Mathematical models and simulation are considered a powerful tool in engineering practice. Those tools are becoming increasingly used for the improvement of wastewater treatment plants design because the conceptual design is complex and ill-defined. In this paper, three alternatives: 1) complete mix activated sludge without nitrogen removal (CAS); 2) complete mix activated sludge with nitrogen removal (CAS-N) and; 3) membrane bioreactor (MBR) processes were designed into two steps: first concept design to calculate the size of process units, then second implement modeling and simulation to improve the accuracy of the conceptual design. In brief, the treatment process design has been verified by using the activated sludge model No. 1 (ASM1) in GPS-X (v.7) simulation software. This application helps not only in sizing the treatment units but also in understanding the plant’s capacity. In the same time, it can assist in studying the future expansion works required for increased hydraulic and organic loadings. For this purpose, Tikrit WWTP was selected as a case study. The used model was validated by comparing the designed values of the plant and the modeling data. The verification of the obtained results from both hand calculations and the results of the program showed a good agreement. A significant difference in the volume of secondary treatment was obtained from design calculations, where the CAS without denitrification system was 9244 m3 (aerobic and secondary tanks), CAS with denitrification system was 11,324 m3 (anoxic, aerobic and secondary tanks) and for MBR system was 7468 m3 (anoxic, aerobic and immersed membrane tanks). From the obtained results point of view, it can be concluded that mathematical models can be considered as worthy tools to complement the established wastewater treatment plant design procedures.
The activated sludge process (ASP) is the most widely used wastewater treatment technology. This is because of its high flexibility which allows the designer to adopt it to any kind of wastewater, it is the most cost effective, and it is also capable of producing high effluent quality that meets the increasingly stringent effluent standards.
In the ASP, microorganisms are responsible for degradation and removal of contaminants within the wastewater [
Since the beginning of the 20th century, modifications at the design and operational conditions of the conventional ASP have been developed to meet increasingly stringent performance demands [
Technological development in the last three decades has led to the application of a membrane bioreactor (MBR) for many medium and large municipal WWTPs. MBR is a system that combines biological treatment with membrane filtration into a single process. The first reported application of MBR technology was in 1969, when an ultra-filtration membrane was used to separate activated sludge from the final effluent of a biological wastewater treatment system and the sludge was recycled back into the aeration tank [
The most challenging step in wastewater treatment plant design is the selection of treatment process which defined as a combination of unit operations and processes capable of meeting effluent permit requirements [
Traditionally, design of WWTP’s is carried out using empirical equations and simplified system descriptions available in guidelines, for example Metcalf & Eddy [
Currently, the use of dynamic modeling and simulation is a common practice in the field of wastewater treatment [
Focusing on the design applications, after the preliminary concept design step based on the guidelines, modeling and simulation tool could be used to: 1) evaluate the proposed process alternatives for new WWTP units; 2) check and validate the design in order to confirm different process units sizing, sludge concentrations, recirculation rates, and effluent compliance to the identified standard of discharge limits; and 3) check the performance of the WWTP under dynamic conditions.
The main objective of this paper is to present and demonstrate a systematic approach to design three alternatives of biological wastewater treatment systems, namely conventional activated sludge (CAS1), conventional activated sludge with removal nitrogen (CAS-N) and membrane bioreactor (MBR) for Tikrit, Iraq WWTP as a case study. The design of the three alternatives was conducted into two steps: 1) use the empirical formula and guidelines to initially design the WWTP units, and 2) apply modeling and simulation tool to verify the initial design of the three alternatives using the software GPS-X (v.7).
