DNV Software has been involved in nearly all offshore field developments on the Norwegian continental sector of the North Sea, and has made observations of accidents involving offshore structures and what can be learned from these.
The North Sea was developed to produce oil and gas for more than 30 years, during which time we have gained valuable experience from design knowledge acquired in previous projects and from accidents of offshore structures. We would like to share some of these experiences, as they have relevance to the process of designing fixed and floating offshore structures for other regions with harsh environmental conditions.
What lessons have we learned from accidents, and how can the risks be categorised?
Types of structures used in the North Sea
Since Ekofisk started production in 1974 a variety of structures in the North Sea have been built, installed, operated and now also decommissioned. All platforms in the North Sea must sustain harsh weather.
It is commonly accepted that the era of large platforms in the North Sea is over. New developments are expected to use subsea production units utilising existing infrastructure or new small export satellites. That being said, a large discovery in the northern region may still demand large structures.
Risk
Risk of structure failure can be categorised into three groups:
- Risk of failure resulting from statistical variations in loads and structural loadbearing capabilities
- Risk of failure due to accidents
- Risk of failure due to a gross error during design, fabrication and operation of structures.
Gross errors (human errors) are the main contributors to the failure of structures. Focusing on gross errors is a key factor to ensure project success. One example may be to perform risk assessments or investigate how human interaction can introduce dependencies between independent systems.
Influencing the design rules
There are at least four major disasters that have been important to the industry with respect to development of appropriate rule requirements. These are the Alexander Kielland disaster in 1980 and the loss of the platform Ocean Ranger in 1984, the explosion on Piper Alpha in 1988 and the sinkage of P36 in 2001.
Design is usually based on a combination of empirical and theoretical rules, meaning that at least some of the governing criteria are based on historical data.
Following the Alexander Kielland tragedy the Norwegian Marine Directorate introduced new rules requiring that a unit should remain stable and afloat even after loss of buoyancy volume corresponding to one column on mobile offshore units. It required that volumes above the main deck had to be of a certain size, watertight and separated from the columns. These rules have later been refined, requiring a certain level of righting arm from watertight volumes above the main deck.
The industry made a detailed review of its rules for mobile offshore units, and stricter requirements were introduced with respect to consequence of fatigue cracks and accidental loads.
The losses of Ocean Ranger and Piper Alpha were caused by an initial event leading to a series of undesired fatal and uncontrolled events. As a consequence, the Norwegian Petroleum Directorate built into its rules that the failure of a single component must not lead to unacceptable consequences.
It is understood that it is difficult to build safety into structures by verification and quality assurance alone. In addition, robustness is required. Robustness in design is more than redundancy – it also includes degree of indeterminacy and reserve strength. A failsafe design is one aspect of this approach.
The Ocean Ranger accident has parallelsin other accidents on mobile offshore units where manholes or doors that should have been closed are left open, as on the P36 platform in Brazil in 2001.
Other accidents have also directly influenced design criteria. Examples of such are accidents related to pipelines. Cracks in pipelines have sometimes resulted in large R&D projects (carried out by DNV, for example) which have influenced the recommended practices (design of stainless steel under cathodic protection). These have already been employed on the Ormen Lange field to avoid similar undesired events.
Large accidents in the North Sea
Several accidents have led to a huge number of fatalities, total loss of the structures and pollution to the environment. The most severe ones are Frigg DP1, Bravo Ekofisk, Alexander Kielland, Piper Alpha, West Gamma and Sleipner A.
Alexander Kielland
The fatal accident of this floater in 1980 was caused by initial cracks, probably introduced during welding operations. Small initial cracks were allowed to develop until the brace between two columns failed. This accident showed the need for a detailed fatigue analysis, assessment of members and accidental loads, and how to avoid gross errors in design, fabrication, installation and operation.
Piper Alpha
The world’s most serious offshore oil disaster was the explosion and fire on Piper Alpha in 1988. It started with a condensate leak which was ignited. The explosion led to an oil pool fire burning through the gas risers, engulfing the platform in flames. The accident illustrated the need to ensure that failure of a single component does not lead to unacceptable consequences, likewise the need for implementation of safety procedures.
Importance of analytical design tools
Offshore designers have gained valuable experiences over the years from previous projects – either in the form of lessons learned and what can be improved, or from analysing accidents.
Engineers have also been able to predict the soundness of a structure by using computational tools and the finite element analysis methodology, relying far less on empirical data. But no matter how many analyses are performed, it is always the responsibility of the designer to ensure that the structures will have the necessary quality and robustness – in other words, it is the engineer’s role to make sure that the structure is fit for purpose.
The focus here is on the importance of analytical design tools in the North Sea. We believe this is applicable for offshore reserves in other parts of the world. What to do in the future to ensure best possible utilisation of the oil and gas fields, and develop fields in deeper water?
North Sea characteristics
It is possible to design North Sea offshore structures using more simplified techniques involving hand-based calculations, but it is likely that such structures would be overly conservative and expensive to build.
