Modelling of Wastewater Treatment Plants

Transcription

1 Modelling of Wastewater Treatment Plants Nevenka Martinello

2 Why do we need WWTP models? to build a WWTP model CASE STUDY - WWTP model in Sweden

3 Why do we need WWTP models? Rise awareness Plant design Process control

4 WWTP models Deterministic models Static vs dynamic models

5 to build a WWTP model Purpose definition Definition of the WWTP model purpose or the objectives of the model application (learn, design, control) Model selection Choice of the models needed to describe the different WWTP units to be considered in the simulation, i.e. selection of the activated sludge model, the sedimentation model, etc. Hydraulics determination Determination of the hydraulic models for the WWTP or WWTP tanks Flows quality characterization Wastewater and biomass characterization, including biomass sedimentation characteristics Calibration Calibration of the activated sludge model parameters Simulations Scenario evaluations

6 Activated Sludge Model 1 Activated Sludge Model 1, 2, 3 were developed by the task group formed from the International Association on Water Quality (IAWQ, formerly IAWPRC). The focus will be on the Activated Sludge Model 1 (ASM1), which is a theoretical mathematical model depicting the biological processes occurring in the activated sludge section of a wastewater treatment plant. It describes carbon oxidation, nitrification and denitrification. Activated sludge model Settling model

7 Activated Sludge Model 1 Model state variables: COD and Nitrogen fractionation Activated sludge model Settling model

8 Activated Sludge Model 1 Model processes Activated sludge model Settling model

9 Activated Sludge Model 1

10 Activated Sludge Model 1 Model Parameters The stoichiometric parameters are: Heterotrophic yield, (g X BH COD formed (g COD utilised ) -1 ) Autotrophic yield, (g X BA COD formed (g N utilised ) -1 ) Fraction of biomass yielding particulate products, (dimensionless) Mass N/mass COD in biomass, (g N (g COD) -1 in biomass (X BA and X BH )) Mass N/mass COD in products from biomass (g N (g COD) -1 in X P ) The kinetic parameters are: Heterotrophic maximum specific growth rate, (day -1 ) Heterotrophic decay rate, (day -1 ) Half-saturation coefficient for heterotrophs, (g COD m -3 ) Oxygen half-saturation coefficient for heterotrophs, (g O 2 m -3 ) Nitrate half-saturation coefficient for denitrifying heterotrophs, (g NO 3 -Nm -3 ) Autotrophic maximum specific growth rate, (day -1 ) Autotrophic decay rate, (day -1 ) Oxygen half-saturation coefficient for autotrophs, (g O 2 m -3 ) Ammonia half-saturation coefficient for autotrophs, (g NH 3 -N m -3 ) Correction factor for anoxic growth of heterotrophs, (dimensionless) Ammonification rate, (m 3 (g COD day) -1 ) Maximum specific hydrolysis rate, (g X S (g X BH COD day) -1 ) Half-saturation coeff. for hydrolysis of slowly biodegradable substrate, (g X S (g X BH COD) -1 ) Correction factor for anoxic hydrolysis, (dimensionless) Y H Y A f P i XB i XP μ H b H K S K OH K NO μ A b A K OA K NH η g k a k h K X η h Activated sludge model Settling model

11 One-dimensional secondary settling tank model It reproduces the clarification-thickening processes and describes the solids profile throughout the settling column, including the underflow and effluent suspended solids concentrations (Takács, 1991). It is a one-dimensional model, where only processes on the vertical dimension are described, whereas horizontal solids gradients and horizontal velocity contributions are neglected (Vitasovic, 1985). The secondary settler is idealized as a settling cylinder with a constant cross sectional area A. Main hypotheses: The concentration of SS is completely uniform within any horizontal plane within the settler; The bottom of the solids-liquid separator represents a physical boundary to separation and the solids flux due to gravitational settling is zero at the bottom; There is no significant biological reaction affecting the solids mass concentration within the separator. Activated sludge model Settling model