In order of the design requirements for the present study, a WWTP located in the area of Tikrit, Iraq was selected as a reference case-study. The WWTP is used to represent the characteristics of municipal wastewater treatment, design population, and the design flow for the plant. The information of this plant was obtained from “ER-GE Design, Engineering, Consulting & Trade Ltd. Co” report issued in 2013 [
Designing of wastewater treatment procedure was depended on the characteristics of wastewater, topographic and geographic features in Tikrit city. The estimation amount of both present and projected populations was essential to know the quantity or volume of wastewater in terms of flow. The population that contributes to the treatment plant is that situated inside the design area served by the sewerage system. The construction of WWTP required massive fund investment, so, the design was divide the project into two phases. The planning timeframe of wastewater treatment plant was assumed 39 years from 2018 until 2057 with two phases (Phase I up to 2033 and phase II up to 2057). According to the report [
Parameter | Unit | Measured average value | Iraqi standard |
---|---|---|---|
PH Total suspended solids, SS | - mg/L | 7.9 350 | 6 - 9.5 <30 |
Biological Oxygen Demand, BOD5 | mg/L | 320 | <20 |
Chemical Oxygen Demand , COD | mg/L | 615.2 | <60 |
Total TKN | mg/L | 60 | <50 |
Ammonia Nitrogen, N-NH4 | mg/L | 45 | <10 |
Organic N | mg/L | 15 | - |
Nitrate Nitrogen, N-NO3 | mg/L | 0 | <25 |
Total P | mg/L | 8 | <3 |
Average wastewater temperature during summer | ˚C | 25 | <35 |
Average wastewater temperature during winter | ˚C | 15 | <35 |
Elevation from mean sea depth | m | 115 | - |
In this paper three alternative configurations of the wastewater treatment systems (CAS, CAS-N and MBR) were selected. The design developed for the units of Screen Chamber and Grit Removal Unit for pre-treatment, Primary Clarifier Unit for primary treatment and Anoxic Tank, Aeration Tank, Secondary Clarifier Unit and Membrane Filter Unit for secondary treatment as they are commonly used in the field of wastewater treatment. The designed process was studied and implemented according to various guidelines principles found in the international literature [
Designs for the given guidelines are determined by defining influent wastewater characteristics, specifying operating preferences (e.g. DO and MLSS concentration in the reactors), selecting safety factors (SF) and setting the effluent requirements. The design outcomes will present in the next section.
The design of the WWTPs includes the units of preliminary treatment, primary treatment, and secondary treatment. The preliminary treatment and primary treatment units were assumed the same design sizes for the three studied alternatives (CAS, CAS-N and MBR) because the flow rate was the same value of each alternative. Design and simulation of sludge processing was not included in the overall WWTP model, due to the lack of some relevant data.
1) Preliminary treatment
Preliminary treatment stage includes two units (coarse screen and grit chamber). The design procedure used in this design stage is described below:
a) Screen chamber
The screen chamber is provided at the binging of WWTP in order to prevent large particles to pass through. The design details are shown in
b) Grit chamber
Grit chamber is designed to remove grit consisting of sand, gravel, cinders or other heavy solid materials that have subsiding velocities or specific gravities substantially greater than those of the organic putrescible solids in wastewater [
2) Primary treatment
Parameter & design criteria* | Equation* & obtained value |
---|---|
Design flow, Qmax= 60,000 m3/day Approach velocity in the channel, Vh =0.6 m/s Depth to width ratio = 1:1.5 No. of screen in unit at less 2 | Area of screen A s = Q max V h = 1.5 m 2 |
Clear spacing between bars, S = 30 mm Thick of bar, tbar =10 mm | Number of bars in the screen: N ∗ t bar + ( N + 1 ) S = W = 37 |
Coefficient of discharge for clean screen, C = 0.7 velocitythroughscreennars V t h = 0.615 m/s | Headloss = 1 C ∗ ( V t h 2 − V h 2 ) 2 g = 1.4 mm |
*Used according to the source: (Metcalf & Eddy 2003) [
Parameter & design criteria* | Equation* & obtained value |
---|---|
Design flow, Qpeak = 82,500 m3/day No. of grit chamber in unit = 2 Detention time, D.T = 3 min Depth: width = 1: 1.2 Depth = 3 m | Volumeofeachtank = Q p e a k ∗ D .T / 2 = 86 m 3 Width = 3.6 m Length = 8 m |
The air-supply requirement = 0.3 m3/min/m | totalAirRequirement = 4.8 m 3 / min |
Quantity of grit to be removed = 0.015 m3/103m3 at Qave | Totalvolumeofgrit = 0.45 m 3 / d |
*Used according to the source: (ATV-DVWK 2000) [
In primary treatment the wastewater still contains non-coarse suspended solids (settleable solids and part of the organic matter), which can be partially removed in sedimentation units. Primary clarifier unit represents the primary treatment stage.
3) Secondary treatment
The wastewater will be allowed for biological treatment in the secondary treatment process which includes three different systems configuration:
a) Carbonaceous removal process (CAS)
In this system, only carbonaceous matter is removed by microorganisms in the reactor (aerobic tank). The computation approach for the design of the activated-sludge process is presented in
b) Crbonaceous, nitrification and denitrification removal process (CAS-N).
