Risk-based simplified fire models and methods

Fire safety design of industry facilities that may experience large scale fires feed by accidental releases of combustible material is challenging. On one hand, current technology allows for effective control of fire risk by implementation of mitigating barriers such as passive and active fire protection systems. However, on the other hand, absolute control of the fire risk is expensive. In general, it is acknowledged in industry that one must trade off cost of safety design and the underlying risk related to the facility. In this process balancing cost and risk, the designers are in need of effective engineering tools to timely make effective decisions in complex design processes as well as under varying operational conditions. The project has delivered models including guidelines that enable the designer to identify optimal risk-based solutions of measures that protect structural components and process equipment exposed to fire loads arising under accidents. Some of the models are tailored for oil and gas facilities, but the methods and guidelines are generally applicable for facilities that process, distribute or store hazardous fluids, including facilities processing hydrogen, methanol and ammonia, i.e., particularly relevant within the emerging renewable industry. The key improvement is related to the assessment of passive fire protection of secondary structures (such as pipe supports and pipe racks). Passive fire protection of such structures is a cost-driving element in the industry. However, the generated knowledge base, models and methods do also enhance the work process related to specification of passive protection of process equipment and main load carrying structures. The established models identify temperature response of the exposed objects, where conventional methods are limited to the fire load itself (heat load per area unit). Heat transfer in a fire scenario is a complex process, but with the assistance of analysis of a huge amount of empirical data generated by the developed advanced models, the project has managed to derive simplified models that reflect the driving factors - global design parameters of the facility (size and general layout), inventory contained by process system, capacity of system depressurisation systems, design of active fire water system as well as the properties of the exposed objects (profile, material type and profile thickness). The knowledge generated is made available through a comprehensive report that describes models and guidelines. The models are also commercially available through software tools developed by project partners.

Metoder, modeller og retningslinjer vil bety en betydelig forbedring av arbeidsprosessen for bestemmelse av branntekniske tiltak på komplekse industrielle anlegg - inklusive anlegg innen fornybarsektoren som håndterer farlige stoff (hydrogen, ammoniakk og metanol). Metodene vil gjøre designerne i stand til å finne gode løsninger som ivaretar et krav til sikkerhetsnivå og samtidig er kostnadseffektive. Dette gjelder særlig bruk av ressurser for å bestemme behov for passiv brannbeskyttelse av sekundærstruktur. I tillegg er det en betydelig effekt at høyere nivå av konsistens på tvers av prosjekter og anlegg industrien vil oppnås. Resultatene (modeller og retningslinjer) fra prosjektet vil bli brukt som sentral referanse i 2023 revisjon av NORSOK Z-013 Risk and Emergency preparedness assessment (se www.standard.no).

Current industry practice does in general not adequately reflect the time-dependent behaviour of the load and the non-linear dynamic response of the steel structures and equipment, which in many cases may lead to overly conservative design of the integrity with respect to such loads. On the other hand, in some cases the general industry practice not reflecting the dynamic behaviour of the phenomenon may lead to insufficient integrity, which lead to increased cost of accidents in the long run. There exist advanced CFD models capable of simulation realistic fires in industrial environments and structural response models that reflect the dynamic behaviour separately, but no consistent fully coupled effective risk-based model for engineering purposes that reflect the interdependence between the nature of the fire load and the response model. The research activities executed in the project will build the bridge between the risk factors causing fires, the time-dependent behaviour of fires originating from oil and gas hydrocarbon systems and the dynamic response of objects subjected to the heat load generated by the fire. The results generated by the research activities will provide the basis for tailoring of effective engineering models build into a risk analysis frame work in compliance with regulatory requirements.