Hi,

I am simulating coal fired utility boiler. summary of cas: E-L, species transport, RNG-k-epsilon, P-1, tangentially fired,all urfs are 1 energy, dpm, species,

I am not able to achieve heat balance for this case.

The fluent report 580 MW energy at the outlet boundary where the temperature is only 670K. The inlet boundary shows -ve values of convective energy. The error is +16%. i.e. that much excess heat is generated in reactions. I have converged case for many iteration but still no improvement in heat balance.

Can any one can comment on this.

Thanks

DG

What about DPM enthalpy source?

Hi Allan,

Thank you for your response.

I have checked the sum of DPM enthalpy source(ES)= -122MW. Also the DPM sensible enthalpy source (SES)= +50 MW

Does this means that this much energy is added into gas phase by char combustion.

I have c+ 1/2 O2 >> CO delH= -9.81e6 J/Kg. The flow rate is 9.2 kg/s. hence the energy generated will be 90.25MW.

But ES value should have been positive as heat is added from dpm to continuous phase and that what Fluent manual says.

Thank you DG

It is difficult to say without seeing all of the terms (and oftentimes difficult even when you have them all) where the problem lies. In my experience, once the outlet temperature stops changing with interation and the wall flux terms even out, then it has converged. The energy balance always seems to make sense in the end and the simulations usually balance within 5%.

The DPM sensible enthalpy should include any latent heat for drying moisture in coal and heat of combustion for carbon and sensible enthalpy for heating particles. The DPM sensible enthalpy is included in the DPM enthalpy term. The DPM enthalpy source will include the heat of formation of any volatiles and the heat of formation of CO as a product.

The heats of formation often seems to skew things in terms of looking at input and output enthalpies, especially since the sign convention seems to go the opposite way. Fluent 12 promises to have a sensible enthalpy balance which would be easier to interpret.

Dear Allan,

The signs of input and output enthalpies are not only opposite but the magnitude is far different than actual calculations. This I have checked when there is very negligible change in heat transfer to wall and ht exch.

The heat product due to coal combustion will be 560 MW at 298K . The heat generated will be lowered at reaction temperature (~520MW). I am specifying the super heaters by using porous volume and estimating the heat transferred to these exchangers by using define source udf. The UDMI is defined for eacg heat exchagner and the value of heat transferred is was taken by using volume integral of the source term written in udf.

The heat transferred to water wall is 166 MW. The total ht to heat exchanger is 389 MW. So total heat transfer is around 555 MW. Where as, this should have been around 460MW.

Does the approach of estimating heat transfer through porous volume by using volume integral works well in fluent?

I had checked this in a case where hot gas was introduce and only heat transferred was observed to all heat exchangers in the same boiler. The heat balance was in the range of 0.05% error.

So may be it is over predicting heat generation. If we observe the heat transfer heat equation (fluent manual 6.3, equation 13.2-6) for non premixed case, then the scaling of heat generated due to reaction which should be done at reaction temperature could not be observed. So is this leading to a error in heat balance.

Pl give your comment.

Thanks

DG

A few comments. How do you calculate "The total ht to heat exchanger is 389 MW"? Is this from Fluent? The value for the sum of the porous cells?

I have used porous cells for the superheater and specified a volumetric heat source (sink) several times and it works okay. In Fluent 6, I don't use a UDF for this - rather just use the input panel to specify that there is a heat source in the porous cells and then specify a constant source. The 389 MW, of course, is much to large of fraction for just the superheaters assuming a boiler with conventional superheaters, generating bank, economizer, etc.

The inlet and outlet enthalpies in Fluent sometimes look unusual but have always worked out in my boiler simulations. For example, for grate fired units like waste-to-energy boilers, I have specfied the fuel as water, volatile species, and char fractions coming up through the grate. The values that Fluent calculates for the enthalpy fluxes are not intutive due to the differences in latent heat of formation for the various species.

Dear Allan,

The approach is same but I had done periodic heat transfer studies to estimate the Nu around tube banks. Based on that heat sink was estimated in udf with tube wall as const temp. So I don't specify const. heat sink but estimate the same.

Can you comment on my observations: 1) " If we observe the heat transfer equation (fluent manual 6.3, equation 13.2-6) for non premixed case, then the scaling of heat generated due to reaction which should be done at reaction temperature could not be observed in equation 13.2-6 "

2) Also, for the heat of char combustion, what is the Hreac value you specify? I mean, what I want to know is, I am modeling char combustion as c + 0.5 O2 >> CO delH= -9.81e6 J/Kg@ 298K. so same value I specify for Hreac. Is it correct?

It sounds as if you have solved your larger issues with the overall heat balance as now you are looking at the details.

Specifically: 1) I think I see what you are getting at in terms of looking at the sensible enthalpy of the reactants when cooled from the reaction temperature, the heat of reaction, and then heating the products back up to the reaction temperature. I don't know how Fluent actually handles it, but that would not be my biggest concern due to all of the other unknowns in modeling the coal combustion.

2) Yes the heat of reaction for carbon -> CO should be something like 9.81 MJ/kg of carbon. (My reference says 9.21 MJ/kg). How do you handle the reactions for the CO generated?

Dear Allan,

No, still I am striving hard to get the heat balance of my boiler.

Point (1) is important when we look for heat balance. The reaction takes place over a range of temp.s and hence the amount of heat generated keeps on varying. I have checked it by hand calculations, which shows that if reaction occurs at 1400K then the overall heat generated decreases by 6% to that of at standard state.

so I started looking into the energy equation into more details.

about point (2) Pl. refer equation 22.9-86 in fluent 6.3 manual, then the fluent does not scale Hreac value to reaction temperature.

I think there is problem. Actual heat generated will be less than std. state. Hence, we will get extra heat generated and same error will be replicated in heat balance.

My reference for DelH is Baum and Street (1970). Kindly let me know details of your reference.

I am handling CO generated by Finite chemistry/EDC model. so CO + 0.5 O2 --> CO2 delH@298K = -283 MJ/Kmol and plus one more gas phase volatile combustion reaction.

I am running my case further, I will let you know advances in my heat balance. But if you can provide me more checks to do the same then it will be great help for me.

Thank you

DG

how heat balance calculation for boiler make fuel baggase? please your answer to my email.. thanks..