This system is designed for the removal of both carbonaceous and nitrogenous substance by microorganisms in the reactor (anoxic and aerobic tank). The computation approach for the design of the ASP is shown in
c) Secondary clarifier design
As part of the proposed secondary treatment for (CAS and CAS-N), a secondary clarifier is designed. The secondary clarifier is essential for the removal of suspended solids that encompass the nutrients. The computation approach for the design of the secondary clarifier is shown in
d) Carbonaceous, nitrification and denitrification removal process with immerged membrane
The computation third approach for the design of nitrification and de-nitrification tanks was same that used in design alternative two (activated sludge with biological nitrogen removal). The design criteria of nitrification and de-nitrification tanks were also same except MLVSS and SRT were used 8000 mg/l and 20 days respectively. The computation approach for the design was found Se= 0.73 mg/l,
p x , b i o = 1842 kg d , P x , T = 2839kg / d ,
V = 6218 m3, Vanox = 1250 m3, Xanox = 2114 mg/l,
F M anox = 3.5 d − d and capacity ratio = 1.13.
Parameter & design criteria* | Equation* & obtained value |
---|---|
Design flow, Qave = 30,000 m3/day Surface loading rate, SLR = 40 m3/m2・d | Area ( A ) = Q SLR = 750 m 2 |
No. of Primary sedimentation tank = 2 Depth = 3 m | Diameter: D = 4 A π = 22 m Weirloadingrate , WLR = 217 m 3 / m .d |
Empirical constants, a, b: 0.018, 0.020 for BOD 0.0075, 0.014 for TSS | expected removal efficiency, R = t a + b . t = 33.5 % BOD , 55.3 % TSS |
*Used according to the source: (Metcalf & Eddy 2003) [
Parameter & design criteria* | Equation* & obtained value |
---|---|
BOD5/BODu = 0.666% BODu/Xb = 1.42% Desired effluent SS = 25 mg/l Biodegradable fraction (generating solids) = 0.65 Desired effluent BOD = 20mg/l | S s s = BOD 5 BOD u ∗ [ BOD u X b * S S b / S S e ] = 15.4 mg/l S sol = S total − S ss = 4.6 mg/l |
Sludge age, θc = 10 days S0 = 212.8 mg/l Yield coefficient, Y = 0.5 MLVSS concentration = 3500 mg/l Endogenous respiration coefficient, Kd = 0.057d−1 at (15˚C) | V = Y θ c Q a v e ( S o − S s o l ) MLVSS ( 1 + K d θ c ) = 5 , 684 m 3 |
Return sludge ratio Qr/Q = 0.8 Concentration of recycle sludge, Xr = 10,000 mg/l | θ c = V ∗ X ( Q W ∗ X r ) + ( Q e ∗ X e ) → sludgewasterate , Q W = 174 m 3 / d |
Hydraulic detention time, HRT = V / ( Q ) = 0.19 d | F M = S 0 HRT ∗ MLVSS = 0.32 d − 1 |
*Used according to the source: (Metcalf & Eddy 2003; ATV-DVWK 2000) [
e) Membrane system
The design of the membrane system includes determining the design flux, the required number of membrane modules, and the aeration requirement for coarse bubble aeration.
A computer program GPS-X (v.7) software package (Hydromantis Inc., Ontario, Canada) was used in this study to verify and optimize the obtained plants design. It was also used to simulate and verify the process performance in normal condition and shock loading conditions (double organic load and increase of influent flow). The predicted effluent concentrations obtained with the process model are then compared to the effluent requirements which were imposed for the designs. These enable to investigate whether the guidelines lead to optimal designs or to over- or under-sized plants. It is important to stress that the same criteria used
Parameter & design criteria* | Equation* & obtained value |
---|---|
Sludge age, θc = 10 days Half-saturation coefficient: Ks = 20 mg/l maximum specific growth rate, m = 4.28 d−1 Endogenous respiration coefficient, Kd = 0.1 d−1 at (15˚C) | EffluentCOD , S = K S ( 1 + K d SRT ) SRT ( m − K d ) − 1 = 1 m g COD / l |
Design flow, Qave = 30,000 m3/day biodegradable COD, bCOD = 1.6 × BOD = 340 mg/l oxidizable ammonia concentration, NOx = 49.3 mg/l yield coefficient, Y = 0.5 growth yield of nitrifying bacteria, yn = 0.12 decay constant of nitrifying bacteria, Kdn (15˚C) = 0.065 d−1 fraction of biomass accumulated during decay, fd = 0.15 inert material in influent wastewater, X o , i = 21.5 m g / l | solids production, p x , b i o Q Y ( S o − S ) 1 + ( k d ) SRT + Q y n N O x 1 + ( k d n ) SRT + f d k d Q Y ( S o − S ) 1 + ( k d ) SRT SRT = 2443.5 kg / d total solids, Px,T P x , T = p x , b i o + Q X o , i = 3088.5 kg / d |
MLVSS concentration =3500 mg/l | V = P x , T MLVSS = 8824 m 3 |
H.D.T = 2 h | V a n o x = H .D .T ∗ Q a v e / 24 = 2500 m 3 |
Effluent nitrate concentration, N O e = 25 mg / l recycle nitrate = N O x − N O e = 24.3 mg/ l I R = N O x / N O r − 1 = 1.03 based on Metcalf & Eddy SDNR = 0.35 g/g∙d (15˚) = 0.31 | X a n o x = Q ∗ S R T V e r a ∗ Y ( S o − S ) 1 + ( k d ) S R T ∗ I R I R + 1 = 1170 mg/l F M a n o x = Q S o V a n o x X a n o x = 3.5 d − 1 Noxfeed = Q ∗ I R ∗ N O r = 751 kg / d NOrreduce = V a n o x ∗ SDNR ∗ X a n o x = 907 kg / d Capacityratio = NOreduce Noxfeed = 1.2 |
