Open Collections

Modelling carbon oxidation in pulp mill activated sludge systems : calibration of ASM3

University of British Columbia

Activated Sludge Model N°3 (ASM3) was chosen as a basis to model an activated sludge system treating effluents from a mechanical pulp and paper mill. The wastewater treatment plant (WWTP) selected, located in Port Alberni, consists of a pH adjuster, primary clarifier (not modeled), five complete mixed bioreactors in series, and a secondary clarifier. The plant treats an average of 74,000 m³ d⁻¹, with the following average influent characteristics (primary effluent): COD of 594 g m⁻³, BOD of 250 g m⁻³, TSS of 27 g m⁻³, pH of 7.2, and temperature of 32°C. A simplified version of the ASM3 model (excluding nitrification and anoxic conditions) was successfully calibrated for this plant. A single set of parameters, with minor variations, was able to fit a variety of batch test respirographs, and performed relatively well when a full-scale simulation was performed using the calibrated parameters values over a time period of nine days. Three measurement campaigns were undertaken in order to calibrate the model properly. The most sensitive parameters (Y[sub STO], Y[sub H], b[sub H], k[sub STO], μ[sub H], K[sub STO], and k[sub H]), as well as the wastewater influent COD fractions (S[sub I], S[sub S], X[sub I], and X[sub S]), were evaluated using three complementary tools: batch respirometric tests, analytical measurements, and mass balance equations. Even though the authors of ASM3 advises not to apply that model for industrial wastewaters and outside the temperature range of 8 - 23°C, the model was found suitable for modelling the COD removal in a WWTP treating pulp and paper effluents at temperatures around 32°C. The importance of this finding is that the application of widely used models, developed originally for municipal wastewater, in pulp and paper applications would simplify considerably the task of modelling and designing WWTP for this industry. Some assumptions of the model, however, proved not to be applicable and some adjustment would be necessary in building a dynamic model for the treatment of these effluents. The least sensitive parameters were not calibrated, so they were either assumed from the literature (f[sub SI], f[sub XI], K[sub X], and K[sub S]) or assumed to be equal to other estimated parameters (b[sub STO] and θ[sub T,bsto], assumed to be equal to b[sub H] and θ[sub T,bH] respectively). Sensitivity and structural identifiability analysis were also performed of the ASM3 simplified model. Only the calibration of the heterotrophic, growth rate (μ[sub H]) presented some identifiability problems, which were solved by estimating an auxiliary parameter (μ[sub OBS]). A few novel calibration procedures were developed for estimating some of the parameters (b[sub H], θ[sub T,bH]), which could be used for any type of effluents. In addition, different methods were used for the estimation of many of the model parameters and wastewater fractions, and some recommendations were done in order to select the best method for estimating different model components.

Thesis/Dissertation

application/pdf

2009-11-02

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/14560

Civil Engineering

University of British Columbia

UBCV

DSpace