Options
Calibration and Verification of a Mathematical Model for the Simulation of Blackwater/Biowaste Digestion
Citation Link: https://doi.org/10.15480/882.496
Publikationstyp
Master Thesis
Publikationsdatum
2005
Sprache
English
Author
Advisor
Title Granting Institution
Technische Universität Hamburg
Place of Title Granting Institution
Hamburg
Examination Date
2004-12
The object of this work is to apply and develop IWA anaerobic digestion model No.1 (ADM1) to the blackwater anaerobic digestion (BWAD) plant. The basic theory of anaerobic digestion (AD) processes and biochemical kinetics were introduced first. Afterwards the model was calibrated based on the performance of a lab-scale BWAD plant (at the mesophilic condition). The calculation includes three scenario studies, i.e. the reference conditions, the different feeding frequency and with high NH4+ input concentration. In order to verify the biochemical kinetics, the batch experiments were executed. According as the Michaelis-Menten kinetics, the maximum uptake rates (km) of butyrate, propionate, acetate are 18, 14, 13 d-1, and their half saturation concentrations (KS) are justified as 110, 120, 160 g COD/m3, respectively. The further two scenario studies were achieved based on the calibrated and verified model. First, the BWAD plant performance is predicted with different sludge retention time (SRT); second, the kitchen refuse (KR) was added into BWAD plant. The model successfully simulated these two scenarios and generated some suggestions for the operation of the real BWAD plant.
The model was discussed from the mathematical point of view subsequently. Disintegration and hydrolysis is not the rate-limiting step (at least not the sole step) for BWAD. They are much faster than the common mesophilic biowaste digestion. Containing disintegration and hydrolysis two steps make the model more flexible and controllable, although they are treated as one step normally. The influence from uptake processes of valerate and butyrate was checked. At least in BWAD, valerate and butyrate have very limited impact on the whole anaerobic digestion processes (ADP). Meanwhile, ADM1 uses the same degraders (i.e. the same uptake rate) to utilise these two acids. However, we suggested that these two acids are either excluded from the model (if they are not important), or included with individual uptake rates. Two methods for implementing acid-base processes were compared (equilibrium processes with differential-algebraic equation (DAE) and dynamic processes with differential equation (DE)). The same simulation results were obtained, which indicates that two methods can be arbitrarily chosen for all each acid-base. As to inhibition, different half inhibitory NH3 concentration had to be used in order to fit in with both the reference condition and high NH4+ input situation. This implies that the threshold of NH3 inhibition could be existent. The coefficients for physicochemical processes kLa and kp were tested by the model. Both of them are not sensitive to the model, so the determination experiments are unnecessary. In our model, kLa = 20 d-1 and kp =100 m3/(d•bar), respectively. It is justified that cations and anions influence pH strongly due to the charge balance, though they do not contribute either OH- or H+. The startup of model needs to be careful because of minus logarithm due to the improper initial conditions.
The model was discussed from the mathematical point of view subsequently. Disintegration and hydrolysis is not the rate-limiting step (at least not the sole step) for BWAD. They are much faster than the common mesophilic biowaste digestion. Containing disintegration and hydrolysis two steps make the model more flexible and controllable, although they are treated as one step normally. The influence from uptake processes of valerate and butyrate was checked. At least in BWAD, valerate and butyrate have very limited impact on the whole anaerobic digestion processes (ADP). Meanwhile, ADM1 uses the same degraders (i.e. the same uptake rate) to utilise these two acids. However, we suggested that these two acids are either excluded from the model (if they are not important), or included with individual uptake rates. Two methods for implementing acid-base processes were compared (equilibrium processes with differential-algebraic equation (DAE) and dynamic processes with differential equation (DE)). The same simulation results were obtained, which indicates that two methods can be arbitrarily chosen for all each acid-base. As to inhibition, different half inhibitory NH3 concentration had to be used in order to fit in with both the reference condition and high NH4+ input situation. This implies that the threshold of NH3 inhibition could be existent. The coefficients for physicochemical processes kLa and kp were tested by the model. Both of them are not sensitive to the model, so the determination experiments are unnecessary. In our model, kLa = 20 d-1 and kp =100 m3/(d•bar), respectively. It is justified that cations and anions influence pH strongly due to the charge balance, though they do not contribute either OH- or H+. The startup of model needs to be careful because of minus logarithm due to the improper initial conditions.
Schlagworte
ADM1
mathematical model
blackwater
kitchen refuse
ECOSAN
Loading...
Name
Calibra_Verifica_of_ADM1_by_BlackWater_Y.Feng.pdf
Size
1.38 MB
Format
Adobe PDF