Quantitative Study on How Model Simplification Affects Pipeline Network Surge Results

This paper presents the methodology and results of a systematic study on how the surge analysis results are affected by different levels of model detailing. By doing this, the basis for a guideline and a strategy for surge analysis requirements, at different phases of a sustainable design, can be developed in the future. A complicated pipeline network with multiple pumps distributing cooling water to multiple sub-units was analyzed. Process design was based on steady state hydraulics in which the sub-units were simplified as “points”. This simplification method does not alter the steady state hydraulics of the main pipeline network loop. However, the detailed piping connections inside the sub-units have an influence in the pressure wave propagation going in and out of the units. Different internal piping configuration details of the sub-units are considered to collect quantitative understanding on the effects they have on the surge. The levels of maximum surge pressures influenced by the pipe length, details in branches, and pipe elevations inside the sub-units are evaluated. Power failure and emergency scenarios were considered to collect data on the sensitivity model details have on location, timing, magnitude of surge pressure, and the occurrence of column separation and collapse during surge events. Special attention was given to the conditions at which the transient operations inside the sub-units create surge outside the unit, and to conditions in which outside transient operations develop un-desirable surge inside the sub-units. The results show the importance of integrating transient hydraulics analysis at an early stage. It is necessary to think about operations in design. The magnitude and sensitivity level of surge influence by various types of model details are summarized. A guideline for surge analysis strategy, considering appropriate levels of model details in an economical way, can be elaborated based on the results of this study.