environment before printing this email. > > This email (including any attached files) is intended solely for the > addressee and may contain confidential information of The University > of Queensland. If you are not the addressee, you are notified that any > transmission, distribution, printing or photocopying of this email is > prohibited. If you have received this email in error, please delete > and notify me. Unless explicitly stated, the opinions expressed in > this email do not represent the official position of The University of > Queensland. > > > > > ------------------------------------------------------------------------ > *From:* Gareth Davies <gar...@gm...> > *Sent:* Tuesday, May 14, 2019 13:07 > *To:* anu...@li...; Ananth Wuppukondur > *Subject:* Re: [Anuga-user] Simulation of bores > Just adding to Steve's boundary-condition point here -- which I agree > with -- I have seen problematic models where the incoming stage was > specified along with zero momentum. This can cause severe reductions > in wave-height near the boundary if the true momentum is not > negligible (e.g. typical situation for an offshore tsunami boundary > condition). The boundary condition Steve mentions should not do that. > > > On 14/5/19 1:01 pm, Stephen Roberts wrote: >> >> Hi Ananth, >> >> >> I would suggest looking at the boundary condition for the Okushiri >> validation test >> >> >> validation_tests/experimental_data/okushiri, in particular the use of >> >> >> Bts = >> anuga.Transmissive_n_momentum_zero_t_momentum_set_stage_boundary(domain, >> wave_function) >> >> >> >> Also you might try the more accurate (though slower algorithm), DE1 >> instead of DE0, via a call to >> >> >> domain.set_flow_algorithm('DE1') >> >> >> How are you setting the BC at the moment? >> >> >> Cheers >> >> Steve >> >> >> ============================== >> Stephen Roberts >> Undergraduate Convenor >> Mathematical Sciences Institute >> Room 4.74 Hanna Neumann Building #145 >> The Australian National University >> Canberra, ACT 2600 AUSTRALIA >> Ph: +61 2 61254445 >> CRICOS: 00120C >> ------------------------------------------------------------------------ >> *From:* Ananth Wuppukondur <a.w...@uq...> >> <mailto:a.w...@uq...> >> *Sent:* Wednesday, 8 May 2019 7:17:46 PM >> *To:* anu...@li... >> <mailto:anu...@li...> >> *Subject:* [Anuga-user] Simulation of bores >> Hi, >> >> I'm simulating tsunami bores in ANUGA in a 9m long flume with stage >> and x momentum as input. The attached figure shows comparison of >> simulations with experimental measurements as well as another >> numerical model BASEMENT which is similar to ANUGA. Black lines are >> experimental measurements, red -ANUGA results and blue - BASEMENT >> results. If we see at the results, the simulations are underpredicted >> by upto 50% within a 8m reach in both the numerical models compared >> to experiments. The first curve is the input condition for the model >> and we can see that even the queried results at the input location >> are not simulated properly in the models to match with the given >> boundary conditions. The results are similar for different bore >> heights and velocities. >> >> Also, if we look at the duration of the bore, it is around 3sec with >> a falling limb behind peak elevation. Instead of a falling limb >> behind the peak elevation, if I extend the duration of the peak >> elevation to 2 sec followed by a falling limb (making total duration >> of bore 5sec), the dissipation in the simulation results is not much >> compared to earlier case, which suggests this is something to do with >> duration of peak elevation in the bore. >> >> Could you please suggest why there is huge dissipation of wave height >> in simulation of bores? The still water depth was 0.13m and bore >> height was around 7cm at the input. >> >> Best regards, >> >> Ananth. >> >> >> / >> / >> >> // >> >> *Ananth Wuppukondur* >> >> B.Tech, M.S (by Research) >> >> PhD Candidate >> >> School of Civil Engineering >> >> The University of Queensland >> >> Brisbane Qld 4072 Australia >> >> *M*+61 0431 453 828 >> >> *E*a.w...@uq... <mailto:a.w...@uq...> >> >> *w*//Personal webpage >> <https://ananthwsharma.wixsite.com/ananthwuppukondur> Google scholar >> <https://scholar.google.ca/citations?user=AROgxX4AAAAJ&hl=en> >> ResearchGate >> <https://www.researchgate.net/profile/Ananth_Wuppukondur> LinkedIn >> <https://in.linkedin.com/in/ananth-wuppukondur-a37a6048> >> >> CRICOS code: 00025B >> >> >> The University of Queensland is embracing the Green Office >> philosophy. Please consider the environment before printing this email. >> >> This email (including any attached files) is intended solely for the >> addressee and may contain confidential information of The University >> of Queensland. If you are not the addressee, you are notified that >> any transmission, distribution, printing or photocopying of this >> email is prohibited. If you have received this email in error, please >> delete and notify me. Unless explicitly stated, the opinions >> expressed in this email do not represent the official position of The >> University of Queensland. >> >> >> >> >> >> >> _______________________________________________ >> Anuga-user mailing list >> Anu...@li... <mailto:Anu...@li...> >> https://lists.sourceforge.net/lists/listinfo/anuga-user > > |
