17 replies to this topic

[apis]

Hi everyone.

I have a question here.Can we specify the part in HYSYS that we want HYSYS do the calculation for us?As an example, if i face with designation of distillation column,can i specify or give an order to HYSYS to calculate the number of stages for my column design?Can anybody show me how to do it?I really2 need your help....

[Zauberberg]

Hello Apis,

You can use the "Shortcut distillation" template in Hysys unit operation palette. You will have to specify:

- % of light key component in bottoms product (based on your product spec.)
- % of heavy key component in tower overhead stream
- External reflux ratio: the ratio that is reasonably to achieve when taking into account environmental conditions (= cooling utility temperature, derived from CAPEX vs OPEX etc.)

This is only the starting point for you. Hysys is just a tool for calculations, it can never perform a design job instead of you.

Best wishes,

[apis]

Hello Zauberberg,

Thank you very2 much for your response and very precious information.But i have a little bit confuse about the things that we need specify when we want to use shortcut distillation.If you don't mind, can you explain to me about the external reflux ratio and what do you mean by CAPEX vs OPEX?I'm sorry Zauberberg if i'm asking the easy question to you because i'm very the new one in this simulation field:-).Or can you tell me the other source that i can looking at to give better understanding to me in this problems?I hope you can help me because i really2 interested in learning about this simulation.A thousand thank you to you first because of your willing to help me.

[abhi_agrawa]

apis,

I believe that CAPEX would be CAPital EXpenditure and by OPEX, OPerating EXpenditure. The optimization meant by Zauberberg would then be (as compared to some base design):
- If you decrease the size of the column (by decreasing the number of trays), you would require more energy energy for separation, your reflux and reboiling requirement would go up. Thus you would have reduced the CAPEX by making the column smaller, the OPEX is higher as you need more reflux and reboiling.
- Similarly if you increase the number of trays, you would need less reflux. So your CAPEX goes up but the OPEX is lower.

Also keep in mind that there will be some minimum number of trays that you would need for the required separation. This would be at infinite reflux and also a minimum reflux requirement at infinite number of trays.

Hope this helps,
abhishek

[Padmakar Katre]

Dear,
I will give the procedure as I promised,

1.Open a new case.Select your components and the fluid package.
2.Define your feed stream.(It should be flashed-i.e. give the required inputs so that Hysys will perform Flash Calculations)
3.From object pellate select the Short-Cut Column.
4.By clicking on the column object you will go the column property view.
5.In column property view in Design Tab-
Connections Tab-
Make the conections i.e.
Energy streams for Reboiler and Condesner
Material Streams to distillate and bottom products
Top product phase-
Parameters-
Give your required product specifications/purities
Mention your reboiler and condenser pressures
External Reflux Ratio

Your simulation will be converged and you will be at the results and to check that go to the performance page and see the results.

Now if you have to vary the reflux ratio and to check its effect on the number of trays and the reboiler and condesner duties or any other dependent variables you can perform a case study

Just go to Tools--Databook and do your case study by giving the range of your reflux ratios and check its effect on your key variables.

But keep one thing in mind it will give a rough idea.You should go for the other softwares to perform your distillation calculations as Mr Zauberberg said even I don't rely on the Hysys results (this is my personal Opinion-I am sorry if some one minds this).
If still you find any query please let me know.

[apis]

Hi padmakar,

Thank you very2 much for the information and steps that you and abishek had give to me. From your steps,now i already can do my shortcut distillation column.I already had done a short cut distillation column for my problem before.But, the problems that i faced here is i had a different result from my friends though all the specs that they all have made are same with me.I had followed all of your and Zauberberg's steps.Can you help to guide me in this learning process padmakar?My second question is, i had choose NRTL as my fluid package.My components types are carboxylic acids.But when i want to enter the simulation environment, the state 'some of the binary coefficient for this fluid package's components are unknown.They will be set to zero'.Can i know what does it mean by that statement and what is binary coefficient means?Can i accept the assumption by HYSYS to assume that binary coefficient will be set to zero?

[Zauberberg]

Hello Apis,

Binary interaction parameters purpose is to quantify the non-ideality of a given system, and express it in a form of numbers. They are absolutely necessary for systems such is yours, and the information from Hysys that they have been set to zero means that - in simple words - Hysys doesn't have these interaction pairs in its library.

I hope that we'll have additional feedback from Joerd and receive some further clarifications regarding this very important issue. What I can tell you for sure, is that you must not rely on the simulation results without knowing interaction parameters, especially for this extremely non-ideal case. Interaction parameters are usually obtained in a laboratory, or you can ask for support from software vendor, or look for more information in similar industries.

