14 replies to this topic

[Amin Mehdipoor]

Hi everyone,

 

I am trying to use the Matrix method in order to rearrange a heat exchanger network in a plant. to do this, I need to have an estimation of retrofit cost and it calls for Heat exchanger cost, piping cost and pressure drop cost. Is there any clear formula to calculate the heat exchanger cost (either new one or expanding an existing one). what about piping cost? and also the pressure drop cost (either in heat exchanger or in piping).

 

I have some equations which come from 90s and I am not sure if they are still valid or not. Moreover, that show significant difference in compare to Coulson and Richardson's series (regarding fixed cost).

 

Is there any reference that I can rely on? 

 

I really appreciate if you could help me.

 

Thanks.

[Shivshankar]

Hi,

 

Check this link, if this can help you.

 

http://www.matche.co...pCost/index.htm

 

Regards

Shivshankar

[Amin Mehdipoor]

Thanks Shivshankar,

 

I've seen this website before. but it doesn't have any reference. based on the currency and the year, I guess it uses Coulson and Richardson's series. but I am not sure.

 

By the way, thanks

[ankur2061]

amin,

 

Check out this link if it can help:

 

http://www.cheresour...ost-estimation/

 

 

Regards,

Ankur.

[Amin Mehdipoor]

Thanks Ankur,

 

I saw your post couple of days ago. It is a great post which helps a lot but the problem is my current equations are far from you proposed! I am using below:

 

Annual heat exchanger area cost= annuity factor * (fixed cost of heat excahnger + b*(Aarea ^ n))

 

in above equation, fixed cost is assumed 40000 (USD), b=388 or 400 and n=1

 

 have no idea if these values are still valid or not (similar situation for piping and pressure drop cost calculations!). I am kind of restricted to use above equation! and that's the reason that I cannot go through the post you proposed. I need a sort of confirmation of what I have now.

 

Thanks 

[pavanayi]

Amin,

Are you a student doing a project as part of your degree?

 

Determining reliable costs for retrofit scenario is harder than that for a new build.  This is because, as a result of a retrofit study,

 

1. A few exchangers will need to be replaced with new ones. The new exchanger might fit on the foundation of the earlier one. Or it might need to be demolished and built again. Either way, you have to add/subtract the additional work.

2. Few modifications might only be repiping the exchangers to change position. The costs associated here are different to a new exchanger.

3. If you want additional exchanger area, the piping costs as calculated conventionally will not be valid because of its inherent assumptions.

 

Each cost estimation tool has some inherent assumptions. One common basis for heat exchanger costing is a plot area of 15mx15m. The cost result is for buying a new exchanger, preparing a foundation and installing it, and running pipes to and from it till the boundaries of this plot area. Some estimates might include basic valves and fittings as well. So, as a result, the accuracy of any such estimates will depend on the assumptions.

 

Have you had a look at paper titled Cost estimation and energy price forecasts for economic evaluation of retrofit projects by Taal, Bulatov, Klemes and Stehlik?

[Amin Mehdipoor]

Dear Pavanayi

 

Thanks for your explanation. Actually I finished my studies and as a duty of my new job, I am trying to figure these out. 

 

The paper that you refer me to, is a great one that I've read and I will use it to see if my results are reasonable or not. I wanted to check if there is any other valid reference or not. As you said, in case of retrofit, it's really complicated and one must consider several assumptions!

[pavanayi]

Amin,

 

If this is a professional scenario, the whole picture changes even before you begin.

 

In your quest, have you come across any real life case study of a successful implementation / application of 'Matrix method to rearrange a heat exchanger network' in a plant?

 

Is this study being performed as part of a feasibility study? Could you include additional details as to the scope of the work and the scope of your involvement? Also, what costing tools you have in your company? Like Aspen Process Economic Analyser?

[Amin Mehdipoor]

Pavanayi:

 

It's not a real life case. it's a feasibility study of Matrix method's capability. So I can add some detail to the work's scope. but not that much. I'm not going to produce a detailed estimation of cost, so not using any advanced tool. just a rough estimation.

 

Do these questions and answers affect the accuracy of the results? or what?

 

Thanks

[pavanayi]

The questions were to enable me to understand how serious was this study being performed, and what tools were at your disposal.

 

Secondly, any study like matrix method or any other heat exchanger network design philosophy will give you different options to choose from. The benefits of each method is usually an economic analysis of return on investment. When your estimated cost itself is ±50% (or even worse off), how reasonably can one conclude one configuration is better than another? This becomes more difficult with a retrofit scenario, when you can only get an order of magnitude cost.

 

Will this be alright as an academic exercise? Probably yes.

Will any engineer give a serious thought about the results of such a study? Probably not.

Thats why I asked about the scope of the work and your involvement in it.

[Amin Mehdipoor]

Pavanayi:

 

Thanks for explanation, now I see your point.

 

The thing is, in this stage, nobody expects me to produce the most possible accurate answer (or cost evaluation). But rather to find out how Matrix can be used in a commercial way. So I am supposed to fit myself in current scope of the work and investigate whether calculations or right or something is missed. Accordingly, I asked if someone know an up to date and valid reference which can be used as a proof of concept. 

 

Now, knowing the situation, do you have any suggestion? 

[pavanayi]

No, unfortunately I dont have any suggestion because, however you do it (accurately, roughly), you will eventually find that neither Matrix method nor any other heat exchanger network synthesis method to date is robust enough to cope with the constraints presented by a real retrofit scenario by itself.

 

But my advice would be to go ahead and finish the attempt as it would give you valuable insights into

1. Heat exchanger network synthesis and the advantages/disadvantages they have

2. Retrofit and its challenges

3. Application of process integration principles to a real plant.

 

This will for sure come handy in future.

[Amin Mehdipoor]

Pavanayi: 

 

Thanks a lot for your useful advices. I hope we find a way to make Matrix useful for real plant situation.

 

cheers

[awool]

Hi there, I have done something similar recently.

I found chemical engineering design, Sinnot 2009, to be very useful for the following

 

all capital costs follow the equation A+ B(S^n), where a b and n are specific constants and S is dependant on the machinery. if you are designing a simple heat exchanger, the capital cost should also be in there too. if you are doing a plate fin heat exchanger, as i am, it is more complicated.

 

compression cost is purely operational; unless you need to recoupe the pressure through new compressor.

calculate your power of the compressor (kW) and then multiply by the number of hours of operation for a time period for example

 

100kWx360days/yearx24hours/day=864MWh/year

 

electiricity is priced at kWh. operational can be calcd from there.

 

hope this helps

[Amin Mehdipoor]

Hi there, I have done something similar recently.

I found chemical engineering design, Sinnot 2009, to be very useful for the following

 

all capital costs follow the equation A+ B(S^n), where a b and n are specific constants and S is dependant on the machinery. if you are designing a simple heat exchanger, the capital cost should also be in there too. if you are doing a plate fin heat exchanger, as i am, it is more complicated.

 

compression cost is purely operational; unless you need to recoupe the pressure through new compressor.

calculate your power of the compressor (kW) and then multiply by the number of hours of operation for a time period for example

 

100kWx360days/yearx24hours/day=864MWh/year

 

electiricity is priced at kWh. operational can be calcd from there.

 

hope this helps

 

Hi Awool,

 

thanks for your help. I was also thinking about Sinnot 5th edition. but as I mentioned, there is a huge difference between what I currently have and what is in Sinnot. Anyway thanks for your reply.