Aerated bioreactor installation factors

[mretier]

Dear BioSTEAM team,
I don't quite understand how the AeratedBioreactor class handles installation factors.

When I create one (say as R1) and check the installation factors (R1.F_BM), a dictionary is returned that doesn't include installation factors for compressors, coolers, etc.:

{'Horizontal pressure vessel': 3.05,
 'Vertical pressure vessel': 4.16,
 'Horizontal pressure vessel (jacketed)': 3.05,
 'Vertical pressure vessel (jacketed)': 4.16,
 'Platform and ladders': 1.0}

When checking R1.installed_cost / R1.purchase_cost a factor of 2.1 is returned. I briefly checked the source code but don't quite get how installed cost is actually calculated. Could you elaborate?

Thanks a lot!

Best,
M

[sarangbhagwat]

Hi @mretier,

The installed cost for each component of a unit in BioSTEAM is calculated based on the baseline purchase cost and bare module, design, pressure, and material factors (so there are other factors besides bare module). The purchase/installed cost of a unit is the sum of component purchase/installed costs. This estimation method is detailed in the parent Unit class (under _load_costs) rather than in individual unit classes as they share this method.

When a bare module factor dictionary F_BM does not include factors for components that do appear in baseline_purchase_costs, a bare module factor of 1 is used for those components when estimating installed cost. Note these could still have values provided for the other factors F_D, F_P, or F_M.

If you have any more questions, please feel free to comment and reopen this issue (closing for now)!

[yoelcortes]

@mretier, check the bare module factors of the auxiliary unit operations, which are stored as attributes. BioSTEAM automatically adds the costs of these special attributes to the parent unit. For example, <AeratedBioreactor>.compressor._F_BM

So BioSTEAM does account for the bare module factors of all auxiliary components, but they are costed separately by the auxiliary components themselves.

Please feel free to reopen if you have any questions.

Happy to help!

[mretier]

@sarangbhagwat, @yoelcortes, thanks for the prompt response.
I assumed installation factors would be available under <AeratedBioreactor>.F_BM for all components. Maybe it would be worth considering collecting all factors there since <AeratedBioreactor>.results(), <AeratedBioreactor>.installed_costs and <AeratedBioreactor>.purchase_costs make data available for all components?

One more point: I just calculated the installation factors for all components [(y, R1.installed_costs[y] / R1.purchase_costs[y]) for y in R1.installed_costs] which returns:

[('Vertical pressure vessel', 2.858823529411765),
 ('Platform and ladders', 1.0),
 ('Compressor - Compressor(s)', 2.15),
 ('Air cooler - Floating head', 3.2021141764681116),
 ('Heat exchanger - Floating head', 3.195635281184021),
 ('Recirculation pump - Pump', 1.2584269662921346),
 ('Recirculation pump - Motor', 3.3),
 ('Agitator - Agitator', 1.0)]

The calculated values for Vertical pressure vessel and Recirculation pump - Pump are off by quite a bit to the installation factors:

> R1.F_BM
{'Horizontal pressure vessel': 3.05,
 'Vertical pressure vessel': 4.16,
 'Horizontal pressure vessel (jacketed)': 3.05,
 'Vertical pressure vessel (jacketed)': 4.16,
 'Platform and ladders': 1.0}


> R1.recirculation_pump.F_BM
{'Pump': 3.3, 'Motor': 3.3}

Can you explain what causes this discrepancy?
Thanks a lot for helping me understand this better!

PS: I don't have permission to reopen this issue.

[yoelcortes]

@mretier,

I would be open to adding a "convenience" method to output all bare module factors as a table. Internally, however, there is no need to compile all bare module factors into the parent dictionary. Best to keep it simple for computational efficiency and consistency with how things are calculated in BioSTEAM.

Your estimation of the installation/bare-module factor is actually quite off. The bare module factor only applies to "baseline" design, pressure, and material assumptions (all of which impact the purchase cost). See the notes in https://biosteam.readthedocs.io/en/latest/API/Unit.html

Think about how changing the material of a vessel from carbon steel to stainless steel could impact the costs. It would be wrong to assume it will proportionally impact the installed equipment cost (of which labor is biggest contributor). Instead, it only impacts the purchase cost and the remaining capital costs remains the same.

If you would like to test biosteam's calculation, please use the correct formula, using the baseline purchase cost and all purchase cost factors (e.g., for a pump would be F_M, F_D, and F_P) and accounting for the number of units in parallel. Additional tests are always welcome :)

Thanks,