What are the most common errors and pitfalls to avoid when using process simulation software?

In my experience, selecting the appropriate software and correlation for hydraulic pipeline calculations is crucial. Many simulation softwares offer dozens of correlations for horizontal, vertical, and inclined pipelines, along with recommendations. However, the type of fluid being simulated must also be taken into account. For single-phase streams, several softwares such as HYSYS, UniSim, and PIPESIM can be used. However, for two or three-phase streams, more advanced software/correlations such as PIPESIM and OLGA should be utilized.

Process Simulations are perhaps the most grinding part of process design as most of the simulation tools do not have physical and chemical properties of majority of the molecules ( barring a few popular ones). Hence choosing a correct "Property Method" becomes absolutely essential as this will decide how close our simulation results are from actual parameters observed in a lab or in a plant. An expert often takes the help of regression in simulators like ASPEN PLUS to adjust a property methods which results in fine-tuned reliable simulation models.

For hydraulic calculations including surge analysis, multiphase flow, chemical reactor calculations, heat exchanger evaluations, upstream and downstream processes, emergency depressurization, etc different software needs to be applied. It is not possible to do them in one software accurately or even the software is not able to do some of them. Thus understanding the needs and finding the right tools is critical. In general, the minimum required skills for a process engineer to select the right software and use it effectively are: - Proper understanding of Fluid mechanics, Thermodynamics, Heat transfer, and mass transfer. - Good knowledge of the process. - Strong analytical skills. - Being aware of the abilities of the software in use.

One of the main issues which are happening during the usage of the simulation softwares, is the fact that, in many cases, the engineers without enough field experience are using the same packages for all cases (for example many of them just choosed Peng-Robinson equation for 99% cases in Aspen). It makes difficulties in soft by itself as it gives many errors during creating some cycle further. Moreover, some process engineers are not paying attend to chemical points of the process. Just knowing variables and equipment is not enough for good simulation. Phases of the flow, it's chemical conversion proximity and etc. has also to be considered during the simulation, and it during the choose of correct simulation software.

En general el Ingeniero de Procesos Trabaja en varias áreas donde emplea simuladores: los dise?os de equipo, composiciones, propiedades de sustancias se determinan mediante simuladores de procesos en estado estacionario, en gran parte de las consultoras del area de Oil&Gas prefieren usar Aspen HYSYS, UniSim o ProII (Aveva). Por otra parte la segunda área se refiere a los cálculos hidráulicos (caída de presión en tuberías) para lo cual se emplean software propios de la empresa consultora (para Fluido Monofásico) o comerciales como PIPESIM, PIPEPHASE o AFT (Fathom, Arrow, Impulse). En algunos casos cuando se requiere evaluar filosofías de control o condiciones transitorias de operación se realizan simulaciones dinámicas.

During design conceptualization, some assumptions made during simulations may actually be far from the real conditions. Parameters such as pressure, temperature, composition, etc may have constraints in the actual scenario. These data may not be readily available but could later on impact the overall performance of a system with regards to safety, operability, and cost. It is important to verify these assumptions from SMEs, regulations, standard practices, and vendors early on to avoid risks.

The use of design criteria are important. All key data used for developing mass balances should be in the design criteria and include the type, and source of that data whether it is an assumption, test data or industry standards.

In HYSYS, for sensitivity analysis, you might assess how changing assumptions, like feed compositions or equipment specifications, impacts key outputs such as conversion rates or energy consumption. This aids in identifying critical factors influencing the simulation results.

If a significant deviation in the process simulation's product rates is observed, one of the ways to address it is by modifying the equations of states (such as Peng Robinson, SRK, etc.) and binary coefficients, taking into account the properties of pseudocomponents based on PVT research. This approach has the potential to significantly improve the phase distribution of components.

Always remember that gargabe in equals garbage out. Tips when using a process simulation software - 1. Understand the assumptions you are making and their limitations. 2. Choose the right fluid package. This is very important especially when evaluating critical parameters that impact safety and integrity of operations. 3. It will always be better to validate your simulation with real data. This would indicate whether you model is really reflecting the actual process.

The predominant error often observed is neglecting dynamic effects in time-dependent processes. When conducting dynamic simulation, it is crucial to account for transient effects and comprehend the limitations associated with steady-state assumptions.

The user shall not rely on the default methods in the softwares such as UniSim and AspenHysys. In most cases, they must be changed to meet the process requirements. Some examples are: - the methods for pressure loss calculations, especially for multi phase flow. In some cases, a different software has to be used for pressure loss calculation such as Olga. Another example is when a pipeline is being depressurised, the velocity in the piping downstream the orifice might be sonic (Mach number of 1) which is not considered by some of the softwares and velocities higher than sonic velocity is calculated by the software and it gives completely wrong pressure profile in the flare system. - the methods for PSV sizing, - etc.

