Nov 28, 2016 -- Engineers most often rely on analytical modeling for engineering design and analysis performed using equations to obtain solution for a problem in consideration. While analytical modeling assists engineers to understand the physics involved in the product operation, it increasingly becomes complex when trying to model phenomena with more complexity.
To simulate and predict the behavior of complex 3D fluid flows, industry relies more on the powerful technique called Computational Fluid Dynamics (CFD). The technique makes use of complex differential equations to solve fluid flows involving turbulence, heat transfer, phase changes or chemical reactions. From its first commercial use since 1980s, CFD is extensively applied today in industries like oil & gas, apart from aerospace and automotive, as it allows engineers to optimize equipment designs quickly. Further, advancements in computational power and computer technologies have enabled engineers to simulate fluid flows with greater level of details.
The major advantage of using CFD is that there are no size or weight limitations, and the engineer is only limited by the availability of the computational power. It thus proves valuable in the development of industrial equipment, since the equipment can be evaluated virtually using multiple variables cost-effectively, when compared to physical testing. Unlike complex and expensive physical tests, CFD allows more control over the required parameters in the equipment development such as fluid film thickness, particle concentration, droplet size, etc. Since any number of parameters can be used for CFD analysis, the test results are comprehensive as compared to the results obtained through physical testing.
The benefits of CFD are more significant in refineries where it is important to maximize uptime to ensure maximum profit. Downtimes can impact operating margins drastically and thus requires owners to keep the plant operational and efficient. With CFD, the performance of different components and plant equipment can be evaluated along with useful life prediction efficiently.
Vital applications of CFD that can help in keeping the plant efficient and functioning:
- Predict abnormal distribution of flow in process equipment to avoid catalyst exhaustion
- Predict and improve flow separation and mixing for gas/liquid/solid particles
- Determine temperature distribution and thermal fatigue for mixing tanks
- Couple CFD results with FEA to determine useful life of the equipment and determine maintenance schedules
- Model layout and spacing of equipment to ensure proper air flow and minimize risk of explosion, over pressure and damage due to fire exposure
- Improve flare metering accuracy for both FPSO and offloading vessels and refiners by comparing the flow profile of the as installed case to that of the ideal installation of the flow meter
Although CFD may appear to be a simple solution, there is a lot more complex process involved behind every simulation to produce accurate predictions. With increasing computational capabilities however, it is now possible to perform complex conjugate simulations, enabling engineers to consider multiple physical phenomena in single simulation.
Meshes of the order of tens of millions can be solved easily compared to one million cells in total just a decade ago. These advancements have allowed engineers to address problems that were previously unthinkable, but at the same time, these have also induced the risk of engineers losing the understanding on numerical methods and its application on particular problems. Modern engineering community has begun relying heavily on user friendly software interfaces and is gradually losing the ability to make engineering judgments beyond what the technology informs them.
The ease with which simulations can be performed without much training invokes the danger of engineers attempting to solve the problems without completely understanding all the issues involved. Controlling the sensitivity of the solution is critical by keeping a check on boundary conditions, solution controls, etc., which makes engineering experience extremely important.
Analysts using CFD tools must completely understand the physics that will be solved by the software, to ensure better and meaningful results. To reduce modeling errors, it is also important to utilize physical data for validation. The CFD model can also be validated with analytical calculations or the model can be cross-verified using multiple software solutions or modeling approaches.
Conclusion
Computational Fluid Dynamics (CFD) is as such not a simple but a complex tool that can be utilized for wide range of industrial applications. As accuracies in this approach continue to increase, physical tests are likely to reduce, as only one validation test will be required to compare the model accuracy. While this stage may still take some time, it is appropriate to consider CFD as a prediction method.
Without proper validation, there are no means to check the accuracy of the simulation and this comes with a good amount of engineering experience in CFD modeling, comparison with analytical approaches and experimental testing.
About Author:
Mehul Patel specializes in handling CFD projects for Automobile, Aerospace, Oil and Gas and building HVAC sectors. He works as a CFD consultant with Hi-Tech CFD for the past 5 years and has successfully executed numerous CFD projects of high complexities.