As part the design of a pipeline, a requirement for the assessment of the effects of a possible seismic hazard on the integrity of the pipeline and associated above ground installations (AGIs) was identified. One such installation was a pig trap site consisting of twin pig traps and associated valves and pipework.
In order to quantify the risk associated with the threat of a seismic hazard it is necessary to quantify the likelihood of occurrence of the hazard and likely effects on integrity (and hence safety) given that the hazard has occurred. The likelihood of occurrence of a seismic hazard, at any given location, depends on the proximity of that location to the presence of ground faults, the likelihood of a seismic event occurring at those faults and ability of seismic waves to propagate from the source of the hazard to the location under consideration.The effect of the seismic disturbance on the integrity of a structure depends on the interactive nature of the structure and ground disturbance and the ability of the structure to withstand the loads imposed due to this interaction.
AFAA were contracted to perform the assessment of the ability of the pipework to withstand the loads due to wave propagation caused by seismic activity. The objective of the analysis was to identify components at risk of failure and modifications that may be required to seismically qualify the design for return periods ranging from 100 to 10,000 years.
The focus of the study was on the determination of stress levels that occur due to the presence of seismic waves arriving at the location of the structure.
In these situations the duration of activity is typically no more than a few minutes (often the most significant activity lasts only a few seconds) and the loads vary considerably over these time frames. This means that the stresses can only be determined in the strictest sense, by performing a dynamic analysis that takes account of the inertial effects and rapidly changing loading conditions. Such analyses involve considerable effort.
In view of this, more approximate techniques were used to give a conservative estimate of the stress levels without recourse to a full dynamic analysis. To this end two approaches were adopted. The first of these was a static assessment in which the base of the model was fixed in space and the peak ground accelerations were applied as a body force to the entire model.
The second was a response spectrum analysis which involved two steps. The first of these is the determination of the modal frequencies and associated mode shapes. The second involves the determination of extent of the deformation associated with each mode shape by scaling the mode shapes in accordance with factors contained within uniform hazard spectra (UHS) that had been derived based on a consideration of possible locations and magnitudes of the seismic event.
It was demonstrated using a combination of the two types of analytical techniques (static and response spectrum), that the stresses and displacements were acceptable and fell within the elastic limit criteria. Pressure vessel and support loads for the response spectrum analysis for the 10,000 year return period were sent to their respective manufacturers, who assessed them and found them to be acceptable.