The development of the finite element computer program Sesam, applicable for small to very large structures, started early in the 1970s when the Ekofisk tank was designed. Sesam has since been used when designing most of the structures in this geographical area. Even before this, DNV used the system when replacing and enhancing its empirical-based ship classification rules with theoretical-based ones.
When developing the North Sea – and very likely other regions as well – the use of finite element (FE) software has contributed significantly in a number of areas.
Cost control and on-time delivery is one area improved by FE software. All project managers are aware of the need to estimate cost and predict delivery – in the North Sea it is not possible to install platforms year round. The use of FE programs assists in predicting the design at an early stage, so that costs and project schedules can be estimated with fairly high precision.
The Brix® Explorer for Sesam allows engineers to define and execute Sesam design tasks. Each work process created may be saved as a template, capturing best engineering practice over time. Templates may be shared throughout an organisation, enabling company-specific guidelines to be built into work processes and ensuring that they are always used. Brix Explorer for Sesam is a system which enables users to implement individual Best Engineering Practice, ensuring optimum performance and high quality.
FE systems such as Sesam support the superelement technique that allows for concurrent and distributed engineering. For example, when the Gullfaks C platform was designed, engineers in Oslo, London and Houston were working on the same FE model simultaneously. Using this technique, several months were saved during the analysis project.
Previously, it was not possible to test a number of alternatives before building the actual structure. Later, more techniques were developed to simulate the real structure in model testing. One could carry out what-if analysis. These days, much of the what-if analysis is done using computers and FE programs. This means that we can investigate many more variations during what-if analysis to determine the optimal design.
In the North Sea, it is necessary to satisfy rules and regulations when constructing offshore structures – for example the British Standard (BS), Norwegian Standards (NS and NPD) and Eurocode. To quality while ensuring compliance, many FE systems have features for automatically performing ‘rule-check’ tasks. This means that the standards must be written clearly and concisely. At the same time, software vendors need to have design competence when incorporating such features in their FE systems.
All North Sea offshore structures are subject to wind, waves and current. During the 1970s, it was not possible to perform hydrostatic and dynamic analysis seamlessly together with structural analysis. Sesam is one example of a system where this has been possible since the late 1980s. It is completely integrated when performing ydrodynamic analysis (and static), structural strength analysis, code checking, fatigue analysis and so on.
Finite element technology was initially available within specialist software tools. Over the years, new techniques such as geometry modelling and concept-based techniques for modelling and strength assessment have been introduced. These techniques have made design of offshore structures (and other structures) much easier, reducing the elapsed time required for each cycle through the design wheel.
Times are changing
It is widely acknowledged that undiscovered oil and gas reserves in the North Sea can only be found using improved production techniques and from undiscovered deepwater fields.
This puts additional requirements on vendors to develop computer programs to help engineers optimise structural design. Examples of this include:
- Coupled analysis where damping effects of lines are correctly accounted for, helping engineers predict a better
response history of floating systems and the fatigue life of the lines. This is already a part of the Sesam system. - More efficient computer programs for predicting the lifetime of pipelines with free spans have been made available
commercially. - Concept technologies used in FE tools will speed up the design process for subsea production structures – either sub-reduce the elapsed time and improve sea or satellite platforms. Concept modelling techniques have been available in SESAM since year 2000.
- Several oil companies have established what are commonly known as re-analysis systems. Their main purpose is to quickly assess the structural integrity if an accident occurs or if the structure is modified. Re-analysis systems established in the late 1980s and early 1990s are now based on concept techniques that allow for a much better 3D visualisation and understanding of the structure, loads and results.
- Many operators have maintenance inspection programs in addition to reanalysis programs. Recently Conoco-
Phillips, Norway established a Structural Inspection Management System for the Ekofisk field. The objective was to provide an engineering environment within which the company can quickly and effectively execute planned and unplanned engineering and inspection planning activities. This system is one of the first to have re-analysis systems integrated with maintenance programs covering five main features: integrated structural inspection planning, execution and reporting, structural analysis and structure integrity status (regulatory conformance), change registration and mitigation (inspection results, modifications), data pool for design premises, and technical documentation and emergency response capability.
Learning from history
We have learned important lessons from experience in oil and gas exploration in the North Sea over the past 30 years. First, we must maintain rules and standards that give the intended safety level. The requirements in the standards should be compatible with available design tools, such as finite element analysis programs. The design standards should be easy to understand for engineers, in order to avoid gross errors. The standards should be specific and not open to interpretation.
Engineers should have relevant education, also including education in preparation of design documentation that can be verified by others. Organisations must take a responsible attitude to competence planning and quality, applying the principles of triple bottom line and Corporate Social Responsibility.
We must identify possible failure scenarios that may lead to critical situations and perform independent verification. Where the consequences of a failure are severe, it is recommended that an independent analysis be carried out. An organisation that puts sufficient emphasis on safety through the operational life of the structures must be maintained to avoid gross errors.