0
Select Articles

Analysis LED Design in Subsea Engineering PUBLIC ACCESS

Challenges for Modeling and Simulations of Components and Systems

[+] Author Notes
Egidio (Ed) Marotta

GE Oil & Gas Multiphysics Simulation Group Advanced Technology Organization

Ed Marotta achieved his B.S. degree in chemistry from the State University of New York at Albany, 1980, post-graduate studies in chemical engineering at SUNY at Buffalo, 1982, and his MS and PhD degree in mechanical engineering from Texas A&M University in 1994 and 1997, respectively.Ed’s present responsibilities involve the management of the Multi-Physics Simulations Group for GE Oil & Gas – ATO. In this capacity, Ed is responsible for leading a group of engineering specialists responsible for performing thermal, diffusion, and multiphysics analyses on all major systems and sub-system components, development of thermal best practices for multiphysics analyses specific to GE products. In addition, Ed is responsible for the development of low-dimensional, lower-order tools for integration and integrity management of subsea systems.

Mechanical Engineering 137(03), S8-S12 (Mar 01, 2015) (5 pages) Paper No: ME-15-MAR-7; doi: 10.1115/1.2015-Mar-7

This article highlights some of the analysis challenges, which must be resolved, such as the examples of complex subsea equipment analyses conducted on trees and manifolds, and a vision for the future as it pertains to modeling and simulation in the context of an analysis-led design {AQ: Some edits are made in this sentence, “This article…led analysis.”. Please check and correct if necessary.}. This is just one roadmap that can have a profound impact on innovation within the Oil & Gas Industry, especially, for subsea and onshore production equipment and systems engineering. The advantages of optimization in design, especially in the early up-front engineering stages, can be best illustrated when designing a heat exchanger. While technical challenges still exist with respect to developing more accurate models, whether fluid, structural, acoustic, magnetic or multiphase flow models for boosting pumps, for subsea applications, the concept of using analysis led design methods in the up-front engineering phase has been well proven in many industries. The challenge still lies in gaining its acceptance as a routine practice for offshore and onshore applications to increase engineering efficiency.

The use of sophisticated modeling and simulation techniques, software, and hardware for the verification of design concepts is of utmost importance. This is especially true as production operators and service providers operate and design, respectively, subsea equipment for high-pressure high-temperature (HPHT) wells and for field equipment that may experience 10,000 ft. of seawater depth. Oil and gas fields reside beneath many inland waters and offshore areas around the world, and in the oil and gas industry the term subsea relates to the exploration, drilling and development of oil and gas fields in underwater locations. Many new reservoirs are exhibiting well conditions that can reach temperatures greater than 350 oF and pressures greater than 15,000 psi, thus the well is labeled as HPHT.

To ensure that hydrocarbons easily flow within subsea pipelines and equipment, thus preventing the formation of hydrates, waxes, and asphaltenes, and that reliability issues, such as metal fatigue, are properly addressed during the design phase, analysis led design will incorporate the use of modeling and simulation software. Such tools as ANSYS Mechanical and/or Abaqus for structural analyses, and Fluent, CFX, and STAR CMM+ for computational fluid dynamics (CFD) investigations will be significantly used. In addition, the interaction of physics such as fluid-structure interactions (FSI), vortex-induced vibrations (VIV), and flow-induced vibrations (FIV) have become of greater significance since these phenomena can lead to early-life fatigue failures (e.g., subsea jumpers).

Optimization of designs prior to validation can reassure the design engineer that the very best design goes into testing, that both functional and performance requirements are met, and maximum cost savings can be realized. Hence, the use of higher fidelity models, (e.g., 3-D versus 2-D or 1-D) with their higher nodal and element count requires novel analysis thinking, a new analyst/engineering skill set, and closer collaboration among the various engineering and design disciplines.

This article highlights some of the analysis challenges, which must be resolved, examples of complex subsea equipment analyses conducted on trees (XTs) and manifolds, and a vision for the future as it pertains to modeling and simulation in the context of analysis led design. This is just one roadmap (i.e., they exist in other industries) that can have a profound impact on innovation within the Oil & Gas Industry, especially, for subsea and onshore production equipment and systems engineering.