Trusting any simulator is a matter of belief: Knowing or at a minimum having a reasonable guess of the most likely outcomes is a necessary sense check. When you see trends that do not fall in your experience, ask yourself why and find the variables that make the big impact. Then ask yourself: how likely is it that these variables are known or calculated correctly Then do the sensitivity analysis with the thought: If I am wrong, how would I know? Well, now you are using the Simulator,.your knowledge and intellect to find answers. And often you will learn something you did not expect. Next time, you are less arrogant and have greater knowledge.

In process simulation, having a suitable property package is crucial. When dealing with mixtures that contain both hydrocarbons and non-hydrocarbons like CO2, H2S, alcohols, glycols, esters, and organic acids, it's best to opt for a cubic equation of state such as Cubic-Plus-Association (CPA) for Aspen HYSYS. This equation combines the SRK cubic EOS with the Wertheim association term, resulting in improved accuracy of component interactions and phase distribution.

One approach to resolving anomalies in process simulation results is to consult appropriate books or literature on the subject matter. These resources can provide insight into potential sources of error or offer alternative methods for achieving the desired outcome. It is important to ensure that the resources consulted are reputable and relevant to the specific simulation being conducted.

Don't forget to use a checklist to verify that you are covering technical and quality aspects. Quality is part of a Top ngineering delivery.

A process engineer has to analyze the results critically and be familiar with typical root causes. There are some unit operations in the common process simulation softwares that have high default values for tolerance levels which can potentially create considerable errors in the results of the simulation. Heat & mass balances in steady-state mode need to be reviewed. The input rates and conditions have to be checked. Each unit operation needs to be reviewed critically to prevent surprises. As an example, when component accumulation occurs in the systems, it can result in very high flow rates to the affected equipment which occurs because the simulator wants to converge thermodynamically.

as engineers, it's our responsibility to meticulously review and verify simulation results. I strongly believe that attention to detail and a comprehensive understanding of the underlying processes are indeed an art for an engineer, and they are paramount when it comes to verifying simulation results.

Documenting your work helps, especially in squashing bugs or identifying areas of concern in succeeding simulation revisions. This is a requirement in some simulation companies as these assumptions form the basis of the design of all other deliverables.

One easy method of documenting your work I have found is to just save benchmark files. I have my process modeled up to this point saved by name. Now "save as" your new file name_tempchange or name_flow update. Get into this habit and you have a ton of files, but organized well and you can easily fix any issues or go back to where you were before something errors or corrupts.

First and foremost, documenting provides a record of your methodology, assumptions, and results, ensuring transparency and reproducibility. This documentation allows others to understand and verify your work and knowledge sharing within your team or organization. Additionally, thorough documentation serves as a reference for future analyses or revisions.

I use to define a simulation sequence to follow the scenario/options for the simulation. A sequence of ID numbers is linked with each option and documented in separated carpets.

Just having a great simulation software will not solve the problem. Remember that it is just a calculator and garbage in will only lead to garbage out. A lot of engineers focus on learning the tool and do not focus on the basic concepts. Tool should be used as a helper, then only you can use it effectively and efficiently

A crucial recommendation for utilizing process simulation software is to consistently update and enhance your skills. Process simulation software undergoes continuous evolution and improvement, with the addition or modification of new features, functions, and capabilities. Hence, it is essential to stay informed about the latest developments and trends in process simulation software, learning how to leverage them effectively and efficiently. To update your skills, consider engaging in activities such as reading manuals, tutorials, blogs, or articles, watching videos, webinars, or podcasts, taking courses, workshops, or seminars, and participating in forums, communities, or networks.

In process simulation, the old mantra applies: "use it or you will lose it". Keep sharpening the saw, as Stephen R. Covey will put it, by keeping abreast of new developments in process simulation and learning how to use the new features in each software.

staying updated as an engineer in simulation is crucial in today's rapidly evolving landscape of new methods and technologies. For instance, emerging concepts like hydrogen utilization and discussions surrounding CCU are reshaping the simulation field. Attending workshops hosted by experts in these areas provides invaluable opportunities to learn about the latest advancements, exchange ideas, and enhance proficiency in applying simulation techniques to tackle these novel challenges.

One prevalent mistake involves inconsistent units in input data. Adhere to consistent units throughout the simulation in accordance with project requirements to prevent miscalculations and misinterpretations of results.

Another aspect here is about version of the software tool which is used for generating results. Most of the times, results generated using lower version of tools provides different results when run into higher/newer version of the same tool.

Overall, the accuracy of simulation can never be taken as 100%. In my opinion, high number of reactions and their complexity, polarity of involved molecules, azeotropic mixtures, solid handling, polymerization, etc. are some of the factors affecting the simulation accuracy. As an example from refinery, you can usually rely on the CDU/VDU unit simulation in Aspen HYSYS. For FCC unit, however, the accuracy is around 40% less than that of CDU/VDU. It is not worth mentioning DCU unit here, which would direct us in completely wrong direction in Aspen HYSYS.

As always, the main pitfall is the amount of difference between the simulation and the real process. Multi-context analysis is not a simple task. The more contexts you will correctly include into the simulation, the better the risks are managed.