Vibrations in a topside pipe system on the Troll-C platform was simulated in order to assess potential damage on the inlet to a separator tank, and to check if any internal damage in non-inspectable areas was likely to have occurred.
Consequence analysis of vibrations in topside pipe system on the Troll-C Platform
On the Troll-C oil production platform, temporal vibrations were observed, probably as a result of slugs travelling through the piping systems. A simulation model in FEDEM, combined with real life measurements was used to assess potential damage on the inlet of a separator tank connected to the pipes.
Figure: Processing facility.
Figure: The FEDEM model of the piping system. Pipes and tank are modelled with 2D shell meshes. The triads show points were accelerometer measurements were taken. Lower: 6 channels were used for driving the simulation.
Accelerometer measurements had been recorded, with a total of 21 channels (directions). 6 measured channels were used to run the dynamic simulation model.
Durability was calculated by S-N approach, using strains from virtual and strain gages. Virtual strain gages can be mounted on locations not accessible with physical gages. However, measurements from physical gages are very useful for comparing and verifying the virtual simulation results.
Figure: Comparison of virtual strain gage reading and physical strain gage measurements.
Possible failiure inside the tank was not possible to assess without shut-down of processing facility. By creating a failiure in the model, it was possible to check how this would influence the gage readings on the outside.
The simulation model driven with physical measurements produced virtual strain gage readings that agreed well with physical measurements. The simulation model was used to assess how much the vibrations influenced the structural life, and to check if there was any structural damage in regions that could not be inspected.