The development of complex products such as subsea trees (i.e., assembly of valves, spools, and fittings used for an oil well that controls flow) , manifolds (i.e., a unit that transfers oil /gas from wellheads into a pipeline), BOPs (blow-out-preventer), jumpers (i.e., piping that connects trees to a manifold), separation systems (i.e., used to separate hydrocarbons from sand and water phases), boosting systems (single and multiphase pumps), piping systems, and riser systems requires, at times, numerous build-and-test hardware prototypes. To validate product performance (i.e., specifications) and government regulatory requirements along with the need to verify stress and fatigue life concerns is tremendously expensive and time-consuming, to say the least. This issue can be addressed up front in the development cycle by evaluating and refining/ optimizing the designs with analysis tools so that fewer test/ validation cycles will be needed in the later parts of the development phase.

Analysis led design is one of the latest strategies in the product development process. Its emphasis is more on up-front engineering with a product quality optimization process that starts early in the conceptual design stage. Product development and manufacturing through analysis led design utilizes digital tools extensively for design, analysis and product optimization with the hope that virtual testing is not too far behind. The end goal is to produce products that meet customer needs quickly and inexpensively and thus, ensure their success.

An oil and gas analysis led design initiative is needed to change the prevalent test-first culture, which would have a major impact on the industry, with significant benefits that include shorter development time, lower cost, improved product robustness, and greater reliability. Analysis led design can significantly shorten product development time by getting designs right the first time where traditional hardware testing can take months or years to validate a design. Leveraging analysis early in the development process can eliminate design changes and repetition of lengthy endurance testing, thus providing tremendous reductions in overall development time due to reduction in long-hour tests that can last for weeks or months.

The importance of the analysis led design approach is clearly illustrated when placed in the context of the V-Model development process. The first leg is the project definition where the user’s needs, performance requirements, conceptual and detail design are formulated prior to implementation and validation. Using analysis led design, the concept, technological characteristics, theory and mechanisms of how the product would perform are extensively studied and clarified. The use of simulations allows for the speed-up of “what-if’ scenarios and optimization.

Figure 1 shows the utilization of analysis led design in the overall development process in the V-model. The analysis activities of modeling and simulation, virtual prototyping, parametric design studies, and optimization are all incorporated to achieve the end goals of minimizing the development cycle, rework, errors in design, costs, and to meet customers’ needs per the requirement’s analysis while enhancing design robustness.

FIGURE 1 Positioning of analysis led design activities in the up-front engineering phase.

Grahic Jump LocationFIGURE 1 Positioning of analysis led design activities in the up-front engineering phase.

Product development is a thorough process that requires a huge amount of resources to be channeled throughout in order to obtain a fully functional, well-developed product. It consists of brainstorming the general concept of the product and deciding its function, design and methods of manufacture, as well as its marketing method.

A good and reliable product must fulfill several characteristics such as being able to function efficiently, being of very high quality, and most importantly being able to satisfy the customers’ needs. Higher-order 3-D simulations allow the design engineer to thoroughly investigate the design’s characteristics in virtual space, without expensive pre-testing of prototypes, to verify the concept’s performance against specifications set by the customer.

Such tools as ANSYS Mechanical and Abaqus for structural analyses, and Fluent, CFX, and STAR CMM+ for computational fluid dynamics investigations are used significantly in the oil and gas industry. Other commercial software tools, such as RIFLEX and ZENRISER, are employed for static and dynamic analysis of slender marine structures. The dynamic behavior of offshore flexible or rigid riser or pipe systems is subject to hydrodynamic loading and vessel motion that need to be modeled to ensure reliable operation. They represent simulation technology used for riser analysis suitable for flexible, metallic or steel catenary riser applications.

In offshore applications, subsea trees and manifolds are employed for production purposes. Due to the harsh corrosive environment and constant cold condition, which approaches near freezing temperature, thermal insulation is used to ensure the reliable and cost efficient management of flow of hydrocarbons from the reservoir to top-side floating production vessels or via subsea piping systems to onshore processing facilities. The objective many times is to determine whether or not the thermal insulation system on a subsea tree and/or manifold is capable of maintaining the produced fluid temperature above hydrate formation temperature (HFT) for a specified period of time (e.g., greater than 12 hours) after shutting down the well. Higher-order fidelity, 3-D, transient computational fluid dynamic modeling is used to simulate such operating requirements. Figure 2 shows the meshing scheme used to model such a system to guarantee convergence in a timely manner.

