Subsea developments in seismically active areas present several engineering challenges. One such area, which is frequently underestimated, is the design of jumpers to resist earthquake loading.

Jumpers connect quite unrelated pieces of subsea equipment such as manifolds on deep foundations, pipeline end terminations (PLETs) on mudmats and wellheads supported on slender well casings. If these connected structures have a different dynamic response, the jumpers will receive additional loads from the out of phase structure movements. This seismic coupling is particularly an issue for HP/HT fields, where stiff heavy jumpers are combined with large manifolds.

Italy-based RINA Consulting aims to design these items in a safe and conservative fashion. “Typically, the seismic design of a structure is performed in one of several ways,” said Omar Zanoli, project lead at RINA Consulting. “One is very simple, where you look at each structure and connection individually; you look at the manifold and you see the seismic response of the manifold, then you look at the jumper, and then the PLET. Each structure is studied independently.

He described the other approach as more complex.

“Here, the idea is to study the entire subsea system in a numerical model, and then look at the response of all the structures in a single design phase,” Zanoli said. “This is certainly a more optimum approach, but it is very time consuming.”

It’s also costly and technically complicated, he added.

Finding Middle Ground

The plan is for RINA to develop a methodology that fits between the two extreme design approaches.

“We are developing a simplified way to see the connection and the relative movement between the different components of the system,” Zanoli added. “The basic idea is to look at the individual components and their natural frequency; this is the way a structure is moving under a cyclic loading. Then you carry out a similar assessment on the jumper, and you discover the natural frequency of the jumper and see if they have different natural frequencies.”

Zanoli explained that there might be cases where the natural frequencies are very close. For example, the manifold is one second, and the jumper is 0.95 second. “In that case, if you compute these stresses from separate models you don’t get the right answer, because there is a coupling between the jumper and the manifold and they exchange loads, one to the other, and the stresses can increase,” he said. “If you’re not able to take into account this coupling between the structures, the design will be wrong.

Zanoli suggested looking at the single response of the individual components first. If they have similar fundamental frequencies, the simplified approach cannot be used. Instead, switch to a complex system model that takes time and cost. Differentiation between cases that require fully coupled treatment and those in which a more straightforward separation and analysis of the individual subsystems is important to producing an efficient design within a reasonable time frame.

“We believe there is an intermediate way that could work and that this is the best solution to design or verify situations for certain conditions that can model, in a very simplified way, the manifold, the jumper and the PLET and run a simplified analysis of this system,” Zanoli continued. “The idea is not to perform real-time history analysis with real earthquakes, but to perform a response spectrum analysis. That is a simplified way to see the response of structures.”

This way the response of the structure and the stresses within the structure can be seen in a realistic way, but it is simplified compared to the full system model. First, perform the analysis of fundamental frequencies, and then use a response spectrum analysis of a portion of the system.

Even though there is a drive for standardization within the oil and gas industry, this is not a solution to the dangers of seismic damage to systems as there is the geographic factor to consider.

“If you are in Egypt, you will have certain seismic input; if you are in the Gulf of Mexico or Chile it’s completely different,” Zanoli said. “You need to adapt your designs to the local conditions and conduct the system analysis every time. I cannot envisage a situation where a supplier can say that the manifold is fully safe for seismic conditions in every situation. The only way to achieve that would be to over engineer the manifold and that will make it prohibitively expensive.”

RINA has already utilized this approach successfully in a project with BP in Egypt that resulted with a safe design of subsea structure for the company.

“We are talking to other operators and will continue to develop this technology,” Zanoli said. “With the technology continuing to evolve at a rapid rate, it may be that increased computer power and analytics will mean that in the future the full system analysis may become less time consuming and not prohibitively expensive. But even if we reach that situation we strongly believe that simple sanity checks will be required for these huge models.”