A novel protocol for model calibration in biological wastewater treatment

The coexistence and competition of canonical and comammox nitrite oxidizing bacteria in a nitrifying activated sludge system - Experimental observations and simulation studies

The second step of nitrification can be mediated by nitrite oxidizing bacteria (NOB), i.e. Nitrospira and Nitrobacter, with different characteristics in terms of the r/K theory. In this study, an activated sludge model was developed to account for competition between two groups of canonical NOB and comammox bacteria. Heterotrophic denitrification on soluble microbial products was also incorporated into the model. Four 5-week washout trials were carried out at dissolved oxygen-limited conditions for different temperatures (12 °C vs. 20 °C) and main substrates (NH₄⁺-N vs. NO₂^--N). Due to the aggressive reduction of solids retention time (from 4 to 1 d), the biomass concentrations were continuously decreased and stabilized after two weeks at a level below 400 mg/L. The collected experimental data (N species, biomass concentrations, and microbiological analyses) were used for model calibration and validation. In addition to the standard predictions (N species and biomass), the newly developed model also accurately predicted two microbiological indicators, including the relative abundance of comammox bacteria as well as nitrifiers to heterotrophs ratio. Sankey diagrams revealed that the relative contributions of specific microbial groups to N conversion pathways were significantly shifted during the trial. The contribution of comammox did not exceed 5 % in the experiments with both NH₄⁺-N and NO₂^--N substrates. This study contributes to a better understanding of the novel autotrophic N removal processes (e.g. deammonification) with nitrite as a central intermediate product.

Modeling nutrient removal and energy consumption in an advanced activated sludge system under uncertainty

Activated sludge models are widely used to simulate, optimize and control performance of wastewater treatment plants (WWTP). For simulation of nutrient removal and energy consumption, kinetic parameters would need to be estimated, which requires an extensive measurement campaign. In this study, a novel methodology is proposed for modeling the performance and energy consumption of a biological nutrient removal activated sludge system under sensitivity and uncertainty. The actual data from the wastewater treatment plant in Slupsk (northern Poland) were used for the analysis. Global sensitivity analysis methods accounting for interactions between kinetic parameters were compared with the local sensitivity approach. An extensive procedure for estimation of kinetic parameters allowed to reduce the computational effort in the uncertainty analysis and improve the reliability of the computational results. Due to high costs of measurement campaigns for model calibration, a modification of the Generalized Likelihood Uncertainty method was applied considering the location of measurement points. The inclusion of nutrient measurements in the aerobic compartment in the uncertainty analysis resulted in percentages of ammonium, nitrate, ortho-phosphate measurements of 81%, 90%, 78%, respectively, in the 95% confidence interval. The additional inclusion of measurements in the anaerobic compartment resulted in an increase in the percentage of ortho-phosphate measurements in the aerobic compartment by 5% in the confidence interval. The developed procedure reduces computational and measurement efforts, while maintaining a high compatibility of the observed data and model predictions. This enables to implement activated sludge models also for the facilities with a limited availability of data.

New insights into modeling two-step nitrification in activated sludge systems - The effects of initial biomass concentrations, comammox and heterotrophic activities