FIGURE 2 Meshing used for the tree - CFD Modeling: 4.7 million polyhedral cells (6-layer fluid boundary inflation).

Grahic Jump LocationFIGURE 2 Meshing used for the tree - CFD Modeling: 4.7 million polyhedral cells (6-layer fluid boundary inflation).

The results of the steady state and transient cool-down time of a multiphase flow is shown by Figures 3 and 4, respectively, where red indicates hot and blue indicates cooler temperatures.

FIGURE 3 Steady-State temperatures in the subsea production tree during normal operation.

Grahic Jump LocationFIGURE 3 Steady-State temperatures in the subsea production tree during normal operation.

FIGURE 4 Transient temperatures after 8 hours of cool down time in the tree.

Grahic Jump LocationFIGURE 4 Transient temperatures after 8 hours of cool down time in the tree.

A “tree” is an assembly of valves, spools, and fittings used for an oil well, gas well, condensate well and other types of wells. The primary function of a tree is to control the flow, usually oil or gas, out of the well. A tree may also be used to control the injection of gas or water into a nonproducing well in order to enhance production rates of oil from other wells.

The modeling of the actual physics (using CFD) of the fluid provides a more realistic insight of the actual thermal performance than traditional methods of determining cooldown times using finite element techniques. This is further magnified once the analysis complexity is increased by the introduction of a multiphase fluid in the flow stream instead of a single fluid (e.g., gas or liquid).

A subsea manifold is a large metal piece of equipment, made up of pipes and valves and designed to transfer oil / gas from wellheads into a pipeline. Manifolds are usually mounted on a template and often have a protective structure covering them. Manifolds vary greatly in size and shape, though these can be huge structures reaching heights of 30 meters (90 feet).

Figure 5 shows analysis results for a subsea manifold for a transient cool-down simulation.

FIGURE 5 Transient temperatures after 8 hours of cool down time in the manifold.

Grahic Jump LocationFIGURE 5 Transient temperatures after 8 hours of cool down time in the manifold.

The thermal design challenge becomes the determination of thermal insulation thickness and its placement to minimize thermal losses and thus prevent hydrate formation. Hydrates are ice-like crystals that form with natural gas and water and at combination of low temperatures and high pressures. Thermal insulation on subsea trees and manifolds is always required to prevent the rapid decrease of produced fluid temperature below hydrate formation temperature.

With focus more on up-front engineering, a product quality optimization process that starts early in the conceptual design stage is necessary. The analysis led design method adopts an active and advanced approach to respond to customers’ demands. Rather than acting only when the problems occur, it considers every possible problem and solves them in advance, early in the product development stage. Product robustness and quality are enhanced with the use of optimization software tools.

Commercial software tools such as HEEDS and Isight allow for multidisciplinary design optimization. Whether the analysis is structural (linear or nonlinear, static or dynamic, bulk materials or composites), fluid, thermal, magnetic, or acoustic or a combination of these types of analyses (e.g., fluid-structure interaction, vortex-induced vibration, etc.) optimization techniques in conjunction with computer-aided engineering tools can help identify the optimal solution or the envelope for safe and reliable operations. When it’s important to predict design sensitivities, or to gain a clearer understanding of the design space, a design of experiments (DOE) study is often the ideal approach. DOE methods allow for the extraction of a great deal of useful design information quickly, with the least computational effort possible.

The advantages of optimization in design, especially in the early up-front engineering stages, can be best illustrated when designing a heat exchanger. Heat exchangers for subsea applications, designed for best performance, must conform to multiple performance requirements. To determine the optimal design point, a multidisciplinary team of engineers must identify the significant parameters that affect the performance of the heat exchanger and also, interactions between significant parameters. Design engineers need to be smart in selecting enough data to accomplish the objective, but within available resources. What are the options? A design of experiments method with screening designs, surface response, or full factorial design techniques or specialized optimization algorithms such as genetic algorithms, simulated annealing, non-linear sequential quadratic or simultaneous hybrid exploration of the design space. The latter can provide both global and local optimization at the same time. Figure 6 illustrates the procedure of connecting intelligent optimization tools with CAE tools for determination of the design space through exploration.

