8 replies to this topic

[Allen]

This is more of a discussion point than a question but I put it here because it relates to students.

As a somewhat older engineer who has never used simulation software, the reading of various threads has raised a question in my mind. As I understand it, simulation software can perform calculations which quite honestly could not be undertaken by hand. It also seems that software is available to those undertaking undergrad courses.

My question is really, is this a good thing? let me clarify. Are students provided with the necessary theoretical grounding which allows them to understand what the simulation software is doing, understand its limitations, and critically assess the results before they gain access to the packages?

Some years ago, an experiment was performed where a calculator was modified to give incorrect answers. This was then used at a fruit stall in a market to calculate what people had to pay. Almost invariably the customers accepted the results even if they were excessively high. Why? well because the calculator said so and it must be right. The customers did not use their own mental calculations to verify the answer.

Do we get a similar situation when students use simulation software?

So my question: Are simulation packages a good thing for students or should they only be in the domain of the qualified engineer?

This is not intended to criticise students, its just a question which has been teasing me.

[Naimanb]

As someone who just recently got their undergrad in Chemical Engineering about a year ago I feel I can weigh in on this.

At the University I attended we did have access to the Aspen simulation package. However, the prerequisites for the course were such that students were supposed to understand the methods being used by the software to solve the calculations, and be able to critically question the answers if they seemed unreasonable. As is the case with everything there were some students who used their critical mind, and then there were others who simply believed the answers the computer spit back out at them.

Personally I believe it's a good thing that undergrad students have access to such powerful simulation packages. Some companies have these packages and expect recently graduated students to know how to use them. I think the really important part is having good professors who make students explain the answers they get using simulation packages in order to demonstrate that they understand the mechanics behind the program.

That's my $.02.

[Qalander (Chem)]

Dear All, since this is a sort of 'An Open Discussion'

I understand that in present age of scientific development/innovations;there has to be an optimum mix of the two approaches in such a way

(1) that Basic minimum calculation skills are not lost.

(2) Whereas learning useful programs to better/speedily attack more complex problems of larger process plants

Best Regards
Qalander

[Zauberberg]

In my opinion, the lack of field experience for a process engineer is the biggest of all issues; it can hardly be overcome by comprehensive learning from textbooks or by software training, or by design experience in an engineering office. Too often, equipment simply does not perform as per design, and the knowledge gained during commissioning, startup, field troubleshooting, equipment performance rating etc., is something which cannot be replaced even by holding PhD in chemical engineering, or by mastering all software in the world. Design is based on process design standards, while identifying and removing the problem out there in the field is the true demonstration of engineering knowledge.

As far as students are concerned, it's good to have both things in parallel - in the sense that someone should be able to do the calculations manually before he/she can think of using simulation software, or at least to have an idea/concept how the calculations should be executed. If there is no basic understanding of equipment design and operating principles, if logical and cause-and effect way of thinking is missing in an engineer's mind, then software knowledge is useless; such person will never know if design work that he/she perfomed is accurate or not, and what are the consequences of such results. It's like the story about pocket calculator. It's the pure example of wrongly chosen profession - you cannot understand something which you are not in love with.

[Naimanb]

In my opinion, the lack of field experience for a process engineer is the biggest of all issues; it can hardly be overcome by comprehensive learning from textbooks or by software training, or by design experience in an engineering office. Too often, equipment simply does not perform as per design, and the knowledge gained during commissioning, startup, field troubleshooting, equipment performance rating etc., is something which cannot be replaced even by holding PhD in chemical engineering, or by mastering all software in the world. Design is based on process design standards, while identifying and removing the problem out there in the field is the true demonstration of engineering knowledge.

I've got something I'd like to say about this. I think experience and knowledge must have a good balance as well. I'm currently working my first job in an engineering office after obtaining my bachelors degree. In attempting to do the design on the projects which I have been given I have had to ask my colleagues many questions in order to gain their knowledge. Quite often it seems that "Tribal knowledge" is relied on very heavily and no explanation can be given for the reasons things are done/designed certain ways other than, "that's how we do it." Experience is great, but there comes a point when it is relied upon too heavily and people cease to be innovative in their designs.

[Qalander (Chem)]

The point very justifiably made by our friend 'Zauberberg' is cause and effect and correlation-ships with field problems, their trouble-shooting in comparision to design.

As regards our young friend 'Naimanb' reference of non explaining the logics behind the 'usual ways' employed by people with many year's experience; this is a sort of generally faced typical un-healthy attitude witnessed in industry.