12 One-dimensional secondary settling tank model Solid Flux theory The total flux J consists of the bulk flux J b = vx and the settling flux Js=v s X and becomes: where: J=J b +J s J= vx+v s X X sludge concentration, v vertical bulk velocity, v s particle settling velocity. Activated sludge model Settling model

13 One-dimensional secondary settling tank model The Takács particle settling velocity model v s particle settling velocity Activated sludge model Settling model X

14 One-dimensional secondary settling tank model The Takács particle settling velocity model v s particle settling velocity Maximum practical settling velocity Activated sludge model Settling model X

15 One-dimensional secondary settling tank model The Takács particle settling velocity model v S = max(, min( v, v (e -rh(x-xmin) e rp(x-xmin) ))) where X min = f ns X f Maximum practical settling velocity Maximum theoretical Vesilind settling velocity Hindered zone settling parameter Flocculant zone settling parameter Minimum attainable concentration of suspended solids in the effluent Non-settleable fraction Settler influent TSS concentration v v r h r p Xmin f ns X f m d -1 m d -1 m 3 (g TSS) -1 m 3 (g TSS) -1 g TSS/m 3 dimensionless g TSS/m 3 Activated sludge model Settling model

16 Information required for the model characterization A relevant step in the implementation of a model is the collection of information from the full-scale WWTP under study Design data Reactor volumes, maximum pumpflow rates and aeration capacities. Operational data Flows quality characterization Reaction kinetics Flow rates, as averages or dynamic trajectories, of influent, effluent, recycle and waste sludge flows. ph, aeration (flow rates, valve openings, etc.) and temperatures. Concentrations of influent and effluent, in-process measurements (as well as some intermediate streams between the WWTP unit processes), as averages or dynamic trajectories: e.g. SS, COD, TKN, NH4-N, NO3-N, PO4-P, etc. e.g. growth and decay rates. Settler characterization Sedimentation tests for the determination of the settler model parameters.

17

18 High-loaded activated sludge process Population equivalent 5 Flowrate 46 m 3 h -1 F/M >.8 kgbod (kgtss d) -1 Sludge Age < 2 d No nitrification TSS reactor = 26 mgtss l -1

19 Benchmark Simulation Model 1 It was created and developed mainly by members of the International Water Association (IWA) Task Group, between It is not linked to any particular simulation platform. Activated Sludge Model 1 [Henze et al., 1986] Biological reactor Benchmark Simulation Model 1 [IWA, ] 1-D Takács (1991) Model Secondary settler

20 Information required for the model characterization A relevant step in the implementation of a model is the collection of information from the full-scale WWTP under study Source of information Design data Reactor volumes, maximum pumpflow rates and aeration capacities. Executive project Operational data Flows quality characterization Reaction kinetics Flow rates, as averages or dynamic trajectories, of influent, effluent, recycle and waste sludge flows. ph, aeration (flow rates, valve openings, etc.) and temperatures. Concentrations of influent and effluent, in-process measurements (as well as some intermediate streams between the WWTP unit processes), as averages or dynamic trajectories: e.g. SS, COD, TKN, NH4-N, NO3-N, PO4-P, etc. e.g. growth and decay rates. Data available from on-line sensors and flowmeters Measuring campaign Respirometry test Settler characterization Sedimentation tests for the determination of the settler model parameters. Settling column test

21 Information required for the model characterization A relevant step in the implementation of a model is the collection of information from the full-scale WWTP under study Source of information Design data Reactor volumes, maximum pumpflow rates and aeration capacities. Executive project Operational data Water quality characterization Reaction kinetics Flow rates, as averages or dynamic trajectories, of influent, effluent, recycle and waste sludge flows. ph, aeration (flow rates, valve openings, etc.) and temperatures. Concentrations of influent and effluent, in-process measurements (as well as some intermediate streams between the WWTP unit processes), as averages or dynamic trajectories: e.g. SS, COD, TKN, NH4-N, NO3-N, PO4-P, etc. e.g. growth and decay rates. Data available from on-line sensors and flowmeters Measuring campaign Respirometry test Settler characterization Sedimentation tests for the determination of the settler model parameters. Settling column test

