In recent years, Computational Fluid Dynamics (CFD) has come to the fore as a valuable and increasingly usable tool for analysis of system performance and assistance with system design and operation. As part of our commitment to constantly improve our understanding of fermentation processes, CPI have invested in the development of an in-house modelling & simulation capability, focusing both on CFD and process modelling.
CFD is often seen as a bit of a mystery, a tool used exclusively for highly complex problems, spending vast computational resource to visualise and analyse the finest details of a problem. What is often overlooked is the fact that CFD can be used just as effectively as a simple tool for solving everyday problems and troubleshooting plant performance and design issues. Fundamentally, CFD allows us to see the inner workings of a process in far greater detail than can be achieved through instrumentation alone, allowing us to analyse process design and identify performance issues and potential design improvements quickly and effectively.
As an example – Say we have 1m³/s of gas flowing through a pipe with a cross sectional area of 1m², the velocity of the gas in this pipe is 1m/s. Right?
Not necessarily… Whilst the average velocity of the gas through the pipe system will of course be 1m/s, the way the gas flow behaves through bends can lead to high and low velocity zones, potentially affecting process performance.
In the situation above, there are distinct high and low velocity zones, which could lead to issues with pipe wear or hot spots if the carried fluid is high temperature or dust laden. There is also a potential issue if the fluid is carrying a suspended solid, as the low velocity zones may allow the solid to drop out of suspension. In a quick 5 minute CFD simulation here, the arrangement of this bend can be optimised to reduce the extents of the high and low velocity points and maintain a smoother flow profile overall.
The Industrial Biotechnology & Biorefining technology of CPI aims to support innovation by providing access to assets and expertise in the area of process scale up. Working with innovators to help take their processes from the idea stage, through to a commercial product backed up by a robust manufacturing process. In order to do this, it is essential that engineers and scientists at CPI have a thorough understanding of the scale-up characteristics of the equipment available on site – ranging from ml scale fermenters all the way through to our 10,000 litre demonstration scale plant.
With that in mind, a CFD modelling exercise has been carried out in order to assess all of CPI’s stirred tank fermenters, with the aim of mapping the changes in performance criteria as we move up through the scales. Whilst the results of this exercise will be useful in helping to scale-up processes, it will also then be useful in scaling down existing processes to lab scale, as we will be able to advise on operating criteria to mimic the performance of the larger scale fermenters in the lab.
In addition to an improved understanding of the performance characteristics of our own fermenters, this exercise then also allows CPI to confidently predict the performance of theoretical fermenter designs outside of the scales available at the National Industrial Biotechnology Facility at Wilton. This in turn helps our customers to scale their process beyond pilot and demonstration scale and through to full production scale.
CFD modelling also proves to be particularly useful in assisting with virtual prototyping of potential new equipment designs, helping to guide and advise laboratory based testing to more effectively use time and resources in the lab. At CPI, this has recently been put to use in the development of a new type of reactor design — Here, a modelling exercise was carried out which involved the rapid testing of around 25 design and operational options for the reactor in CFD in order to quickly identify areas with the greatest potential for performance improvement against the base design.
In theory all of the proposed options should have been able to realise some performance improvement, however physical prototyping of all design options for testing in this case was simply not feasible due to the massive time and cost demands involved. CFD modelling then provides a quick and effective method of quantifying the comparative performance between the various options, narrowing down the list of design options to those most likely to realise a performance benefit. In this case then only those most promising options were put forward for physical testing, resulting in a significant cost and time saving in the project and leading to more focused development of an optimised reactor design in a relatively short time period.
These examples all highlight the growing importance of CFD simulation not only in industrial biotechnology, but in process design in general. Whether it be a full reactor design model or a simple design check on a pipe arrangement, with a greater understanding of the performance characteristics of your process, more informed decisions can be made around improving that performance through design and operational adjustments, leading to a more robust, reliable & efficient process in the end.