Best Regards
Qalander

[ankur2061]

Allen,

In my opinion a simulation software is as good or as bad as the person handling it. To me, in a simulation software

Garbage in = Garabge out

Frankly speaking simulation softwares have been to a certain extent responsible for the demise of math and calculation skills and to say the least, a lack of creative thinking.

Personally, my curiosity levels and my inherent lack of blind faith in simulation softwares has always led me to do my own calculations However, I can be pointed as guilty of having extensively used MS-excel to create spreadsheets based on my limited understanding of the vast and beautiful subject of chemical engineering. But in my defense, I would like to say that I have invariably checked all my calculations by hand and have extensively used notes, footnotes, definitions & references in the spreadsheets I have created to get a clear picture of the origin of the calculation, the purpose and the fundamentals behind the calculation. To that extent, I certainly am grateful to Microsoft for creating such a wonderful software such as MS-Excel.

To end, I would just like to say that what is required is a judicious blend of personal calculation skills as well as usage of advanced simulation software.

Regards,
Ankur.

[pjadams]

After reading all the expert opinions, I have to admit that I am in a category of engineers who almost blindly believe the simulation results. I think the technology has researched a point that, whether you like or not, you have to depend on the "garbage" coming out of the machine. Sometimes it is beyond my capability to question all the results from the cases I am dealing with.

For instance, I would create a similation model for a whole gasification plant, starting from gasifier, to syngas scrubbing, CO-shift, low temperature heat recovery, sour water stripping, Selexol unit, Claus sulfur recovery, hydrogen purification, methanation unit, etc, etc......

Each of the units I mentioned could have 10 heat exchangers, 2 - 3 distillation columns, 10 pumps and 10 K.O. drums, the totally could end up with 800 pieces of hardware, costing 4 billion $ to build. How can I come up with a stream table (or HnM Balance sheet) in a few month if I use a little TI calculator? How do I close the loops of recycle streams with the consideration of all the VLE and chemical reaction involves, Henry Law, electrolytes, sulfur species in the solution? One simulation model can involve solving several thousands simultaneous equations. It's just simply can not be done (by hand)! And I have never tried.

I still use Excel spreadsheet to size the K.O. drums, columns, scrubbers and strippers, though. Or do a little calculation on heat exchanger duties, heat transfer co-efficients...

Maybe I don't know what I am doing, but I crank out numbers every days. Is it a scary world out there? I use Aspen Plus, Hysys, Pro-II, Tasc, Acol if you know what I am talking about. I am going to try CFD (or any programs with funny names) if I can get my hands on it. In other words, I like all sorts of digital generating programs and I trust that people created them have good intentions, and if you know how to use them it will generate the desired results. Obviously I have a lot to learn.

-- A novice simulator

[Qalander (Chem)]

After reading all the expert opinions, I have to admit that I am in a category of engineers who almost blindly believe the simulation results. I think the technology has researched a point that, whether you like or not, you have to depend on the "garbage" coming out of the machine. Sometimes it is beyond my capability to question all the results from the cases I am dealing with.

For instance, I would create a similation model for a whole gasification plant, starting from gasifier, to syngas scrubbing, CO-shift, low temperature heat recovery, sour water stripping, Selexol unit, Claus sulfur recovery, hydrogen purification, methanation unit, etc, etc......

Each of the units I mentioned could have 10 heat exchangers, 2 - 3 distillation columns, 10 pumps and 10 K.O. drums, the totally could end up with 800 pieces of hardware, costing 4 billion $ to build. How can I come up with a stream table (or HnM Balance sheet) in a few month if I use a little TI calculator? How do I close the loops of recycle streams with the consideration of all the VLE and chemical reaction involves, Henry Law, electrolytes, sulfur species in the solution? One simulation model can involve solving several thousands simultaneous equations. It's just simply can not be done (by hand)! And I have never tried.

I still use Excel spreadsheet to size the K.O. drums, columns, scrubbers and strippers, though. Or do a little calculation on heat exchanger duties, heat transfer co-efficients...

Maybe I don't know what I am doing, but I crank out numbers every days. Is it a scary world out there? I use Aspen Plus, Hysys, Pro-II, Tasc, Acol if you know what I am talking about. I am going to try CFD (or any programs with funny names) if I can get my hands on it. In other words, I like all sorts of digital generating programs and I trust that people created them have good intentions, and if you know how to use them it will generate the desired results. Obviously I have a lot to learn.

-- A novice simulator

Dear PJ Hello/Good Morning
I must admire your vehement style of expressing the complex calculations and tedious time involving issues.
Regards
Qalander