Best of luck,

[Padmakar Katre]

Hi padmakar,

Thank you very2 much for the information and steps that you and abishek had give to me. From your steps,now i already can do my shortcut distillation column.I already had done a short cut distillation column for my problem before.But, the problems that i faced here is i had a different result from my friends though all the specs that they all have made are same with me.I had followed all of your and Zauberberg's steps.Can you help to guide me in this learning process padmakar?My second question is, i had choose NRTL as my fluid package.My components types are carboxylic acids.But when i want to enter the simulation environment, the state 'some of the binary coefficient for this fluid package's components are unknown.They will be set to zero'.Can i know what does it mean by that statement and what is binary coefficient means?Can i accept the assumption by HYSYS to assume that binary coefficient will be set to zero?

Dear,
If you are having the carboxylic acids components then instead to use only activity models like NRTL,you do one thing use NRTL or Uniquac for liquid phase and and viral equation of state for vapor phase.I will justify my reply, If you are using the built-in binary parameters, the ideal gas model should be used. All activity models, with the exception of the Wilson equation, can automatically calculate three phases given the correct set of energy parameters. The vapour pressures used in the calculation of the standard state fugacity are based on the pure component coefficients in HYSYS' library using the modified form of the Antoine equation.
When your selected components exhibit dimerization in the vapour phase, the Virial option should be selected as the vapour phase model. HYSYS contains fitted parameters for many carboxylic acids, and can estimate values from pure component properties if the necessary parameters are not available.

One another suggestion is from the simualtion basis manager page i.e. beforw swithing to simulation envirnment follow steps,

1.Go to fluid package tab
2.Select the Fluid package selected here mainly your activity model and click view button.
3.Then go in the binary coeff tab and check for the presence of the BIPs if not present calculate by the Unifac method of estimation with ref to certain temp else you just click on estimate unknowns only.
4.Once you get all the BIPs then enter into main simulation envirnment and do simulation and analyse the results.

Even If now you find any problem/doubt please ask us.I hope Mr Jored, JoeWong and so many Jedies will help you to reach to the solution.

[apis]

Hi padmakar and zaberberg,

I already did the steps that you had told me before.Thanks a lot for the brief explaination.After i've done the steps that you had told me before,and after i compared with my friends short cut distillation column, i had found that the slightly differences between our shortcut distillation column are at reboiler and condenser duty.Although we have just slightly differences on that matter, but that's become a very2 big problems to me.So,what is confusing me now,is there anything steps that maybe i didn't put or maybe any specs that i didn't consider in doing this short cut distillation column?If you don't mind,can you check my shortcut distillation that i've done?If you can,i'll send to you or post my HYSYS file in this forum.I hope you can help and guide me in this learning matter.Thanks in advance and I'll very appreciate it.

[joerd]

I fail to see why the slight differences with your friend's simulation bother you so much. The only thing to figure out why you have a difference is to meticulously go through his simulation and your own, and see where you have something different.
Although this may give you some insight on the sensitivity of your problem to the input parameters, I can hardly believe that anyone here would have the time to do it for you.

[apis]

I fail to see why the slight differences with your friend's simulation bother you so much. The only thing to figure out why you have a difference is to meticulously go through his simulation and your own, and see where you have something different.
Although this may give you some insight on the sensitivity of your problem to the input parameters, I can hardly believe that anyone here would have the time to do it for you.

Hi joerd,

Thank you for your response.Actually i'm the new one in this simulation field.So,i do not have much knowledge and information about this simulation.But, refer to the steps that you had said,"meticulously go through his simulation and your own, and see where you have something different.",i've already done it before and i've also tried to do the same cases several time, but the result is still the same;I still have differences between his simulation.I've check severel time everything that i know in his and my own simulation.What is confusing me now is i don't know either i already had put enough information in that simulation or maybe i had missing any consideration or something else in my simulation.That's why this slightly differences between my friends and my simulation become a big problem to me because i don't know which one is the true. The differences only on reboiler and condenser duty.The number of stages,reboiler and condenser temperature,percentage of product composition all give me the same result with my friends.That's make me very confuse and weird.Please help me joerd.I'm really2 need somebody to help me in this simulation learning process.Please guide and advice me if anything i've need to do.Thanks in advance.

[joerd]

It is nearly impossible to tell what could be different between your simulation and your friend's - and if it's important - without the actual simulation files, or at least the print-out of the column data. So, if you could post those (yours as well as your friend's) we might be able to tell you where the difference comes from, and you might learn something in the process...

[DarrenO]

One way you can use HYSYS to help you find the differences between the two cases is by storing out the XML format of both cases and then doing a difference between the two files using something like Windiff. It would probably be a good starting point.

