Life Cycle Cost & Optimum Compression ratio Per Stage of Centrifugal Compressor Package 3

[Sirius P.Eng.]

We are currently engaged in a conceptual study for a Natural Gas Pipeline Gas Turbine Driven Centrifugal Compressor Facility.

The overall compression ratio is 2.614. Even though the discharge temperature is below the limits commonly imposed for per stage, I am considering 2 preliminary concepts which I have run in HYSYS with a polytropic efficiency of 70%:

1. Single-stage design (11.32 MW brake power)
2. 2-stage Design (10.73 MW brake power) - 1st Stage compression ratio 1.626, 2nd stage compression ratio is 1.625.

Obviously, the 2-stage design has the advantage of lower compression power - 600 kW less; I have estimated the fuel gas savings to be about US$6 million over the 25 years life of the facility. However, I expect the capital investment and yearly maintenance cost for the 2-stage design to be higher.

1. Can anyone share any correlations or simple formulas for:

- Estimating the total capital cost of the gas turbine driven centrifugal compressor (i.e gas compressor, gas turbine, interstage cooler, piping, etc.) -
for instance, in US$/MW or US$/BHP, etc.
- Estimating the yearly maintenance costs (parts, servicing, etc.)
- Other factors to be considered in the life cycle cost of the

2. For the 2-stage design, the maximum allowable pressure drop across the interstage cooler is 70 kPa. I understand that the most economical approach is to use equal compression ratios per stage. How can I relate mathematically, the relationship between the compression ratio per stage, suction pressure and discharge pressure and the pressure drop across the interstage cooler? - in my initial HYSYS model I used a trial-and-error approach.

 

[apetri]

you may wish to contact a vendor to obtain accurate values, while there are books and papers discusing these topics (with different details, to mention two, Centrifugal Compressors: A Basic Guide or Compressor Performance: Aerodynamics for the User) there are several aspects to consider, for example if you wish to derive a optimum multistage compression ratio the efficiency of the n stages of a compression might not be equal.
Manufacturers have access to different tools and sources as CFD modeling, extensive databases etc.
Said that, if you wish to optimize different parameters when simulating polytropic stages you can adopt tools which include a minimizer, I utilize Excel, Python o similar tools in union with PRODE Properties (which simulates polytropic stages also including phase equilibria),
you may be able to obtain similar results.

 

[rotw]

To my experience, if you have a motor driven compressor, you size the motor to fit the compressor.
When it is a gas turbine, it is most likely the reverse process, which means you size the compressor to fit the turbine.
Sometimes if the compressor is "too efficient" the turbine runs at partial load, and the efficiency gained in optimizing compressor is lost in the turbine running non optimal, let alone the emission problem and potentially accelerated turbine degradation.
You need to be careful about the max ambient temperature at the site of installation.
A compressor efficiency of 70% is low / too conservative unless you are considering a very low flow application. If the real figure is 80%, you end up with 10% power margin with the selected turbine which can be ok or not ok.
Also turbines are not only selected based on power rating only, it is a complex multi-criteria problem, for instance references and end-user desiderata are key. This can be a decisive factor in your project. If you are including an inter-stage cooler, you need to make sure you either have cooling water available at the site of installation or the site shall have favorable ambient conditions(otherwise it will result in extra increase of process cooler size).
Consider also that not only the design operating point effect the overall picture, if you have off-design cases you need to take this into account (unless they would occur marginally) and verify the corresponding power margins on the performance envelope of the turbine (speed vs. power) at max. ambient temperature.

Back to your main question, I suggest you build a "Net Present Value" model whereby you could consider the following factor:
- Inflation and "cost of capital" and run LCC over project lifetime
- Turbine fuel consumption at design and any alternative conditions
- Compressor capacity at design and any alternative conditions (if capacity is an output)
- Equipment cost of the turbine and compressor and auxiliary systems
- Capital spare part costs (turbine and compressor) and schedule for major overhaul to cast in your LCC

You need to approach OEM to get realistic figures.

If you plan an escape, you must succeed as if you fail, you will be punished for trying. Never say or write down your plan. Heart is the only place where secrecy is granted.

 

[georgeverghese]

Major cost item in this case will be the GT, not the compressor, regardless of whether it is 1 or 2 stage. Agreed, use polytropic eff of say 80%. Stage discharge temp at design case max operating flow should not exceed approx 140degC. The usual practice is to choose a GT with high run time between overhauls, a supplier who has rotor change out facilities not too far from where you are, and select a model preferably the same as some other GT in your company so you have commonality for spares. Avoid a speed reduction gearbox if you can for obvious reasons. And dont fall for these new GTs' with high thermal efficiency > 30% on LHV. Go for some old fashioned machine that is well known in the business with TE 25% or less.
These days, low NOx GTs' are the flavour of the day - but they are a pain to operate and NOx levels in actual operations dont meet expectations in the long run. Stay away from DI water injection at the GT air inlet if you can.