In this study, the conventional two-step nitrification model was extended with complete ammonia oxidation (comammox) and heterotrophic denitrification on soluble microbial products. The data for model calibration/validation were collected at four long-term washout experiments when the solid retention time (SRT) and hydraulic retention time (HRT) were progressively reduced from 4 d to 1 d, with mixed liquor suspended solids (MLSS) of approximately 2000 mg/L at the start of each trial. A new calibration protocol was proposed by including a systematic calculation of the initial biomass concentrations and microbial relationships as the calibration targets. Moreover, the impact assessment of initial biomass concentrations (X) and maximum growth rates (μ) for ammonia-oxidizing bacteria (AOB), nitrite-oxidizing bacteria (NOB), comammox Nitrospira, and heterotrophs on the calibration accuracy were investigated using the response surface methodology (RSM). The RSM results revealed the strongest interaction of X_AOB and μ_AOB on the model calibration accuracy. All the examined model efficiency measures confirmed that the extended model was accurately calibrated and validated. The estimated μ values were as follows: μ_AOB = 0.38 ± 0.005 d^-1, μ_NOB = 0.20 ± 0.01 d^-1, μ_CMX = 0.20 ± 0.01 d^-1, μ_HET = 1.0 ± 0.03 d^-1. For comparison, when using the conventional model, μ_AOB and μ_NOB increased respectively by 26 and 15 % (μ_AOB = 0.48 ± 0.02 d^-1 and μ_NOB = 0.23 ± 0.005 d^-1). This study provides better understanding of the effects of the initial biomass composition and the accompanying processes (comammox and heterotrophic denitrification) on modeling two-step nitrification.

Enhancement of biological nitrogen removal performance from low C/N municipal wastewater using novel carriers based on the nano-Fe3O4

The aim of this work was to study the effects of the magnetic microparticles (MMP) on nitrogen removal under low C/N conditions. A 30-day anaerobic/oxic progress illustrated that nitrification and denitrification were promoted in the presence of MMP. MMP could facilitate the production of extracellular polymeric substances (EPS) and act as pH buffering in aerobic conditions. The high-throughput sequencing displayed that, compared with the sludge without MMP, the relative abundance of Dokdonella and Comamonas which are capable of both nitrifying and denitrifying were 8.7% and 1.29% higher in anaerobic sludge and 7.11% and 0.97% higher in aerobic sludge with MMP, respectively. The relative abundance of Pseudomonas with the excellent capability of EPS secretion was also observed 4.33 times higher than that without MMP in the aerobic sludge. Based on the superior performance above, MMP is a promising additive to enhance nitrogen removal efficiency for low C/N wastewater.

A protocol for dynamic model calibration

Ordinary differential equation models are nowadays widely used for the mechanistic description of biological processes and their temporal evolution. These models typically have many unknown and nonmeasurable parameters, which have to be determined by fitting the model to experimental data. In order to perform this task, known as parameter estimation or model calibration, the modeller faces challenges such as poor parameter identifiability, lack of sufficiently informative experimental data and the existence of local minima in the objective function landscape. These issues tend to worsen with larger model sizes, increasing the computational complexity and the number of unknown parameters. An incorrectly calibrated model is problematic because it may result in inaccurate predictions and misleading conclusions. For nonexpert users, there are a large number of potential pitfalls. Here, we provide a protocol that guides the user through all the steps involved in the calibration of dynamic models. We illustrate the methodology with two models and provide all the code required to reproduce the results and perform the same analysis on new models. Our protocol provides practitioners and researchers in biological modelling with a one-stop guide that is at the same time compact and sufficiently comprehensive to cover all aspects of the problem.

Incorporation of the complete ammonia oxidation (comammox) process for modeling nitrification in suspended growth wastewater treatment systems

The newly discovered process complete ammonia oxidation (comammox) has changed the traditional understanding of nitrification. In this study, three possible concepts of comammox were developed and incorporated as part of an extended two-step nitrification model. For model calibration and validation, two series of long-term biomass washout experiments were carried out at 12 °C and 20 °C in a laboratory sequencing batch reactor. The inoculum biomass was withdrawn from a large biological nutrient removal wastewater treatment plant. The efficiency of the examined models was compared based on the behaviors of ammonia, nitrite, and nitrate in the studied reactor. Predictions of the conventional approach to comammox, assuming the direct oxidation of ammonia to nitrate, were slightly better than the two other approaches. Simulation results revealed that comammox could be responsible for the conversion of >20% of the influent ammonia load. Therefore, the role of commamox in the nitrogen mass balance in activated sludge systems should not be neglected and requires further investigation. Furthermore, sensitivity and correlation analysis revealed that the maximum growth rates (μ), oxygen half-saturation (K_O), and decay rates (b) of the canonical nitrifiers and comammox were the most sensitive factors, and the highest correlation was found between μ and b among all considered kinetic parameters. The estimated μ values by the best model were 0.57, 0.11, and 0.15 d^-1 for AOB, NOB, and comammox bacteria, respectively.