Return sludge ratio Qr/Q = 0.8 concentration of recycle sludge, Xr = 10,000 mg/l | θ c = V ∗ X ( Q W ∗ X r ) + ( Q e ∗ X e ) sludgewasterate , Q W = 311 m 3 / d |
*Used according to the source: (Metcalf & Eddy 2003) [
Parameter & design criteria* | Equation* & obtained value |
---|---|
Design flow, Qave = 30,000 m3/day Surface loading rate, SLR = 16 m3/m2・d | Area ( A ) = Q SLR = 1875 m 2 |
No. of Primary sedimentation tank = 2 Depth = 3.7 m | Diameter: D = 4 A π = 35 m Weirloadingrate , WLR = 375 m 3 / m .d |
*Used according to the source: (Metcalf & Eddy 2003; ATV-DVWK 2000) [
Parameter & design criteria* | Equation* & obtained value |
---|---|
Maximum operating flux = 40 L/m2・h Filtration ratio = 0.968 Operating ratio = 0.988 Peaking factor = 2.75 | Designflux = ( Maximumoperatingflux ) ( Filtrationratio ) ( Operatingratio ) Peakingfactor = 14 L / m 2 ⋅ h |
Membrane area per module = 31.6 m2/module | Numberofmodules = Dailyaverageflow ( Designflux ) ( Membraneareapermodule ) = 2826modules |
Membrane area=89301.6 m2 SADm = 0.54 m3/m2・h Filtration = 900 sec Backwash = 30 sec | coarse bubble aeration Q A , m = ( SAD m ) ( Membranearea ) filterationtime cycletime = 46 , 667 m 3 h |
Membrane packing density=45 m2/m3 | V m = A m packingdensity = 1985 m 3 |
*Used according to the source: (Park et al. 2015) [
for the design requirements are applied in the evaluation step.
In order to carry out the WWTPs simulation, the same inputs used for the designs (influent characteristics and operating preferences together with the design outcomes obtained from the guideline are used to develop the process model. The major models used to evaluate the designs WWTPs includes mainly the Activated Sludge Model No. 1 (ASM1) for biological processes, the BOD based influent model for influent characterization, and the model simple1d for clarification process [
Daily average value of influent flow rate, influent concentrations [Biological Oxygen Demand (BODin), Total suspended solids (Xin), and Total Kjeldahl Nitrogen (TKNin)] were used as inputs for the model and Mixed Liquor Suspended Solids in the aeration tank (MLSSr). The value of DO concentration was set at 2 mg/L for oxygen transfer efficiency of the fine bubble diffusers.
In order to verify the design results for the WWTPs (CAS, CAS-N and MBR), the adjustment of some operational parameters have to be taken into account. Steady-state simulation provides a solution to the system based on the average influent flow to the system. Those parameters were adjusted one by one until the model fits well.
In order to understand the differences obtained between some values before and after model calibration several factors have to be taken into account: 1) In fact, the design allows selecting two important characteristics of the treatment plant: the MLSS and the SRT. However, it is impossible to impose both on the dynamic process model. When imposing an MLSS concentration, the BSM1
Parameters | Units | Calculations based on guidelines | Calculations based on model | ||||
---|---|---|---|---|---|---|---|
CAS | CAS-N | MBR | CAS | CAS-N | MBR | ||
MLVSS in anoxic tank, Xanox | mg/l | - | 1170 | 2114 | - | 1766 | 2265 |
MLVSS in aerobic tank, | mg/l | 3500 | 3500 | 8000 | 3500 | 3502 | 8012 |
Total solid retention time, SRT | d | 10 | 10 | 20 | 10 | 10.5 | 18 |
Waste activated sludge | m3/d | 174 | 311 | 311 | 355 | 547.2 | 370 |
RAS recycle ratio | % | 80 | 1.03 | 0.94 | 50 | 1.5 | 1.2 |
simulations result in change the SRTs. The differences in the SRTs can be explained by the different model structures of both models that do not allow fixing the MLSS and the SRT of the system at the same time. Note that, when fixing the SRTs (and leaving the MLSS to evolve to lower values) would make that the effluent organic compounds concentrations would come higher and closer to the values the plant was designed for. 2) Sludge production is much higher after modeling compared to the expected sludge production from the hand calculations. The different between default kinetic and stoichiometric parameters for both the used guidelines and model structure, may lead to those differences [
The results and discussion part is divided into two parts. The first part deals with the empirical design of the wastewater treatment plants, while the second one deals with the application of modeling and simulation tool to verify the initial design WWTPs using GPS-X (v.7).