[apis]

It is nearly impossible to tell what could be different between your simulation and your friend's - and if it's important - without the actual simulation files, or at least the print-out of the column data. So, if you could post those (yours as well as your friend's) we might be able to tell you where the difference comes from, and you might learn something in the process...

Hi joerd,

Thank you very much for your response.Here i post both my friend's and my short cut simulation HYSYS file.When you open this HYSYS files,and click on short cut distillation tower,at performance tab,you can see slightly differences only on condenser and reboiler duty.This "slightly differences" is the things that confusing me although we had use all the same things;the specs,fluid packages,temperature,pressure,etc..Please help me joerd.I'am very appreciate because of your willingness to help me.Again,thanks in advance.I'll wait for your precious response.Thank you very2 much.

Attached Files

•  mine.zip   14.76KB   99 downloads
•  my_friend.zip   15.04KB   77 downloads

[apis]

One way you can use HYSYS to help you find the differences between the two cases is by storing out the XML format of both cases and then doing a difference between the two files using something like Windiff. It would probably be a good starting point.

Hi DarrenO,

Thank you for your response.I'm very interested to know the method that you had said before.But, because i'm the new learner in this field,i have a little bit difficult to understand what you had said.If you don't mind,can you explain to me more brief about this method and maybe several steps that i need to take in doing this method?I really2 need to know and try it DarrenO.I'll wait for your precious explaination.Thanks in advance.

[joerd]

Apis,

The method from DarrenO refers to exporting the simulation to a text file or XML file via menu Simulation / HYSYS XML. Then select Export Case as XML, or select the Input Summary Format checkbox and copy the text file. There are a number of tools available for download to find the differences in text files. You can also paste them side-by-side in an Excel sheet and let Excel do the comparison.

I looked at your simulation files, and can assure you that for all practical purposes there is no difference between them. The reboiler/condenser duty differs only by 3/1000 of a percent, which is within the numerical error margins for Hysys. If you add a shortcut column to your friend's simulation, and define the feed stream equal to his feed stream, and put in all the specifications, you end up with your results. So, it looks like the difference has to do with the internal setup of the case inside Hysys, and the solution path that Hysys takes. Too bad that you run in to this when you're still not very confident about your simulation skills, however, it is a good learning opportunity: 1) simulation programs are numerical solvers, so the accuracy is limited to the accuracy of the input, of the solution method, of the approximations inherent in the thermodynamics package. 2) in the end, you'll want to design a working chemical plant. The equipment that you will design will not be different if you have a tiny simulation error/imbalance. 3) you are very right to be cautious about the results, and question everything that doesn't make sense. If you keep this attitude, you'll produce better designs because you will spot errors that do make a difference on the final design as well.

Keep up the good work, and don't fret about this simulation.

[apis]

Hi joerd,

Thank you very much because of your briefly explanation about the short cut distillation.From your help,now i know where is my mistake.A thousand thank you to you.Joerd, i have another question here.Now,i already had my short cut distillation column.Now my next learning steps is to change my shortcut distillation to the rigorous distillation column.I had tried to do it by try and error with the data that i get from my shortcut distillation,but it's still unconverged.So,if i'm lucky one, can you tell me how we want to substitute this short cut distillation to the rigorous one.What is the specs or input(or data from short cut distillation) that we need to put in the rigorous distillation column?What is the things that we need to consider in substituting this things?I hope you can help me again joerd.Thanks in advance.

[joerd]

In your shortcut column you have already specified a column pressure, and the results give you a minimum number of theoretical stages. The shortcut column allows you to specify a number of theoretical stages, which is typically 1.2 - 1.3 times the minimum.
When you set up the rigorous model, you have to specify the pressure and the number of stages.
Then, there are only two more degrees of freedom: if you specify top and bottoms product purity (like you have in the shortcut distillation column), the condenser and reboiler duty are calculated, and vice versa.
Typically, you will want to specify the required product purity, and let Hysys calculate the duties. However, this does not always converge. So, you may specify duties first, and try to get the column to converge, then modify the duty by hand until you get close to your spec, and then switch to the product spec.
Alternatively, you can specify overhead or bottoms product rate, and one duty, and try to get close to your real specifications before you switch these on.
A very good method is to take the results of the shortcut column profile, and use these as the estimate for the column temperature, vapor/liquid rates, and compositions.

Sorry for the unstructured posting, it's still Monday morning here, but I hope it will help you solve your problem.

UPDATE: I took a quick look at it. It is a difficult separation, so I had to switch to Simultaneous Correction solver to get convergence, then could switch back to Hysim IO. I first set condenser and reboiler duty, then backed into the purity specs. See attached file.

Attached Files

•  mine.zip   208.1KB   101 downloads