Inverse modeling of laboratory experiment to assess parameter transferability of pesticide environmental fate into outdoor experiments under paddy test systems

BACKGROUND

Extraction of environmental fate parameters for pesticides by inverse modeling in laboratory experiments has evolved to become a common practice in higher tier exposure modeling. This study focuses on flooded paddy soil conditions using a simple container test system. Four active ingredients of paddy herbicide were tested. The results were parameterized and transferred to analyze the effect of formulation types on the outdoor experimental data via inverse analyses of two structurally-compatible mathematical models, namely: pesticide concentration in paddy field for laboratory (PCPF-LR) and PCPF for outdoors (PCPF-1R_v1.1 ).

RESULTS

After in-laboratory calibration, the PCPF-LR model revealed statistically acceptable or ideal simulations of pesticide concentrations in both the aqueous and soil phases (e.g. Nash-Sutcliffe efficiency > 0.7), in addition to determining the apparent sorption from the laboratory data. The extracted persistence indicators (degradation half-life, DegT₅₀ ) in the aqueous phase were 1.4-38.7 times higher than those of the dissipation (DT₅₀ ) due to the exclusion of partitioning and phase transfer processes (diffusion and sorption). In the outdoor experiment, 72% of the outdoor-calibrated simulations of the PCPF-1R_v1.1 model, showed statistically acceptable representations of the concentrations in paddy water. Furthermore, the DegT₅₀ as 'bulk' degradation in paddy water was statistically insignificant between the formulation types; however, the DT₅₀ demonstrated statistically different results.

CONCLUSION

The laboratory/outdoor data interconnections using proposed modeling approach facilitate the data-specific model calibration and analysis. These can be useful in the exposure modeling of paddy pesticide by manipulating the parameter uncertainties associated with the experimental constraints. © 2020 Society of Chemical Industry.

A Methodology for Global Sensitivity Analysis of Activated Sludge Models: Case Study with Activated Sludge Model No. 3 (ASM3)

The main objective of this study was to demonstrate a computational approach of global sensitivity analysis (GSA) integrated with functional principal component analysis (fPCA) for activated sludge models through aggregation of time-dependent model response patterns into time-independent coefficients of functional principal components (PCs). This proposed approach addresses the main issue of time-varying character of GSA indices when calculated solely on the time-dependent model outputs. The GSA-fPCA methodology was implemented using the rigorous model Activated Sludge Model No. 3 (ASM3) as case study. The approach transforms the time-dependent model outputs into functional PCs prior to calculation of GSA indices to remove the time-varying character of the calculated GSA indices. This work focused on the evaluation of the following key computational factors that may significantly influence the performance of the GSA-fPCA methodology: (a) model parameter sampling range, (b) model simulation period, (c) basis functions system, and (d) state of the system being modeled-batch or continuous activated sludge process. Results show that first few functional PCs capture up to 100% of the curve patterns in the time-dependent model outputs. The sensitivity indices calculated from the PC scores via Morris' GSA technique elucidated parameter sensitivity patterns inherent to the complex mathematical structure of ASM3. PRACTITIONER POINTS: Functional principal components-mediated GSA technique to remove time-varying character of sensitivity indices derived from time-dependent dynamical models. Technique amenable to improving efficiency of capturing response patterns into few functional principal components through various basis functions. Identifying priority parameters for ASM3 model calibration requires specification of target model outputs to which parameter sensitivities are calculated. GSA-fPCA offers a comprehensive numerical approach to manipulating models depending on the intended applications: simple fast-responding models to complex models.