The MBR has the lowest volume compared to CAS processes due to use an immerged membrane in the aeration tanks with surface area equals 89,302 m2. The membrane is used as a filter instead of gravity sedimentation tanks, which results in a smaller footprint than CAS processes.
The second goal of this study is to verify the design guideline with ASM1-based process model. Verification processes were implemented through the examination of the process unit’s sizes by means of the performance of CAS, CAS-N and MBR under two scenarios (stable and unstable conditions) to ensure the accuracy of the plants to reach the design effluent requirements.
In this stage, the analysis was prepared based on the steady state performance. To assess the performance of each system to remove organic matter, the flow rate was kept constant at 30,000 m3/d and the domestic wastewater characteristics of Tikrit WWTP is the same as given in
The units included of each stage | Units | Footprint required of each stage | |||
---|---|---|---|---|---|
CAS | CAS-N | MBR | |||
Preliminary treatment | Screen chamber | m2 | 1.5 | 1.5 | 1.5 |
Grit chamber | m3 | 172 | 172 | 172 | |
primary treatment | Primary clarifier | m3 | 2250 | 2,250 | 2250 |
Secondary treatment | Anoxic tank | m3 | - | 2500 | 1250 |
Aerobic tank | m3 | 5684 | 8824 | 6218 | |
secondary clarifier | m3 | 6937 | 6937 | - |
The conditions at any wastewater treatment process are usually altered due to the change in flow-rate and composition of wastewater with the time. The unstable conditions lead to complexity in the operation of the activated sludge processes, which requires accurate and well developed modeling techniques. In order to assess the changed conditions on the performance of WWTP, two scenarios were investigated.
The aim of the first scenario is to investigate the performance of the three processes under a hydraulic shock load. So, the flow rate was increased from 20,000 m3/d to 60,000 m3/d.
The aim of the second scenario is to investigate the performance of the three processes under an organic shock load. The flow rate was kept constant at 30,000 m3/day for CAS, CAS-N and MBR. While, the double of the influent organic load was used (TSS = 700 mg/l, BOD = 640 mg/l, COD = 1,208 mg/l and TKN = 120 mg/l). Results represented in
After the application of those scenarios, good simulation results were obtained to assure the reliability of the program in the running of different scenarios. Therefore it can be demonstrated that the empirical design is fruitful, and the sizes of the process units achieve the requirements of Iraqi code.
BSM1 simulations can be used to study the change of the reactor volume. The volume can be reduced until the predicted effluent concentrations reach the design effluent requirements. In this case the dynamic simulations are applied with gradual changes in the volume of the aerobic reactors. The results represented in
compared to the BSM1 simulations. This may be due to the differences in the model structure and their default model parameters [
The plant size can be significantly reduced if the safety margins included in the dimensioning guidelines is removed. In Benedetti et al. [
The present study set out to design three configurations of WWTPs and verify the designed process by the application of GPS-X model. From the methodology of conceptual design described in this paper, results illustrated that proper design construction combined with good produce effluents can be developed to support optimal treatment process selection. In general, using application of GPS-X is very helpful tool in verifying the pre-design of WWTPs. In addition, this application helps in understanding the plant’s performance under different conditions as well as in deciding the future expansion works needed for increased hydraulic and organic loadings. It can be concluded that the verification of design guidelines is not easy issue and there is still further work to do, however, both design guidelines and dynamic process model simulations have their role to play and to make better understanding of the impact of various factors of the process design.
Arif, A.U.A., Sorour, M.T. and Aly, S.A. (2018) Design and Comparison of Wastewater Treatment Plant Types (Activated Sludge and Membrane Bioreactor), Using GPS-X Simulation Program: Case Study of Tikrit WWTP (Middle Iraq). Journal of Environmental Protection, 9, 636-651. https://doi.org/10.4236/jep.2018.96040