22 Data collected from experimental work Measuring Campaign A measuring campaign was necessary in order to perform a dynamic simulation of the treatment process, since it requires the knowledge of the diurnal variation of the influent and effluent wastewater quality. A total of 48 samples, 24 for the influent and 24 for the effluent, were collected and then analyzed for standard pollutants: COD (filtered and non-filtered); BOD (filtered and non-filtered); TSS, Total Suspended Solids; VSS, Volatile Suspended Solids; NH4-N, ammonium; NO 2 -N + NO 3 -N, sum of nitrate and nitrite; Total nitrogen.

23 mg/l [mg N/l] mg/l N tot [mg/l] mg/l Alkalinity [molhco3/m3] mg/l COD [mg/l] mg/l BOD [mg/l] mg/l [mg SS/l] Data collected from experimental work Measuring Campaign Influent quality trajectories COD BOD TSS and VSS Non- Filtered Non- Filtered Filtered 5 Filtered Time [hours] Time (h) NH 4 -N [hours] Time (h) Time [hours] Time (h) N tot Alkalinity 2 1 NO 3 -N Time [hours] Time [hours] Time [hours] Time (h) Time (h) Time (h)

24 Data collected from experimental work Respirometry test It is a tool that measures and interprets the rate at which biomass consumes dissolved oxygen. It offers the possibility of indirectly estimating three kinetic parameters of the heterotrophic metabolism required from the Activated Sludge Model 1: i) the maximum specific growth rate, µ maxh ; ii) the decay rate coefficient, b h ; and iii) the half-saturation coefficient, K s. The values of these parameters are not found directly in the test results; therefore, a mathematical sub-model was created for reproducing the process of the test performed. From the calibration of this sub-model, some possible sets of values for the three parameters were obtained. These were then inserted in the full-scale model and further calibrated.

25 Data collected from experimental work Respirometry test aerated activated sludge + nitrification inhibitor + nutrients + high oxygen content oxymeter stirring

26 mg O 2 /l 1 8 Data collected from experimental work Respirometry test Oxygen concentration Acetate addition : 1: 2: 3: 4: 5: 6: hours

27 mg O 2 /l/h mg O 2 /l 1 8 Data collected from experimental work Respirometry test Oxygen concentration Acetate addition : 1: 2: 3: 4: 5: 6: Oxygen Uptake Rate Acetate addition : 1: 2: 3: 4: 5: 6: hours hours

28 So[mgO2/l] OUR [mgo2/l/h] B 15 Data collected from experimental work t [min] Dissolved oxygen Sub-model calibration Respirometry test t [min] Oxygen Uptake rate t [min] t [min] model output experimental measurements

29 SBL (cm) Data collected from experimental work Sludge blanket level X Settling Column test Time (min) 2.57 gtss/l 2.8 gtss/l 1.56 gtss/l,76 gtss/l

30 ln (Vs) SBL (cm) Data collected from experimental work Sludge blanket level X Zone Settling Velocity Settling Column test Time (min) 2.57 gtss/l 2.8 gtss/l 1.56 gtss/l,76 gtss/l V s =v o exp(αx) y = x X (gtss l -1 )

31 mg/l mg/l COD 4 BOD 1 TSS NH 4 -N NO 3 -N.16 Calibration Comparison between measurements and model outputs 6 N tot

32 mg/l mg/l Calibration Comparison between measurements and model outputs COD 4 BOD 1 TSS NH 4 -N.16 NO 3 -N 6 N tot

33 Thank you for your attention! ANY QUESTIONS?

34