Modeling nitrous oxide production by a denitrifying-enhanced biologically phosphorus removing (EBPR) activated sludge in the presence of different carbon sources and electron acceptors

In this study, the IWA Activated Sludge Model No. 2d (ASM2d) was expanded to identify the most important mechanisms leading to the anoxic nitrous oxide (N₂O) production in the combined nitrogen (N) and phosphorus (P) removal activated sludge systems. The new model adopted a three-stage denitrification concept and was evaluated against the measured data from one/two-phase batch experiments carried out with activated sludge withdrawn from a local, large-scale biological nutrient removal wastewater treatment plant. The experiments were focused on investigating the effects of different external carbon sources (acetate, ethanol) and electron acceptors (nitrite, nitrate) on the mechanisms of N₂O production in enhanced biological P removal by polyphosphate accumulating organisms (PAOs) and external carbon-based denitrification by ordinary heterotrophic organisms (OHOs). The experimental results explicitly showed that N₂O production was predominantly governed by the presence of nitrite in the reactor regardless of the examined carbon source and the ratio COD/N in the reactor. The model was capable of accurately predicting (with R² > 0.9) the behavior of not only N₂O-N, but also NO₃-N, NO₂-N, soluble COD, and PO₄-P. The simulation results revealed that only OHOs were responsible for N₂O production, whereas the present denitrifying PAOs reduced only nitrate to nitrite.

Strategies for calibrating models of biology

Computational and mathematical modelling has become a valuable tool for investigating biological systems. Modelling enables prediction of how biological components interact to deliver system-level properties and extrapolation of biological system performance to contexts and experimental conditions where this is unknown. A model's value hinges on knowing that it faithfully represents the biology under the contexts of use, or clearly ascertaining otherwise and thus motivating further model refinement. These qualities are evaluated through calibration, typically formulated as identifying model parameter values that align model and biological behaviours as measured through a metric applied to both. Calibration is critical to modelling but is often underappreciated. A failure to appropriately calibrate risks unrepresentative models that generate erroneous insights. Here, we review a suite of strategies to more rigorously challenge a model's representation of a biological system. All are motivated by features of biological systems, and illustrative examples are drawn from the modelling literature. We examine the calibration of a model against distributions of biological behaviours or outcomes, not only average values. We argue for calibration even where model parameter values are experimentally ascertained. We explore how single metrics can be non-distinguishing for complex systems, with multiple-component dynamic and interaction configurations giving rise to the same metric output. Under these conditions, calibration is insufficiently constraining and the model non-identifiable: multiple solutions to the calibration problem exist. We draw an analogy to curve fitting and argue that calibrating a biological model against a single experiment or context is akin to curve fitting against a single data point. Though useful for communicating model results, we explore how metrics that quantify heavily emergent properties may not be suitable for use in calibration. Lastly, we consider the role of sensitivity and uncertainty analysis in calibration and the interpretation of model results. Our goal in this manuscript is to encourage a deeper consideration of calibration, and how to increase its capacity to either deliver faithful models or demonstrate them otherwise.

A systematic model identification method for chemical transformation pathways - the case of heroin biomarkers in wastewater

This study presents a novel statistical approach for identifying sequenced chemical transformation pathways in combination with reaction kinetics models. The proposed method relies on sound uncertainty propagation by considering parameter ranges and associated probability distribution obtained at any given transformation pathway levels as priors for parameter estimation at any subsequent transformation levels. The method was applied to calibrate a model predicting the transformation in untreated wastewater of six biomarkers, excreted following human metabolism of heroin and codeine. The method developed was compared to parameter estimation methods commonly encountered in literature (i.e., estimation of all parameters at the same time and parameter estimation with fix values for upstream parameters) by assessing the model prediction accuracy, parameter identifiability and uncertainty analysis. Results obtained suggest that the method developed has the potential to outperform conventional approaches in terms of prediction accuracy, transformation pathway identification and parameter identifiability. This method can be used in conjunction with optimal experimental designs to effectively identify model structures and parameters. This method can also offer a platform to promote a closer interaction between analytical chemists and modellers to identify models for biochemical transformation pathways, being a prominent example for the emerging field of wastewater-based epidemiology.