FIGURE 6 Exploration of the design space through optimization algorithms

Grahic Jump LocationFIGURE 6 Exploration of the design space through optimization algorithms

Figure 7 shows the results of performing an intelligent search of the design space where the exploration provided three viable design solutions versus the traditional one solution with DOE surface response. The best choice will be determined on the basis of cost and manufacturability since each meets the performance requirements.

FIGURE 7 Results of the design exploration space with smart optimization algorithms.

Grahic Jump LocationFIGURE 7 Results of the design exploration space with smart optimization algorithms.

The design exploration space involved 183 design evaluations where 137 were deemed feasible (~75%), 7 were infeasible (~4%), and 39 were errors (~21%), having convergence issues in the CFD analysis.

The benefits from performing an optimization became apparent when the results revealed that more than one design option met all performance specifications and constraints imposed on the analysis. While the computational surface response analysis did provide one good design option from a total of 21 designs that were evaluated, the optimization study evaluated 183 options with three options providing the best solution. Total time duration for the surface response evaluations was 6 weeks; the optimization study required 2 weeks. The increase in engineering efficiency can clearly be quantified from this investigation.

A1-D simulation model is a mathematical representation of a system and its dynamic behavior in the physical world. For example, bringing models together allows engineers to better simulate and evaluate how a multiphase pump design performs in context, when given a gas volume fraction (GVF) load, requirements for total pumping head versus volumetric flow rate for best efficiency point, and even on specific reservoir locations with their unique inlet boundary conditions.

In thermo-fluid applications it is known that 3-D CFD simulations can provide detailed insights about fluid and flow properties in complex 3-D domains. However, from a systems level perspective, 1-D CFD simulations can give important information with respect to performance of an entire system of internal flows. The drawbacks of the two simulation methods are that the former requires high computational costs while the latter cannot capture complex local 3-D features of the flow. Therefore, the two simulation methods must become complementary; indeed a coupling of the two methods can incorporate the strongest attributes of the two methods while minimizing their drawbacks. Here lies the challenge for subsea applications.

The coupling possibility is not limited to the CFD field but can extend to multiphysics. An example of multiphysics one-way coupling is the simulation of vibrations in piping systems, fluid-structure interactions, (e.g. compressed gas systems, blow-down systems); this simulation is performed by modeling the pressure wave propagation inside the piping system with 1-D modeling and passing the forces exerted by the internal flow to a structural analysis tool for mechanical analysis. The interest in multiphysics modeling with one-way and two-way coupling is of the utmost importance in subsea applications.

For subsea offshore applications, this involves the simulation of all potential subsea equipment that may encompass what some people have coined, “the subsea factory.” The dynamic response modeling of an entire subsea production factory would allow the placement of various installations such as trees, manifolds, separation systems, single phase and multiphase boosting pumps and compressors in strategic locations where parasitic losses due to pressure drops, in piping systems such as jumpers, can be minimized while maximizing the safe and reliable production of valuable hydrocarbons.

Figure 8 shows a schematic of a two-point coupling model. In the two-point coupling model the outlet of the upstream is modeled using 1-D physics, which then is coupled with the inlet to the 3-D modeled bend, using a CFD tool, and the outlet of the bend is coupled with the inlet of the downstream section of pipe, again modeled using 1d physics. Thus, a coupling of the 1-D horizontal and vertical pipe sections with the 3-D modeled elbow is achieved and analyzed.

FIGURE 8 Schematic of the two-point 3-D/1-D coupling model.

Grahic Jump LocationFIGURE 8 Schematic of the two-point 3-D/1-D coupling model.

As shown in Figure 9, the combination of coupling 3-D to 1-D modeling gave reasonable prediction of the flow field and thus, the forces acting on the bend induced by the fluid. This coupling technique allows for decreased computational needs while increasing the speed for the analysis.

FIGURE 9 Velocity profile on the horizontal cross section of the bend.

Grahic Jump LocationFIGURE 9 Velocity profile on the horizontal cross section of the bend.

In summary, while technical challenges still exist with respect to developing more accurate models, whether fluid, structural, acoustic, magnetic or multiphase flow models for boosting pumps, for subsea applications, the concept of using analysis led design methods in the up-front engineering phase has been well proven in many industries. The challenge still lies in gaining its acceptance as a routine practice for offshore and onshore applications to increase engineering efficiency.

Copyright © 2015 by ASME
View article in PDF format.

References

Figures

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In