Selected process safety systems analysed using dynamic simulations: A study of the Flare and the High Integrity Pressure Protection Systems
Translated title
Selected process safety systems analysed using dynamic simulations
Author
Outomuro Somozas, Ana Xiao
Term
4. term
Education
Publication year
2020
Submitted on
2020-06-04
Pages
110
Abstract
Industrielle anlæg bruger fakkelsystemer til sikkert at afbrænde gasser under nødsituationer og HIPPS (High Integrity Pressure Protection System) til at forhindre farligt overtryk. Specialet undersøger, om tidsafhængige (dynamiske) simuleringer i Aspen HYSYS Dynamics V9 kan forbedre dimensionering og vurdering af sådanne sikkerhedssystemer sammenlignet med traditionelle stationære (steady state) beregninger. For fakkelsystemer bygger praksis ofte på steady state-dimensionering, som er bevidst konservativ, men kan føre til upraktisk store systemer. API Standard 521 anbefaler dynamisk modellering for bedre at afspejle de skiftende forhold under nødtrykaflastning. Her er nødtrykaflastning af fakkelnetværkene for tre anlæg - forskellige i samlet størrelse, designfakkelrate, rørdimensioner og gasmængde (hold-up) - modelleret og sammenlignet med steady state-design. Resultaterne viser, at jo større fakkelsystem, desto større er det skjulte potentiale for at fjerne flaskehalse, målt som en større forskel mellem steady state-designraten og den dynamiske top-faklerate. For HIPPS analyseres den tilgængelige procesreaktionstid til ventillukning, samt hvordan modellens robusthed påvirker resultaternes følsomhed. Anvendelse af Tulsa Unified Model som rørstrømningskorrelation gav mere konservative forudsigelser.
Industrial plants use flare systems to safely burn off gases during emergencies and High Integrity Pressure Protection Systems (HIPPS) to prevent dangerous overpressure. This thesis examines whether time-dependent (dynamic) simulations in Aspen HYSYS Dynamics V9 can improve the design of these safety systems compared with conventional steady-state (time-invariant) calculations. For flare systems, industry practice often relies on steady-state design, which is deliberately conservative but can make systems impractically large. API Standard 521 recommends dynamic modeling to better reflect how conditions change during emergency depressurization. Here, emergency depressurization of the flare networks at three facilities - differing in overall size, design flare rate, pipe dimensions, and total gas inventory - was modeled and compared with steady-state designs. The results show that the larger the flare system, the greater the hidden potential for debottlenecking, seen as a larger gap between the steady-state design rate and the dynamic peak flare rate. For HIPPS, the study analyzes the available process response time for valve closure and explores how model robustness affects the sensitivity of those results. Using the Tulsa Unified Model as the pipe-flow correlation produced more conservative predictions.
[This abstract was generated with the help of AI]
Keywords
Documents