Efficient model calibration method based on phase experiments for anaerobic-anoxic/nitrifying (A2N) two-sludge process

A systematic calibration and validation procedure for the complex mechanistic modeling of anaerobic-anoxic/nitrifying (A2N) two-sludge system is needed. An efficient method based on phase experiments, sensitivity analysis, and genetic algorithm is proposed here for model calibration. Phase experiments (anaerobic phosphorus release, aerobic nitrification, and anoxic denitrifying phosphate accumulation) in an A2N sequencing batch reactor (SBR) were performed to reflect the process conditions accurately and improve the model calibration efficiency. The calibrated model was further validated using 30 batch experiments and 3-month dynamic continuous flow (CF) experiments for A2N-SBR and CF-A2N process, respectively. Several statistical criteria were conducted to evaluate the accuracy of model predications, including the average relative deviation (ARD), mean absolute error (MAE), root mean square error (RMSE), and Janus coefficient. Visual comparisons and statistical analyses indicated that the calibrated model could provide accurate predictions for the effluent chemical oxygen demand (COD), ammonia nitrogen (NH₄⁺-N), total nitrogen (TN), and total phosphorus (TP), with only one iteration.

Improving the efficiency of dissolved oxygen control using an on-line control system based on a genetic algorithm evolving FWNN software sensor

This work proposes an on-line hybrid intelligent control system based on a genetic algorithm (GA) evolving fuzzy wavelet neural network software sensor to control dissolved oxygen (DO) in an anaerobic/anoxic/oxic process for treating papermaking wastewater. With the self-learning and memory abilities of neural network, handling the uncertainty capacity of fuzzy logic, analyzing local detail superiority of wavelet transform and global search of GA, this proposed control system can extract the dynamic behavior and complex interrelationships between various operation variables. The results indicate that the reasonable forecasting and control performances were achieved with optimal DO, and the effluent quality was stable at and below the desired values in real time. Our proposed hybrid approach proved to be a robust and effective DO control tool, attaining not only adequate effluent quality but also minimizing the demand for energy, and is easily integrated into a global monitoring system for purposes of cost management.

Utilization and control of ecological interactions in polymicrobial infections and community-based microbial cell factories

Microbial activities are most often shaped by interactions between co-existing microbes within mixed-species communities. Dissection of the molecular mechanisms of species interactions within communities is a central issue in microbial ecology, and our ability to engineer and control microbial communities depends, to a large extent, on our knowledge of these interactions. This review highlights the recent advances regarding molecular characterization of microbe-microbe interactions that modulate community structure, activity, and stability, and aims to illustrate how these findings have helped us reach an engineering-level understanding of microbial communities in relation to both human health and industrial biotechnology.

Comparison of bacterial communities of conventional and A-stage activated sludge systems

The bacterial community structure of 10 different wastewater treatment systems and their influents has been investigated through pyrosequencing, yielding a total of 283486 reads. These bioreactors had different technological configurations: conventional activated sludge (CAS) systems and very highly loaded A-stage systems. A-stage processes are proposed as the first step in an energy producing municipal wastewater treatment process. Pyrosequencing analysis indicated that bacterial community structure of all influents was similar. Also the bacterial community of all CAS bioreactors was similar. Bacterial community structure of A-stage bioreactors showed a more case-specific pattern. A core of genera was consistently found for all influents, all CAS bioreactors and all A-stage bioreactors, respectively, showing that different geographical locations in The Netherlands and Spain did not affect the functional bacterial communities in these technologies. The ecological roles of these bacteria were discussed. Influents and A-stage bioreactors shared several core genera, while none of these were shared with CAS bioreactors communities. This difference is thought to reside in the different operational conditions of the two technologies. This study shows that bacterial community structure of CAS and A-stage bioreactors are mostly driven by solids retention time (SRT) and hydraulic retention time (HRT), as suggested by multivariate redundancy analysis.