Menu

VMGSimDynamics - Modeling Pipe Rupture Scenarios

Modeling pipe rupture, in particular full-bore rupture, is very challenging and most programs designed for modeling these scenarios are single purpose tools. 

The pipe segment unit operation in VMGSimDynamics performs a full set of mass, momentum, energy and compositional balances, so it can be used to model these kinds of events with confidence. Heat transfer from fluid to the pipe wall and the surroundings is also calculated.

In order to validate our modeling, we set up VMGSimDynamics models for two of the cases from a set of blowdown tests performed on the Isle of Grain on behalf of the UK Health and Safety Executive (HSE). The data has been extracted from a report (1) in which the HSE compared measured data to a widely used program for this kind of work (PipeTech). 

The detailed comparison below shows that the VMG predictions match the measured data well, and, in our opinion, as well as or better than the PipeTech program. 

Mark Beyleveld, B.Sc.(Eng), B.Com.

Please contact your local VMG office for more information. 

VMGDynamics-logo.png

Pipe Blowdown Validation 

The pipe module in VMGSim has been validated against experimental data for two test cases of pipe blowdown: a full-bore rupture, and a smaller hole at one end of the pipe. The test data for the two cases, P40 and P47, are reported in a document published by the UK Health and Safety Executive (1). In the report, the actual data was compared against the predictions of the PipeTech program, which is considered state of the art for this kind of simulation. The data below shows that the quality of the VMGSim simulations is comparable to the PipeTech results.

Case P40 - Full Bore Rupture

Case P40 was a 6" schedule 40 pipe of 100m length where a full bore rupture was simulated by slicing the end of the pipe using a guillotine. The initial conditions were LPG (95% propane, 5% butane) at 21.3 bar and 17C.

Plot Trends
  • Measured - measured data from report
  • PipeTech - PipeTech data from report
  • VMG - VMG results when including fluid acceleration dP
  • VMG No Kin - VMG results ignoring fluid acceleration dP
Results

Pipe Inventory

Pipe_Blowdown_1.png

The trends for VMGSim and PipeTech have similar shape, but VMGSim tracks closer to the actual data up to 15s, where the measured data flattens and then rises again. The rise at the end of the experimental data is hard to explain, since the mass of vapor that could flow back into the pipe is minimal.

Discharge Pressure

Pipe_Blowdown_2.png

The discharge pressure is defined by the choke pressure up to the point where the choke condition goes away. After that, the pressure should be atmospheric. It is not clear how the actual data was measured, but it is hard to envisage how the outlet could have dropped to 1/2 bar abs. The "bumps" in the VMGSim base simulation are caused by pressure waves traveling up and down the pipe, which impact the vapor faction and velocity at the outlet.

Inlet Pressure

Pipe_Blowdown_3.png

The pressure at the closed end drops until the vapor pressure of the LPG is reached, and then continues to drop as the whole pipe de-pressures and the material cools. The wiggle at 20s is because momentum causes the pressure to drop below atmospheric, and then recover again.

Discharge Temperature

Pipe_Blowdown_4.png

The jumps in temperature at 21s and after are because the flow direction at the outlet reverses briefly, drawing in air at 17C. No attempt has been made to model thermocouple lag. 

Inlet Temperature

Pipe_Blowdown_5.png

Discussion

Overall, the trends match the measured data well. The blowdown takes about 20s, and the simulation correctly reflects this. There is very little difference between the profiles with acceleration dP enabled in the pipe and those without. This matches our experience that the acceleration term has significant impact only if the system is totally liquid/incompressible.

 

Case P47 - 10mm Orifice at the End of the Pipe

The initial conditions for this case are similar to the P40 case, but blowdown takes place through a 10mm orifice, which has approximately 1/230 of the full bore rupture area. The blowdown time is about 140s - i.e. about 6 times as long.

Plot Trends
  • Measured - measured data from report
  • PipeTech - PipeTech data from report
  • VMG - VMG results when including fluid acceleration dP
  • VMG No Kin - VMG results ignoring fluid acceleration dP
  • VMG LM No Kin - VMG results using Lockhart-Martinelli and ignoring fluid acceleration dP 
Results

Pipe Inventory

Pipe_Blowdown_6.png

The VMGSim results are significantly better than PipeTech. A no-slip model predicts emptying almost all of the liquid by the end of the run. By contrast, Lockhart-Martinelli matches the final inventory almost exactly, suggesting that phase slip has a significant impact on results for this case.

Discharge Pressure

Pipe_Blowdown_7.png

Again, the VMGSim results are better than the PipeTech results.

Inlet Pressure

Pipe_Blowdown_8.png

Discharge Temperature

Pipe_Blowdown_9.png

The kick in temperature at the end occurs when the discharge velocity becomes low, so that the heat transfer from the walls exceeds the convection of heat with the cold fluid.

Inlet Temperature

Pipe_Blowdown_10.png

Mark Beyleveld, B.Sc.(Eng), B.Com. 

Please contact your local VMG office for more information.

Reference

(1) An investigation into the performance of the PipeTech computer code in calculating Isle of Grain pipeline blowdown tests

www.hse.gov.uk/research/rrpdf/rr774.pdf 

All runs were made using VMGSim build 9.5.52

To Top