Towards good modelling practice for parallel hybrid models for wastewater treatment processes.

This study explores various approaches to formulating a parallel hybrid model (HM) for Water and Resource Recovery Facilities (WRRFs) merging a mechanistic and a data-driven model. In the study, the HM is constructed by training a neural network (NN) on the residual of the mechanistic model for effluent nitrate. In an initial experiment using the Benchmark Simulation Model no. 1, a parallel HM effectively addressed limitations in the mechanistic model's representation of autotrophic bacteria growth and the data-driven model's incapability to extrapolate. Next, different versions of a parallel HM of a large pilot-scale WRRF are constructed, using different calibration/training datasets and different versions of the mechanistic model to investigate the balance between the calibration effort for the mechanistic model and the compensation by the NN component. The HM can improve predictions compared to the mechanistic model. Training the NN on an independent validation dataset produced better results than on the calibration dataset. Interestingly, the best performance is achieved for the HM based on a mechanistic model using default (uncalibrated) parameters. Both long short-term memory (LSTM) and convolutional neural network (CNN) are tested as data-driven components, with a CNN HM (root-mean-squared error (RMSE) = 1.58 mg NO3-N/L) outperforming an LSTM HM (RMSE = 4.17 mg NO3-N/L).

A comprehensive, open-source data model for wastewater-based epidemiology.

The recent SARS-COV-2 pandemic has sparked the adoption of wastewater-based epidemiology (WBE) as a low-cost way to monitor the health of populations. In parallel, the pandemic has encouraged researchers to openly share their data to serve the public better and accelerate science. However, environmental surveillance data are highly dependent on context and are difficult to interpret meaningfully across sites. This paper presents the second iteration of the Public Health Environmental Surveillance Open Data Model (PHES-ODM), an open-source dictionary and set of data tools to enhance the interoperability of environmental surveillance data and enable the storage of contextual (meta)data. The data model describes how to store environmental surveillance program data, metadata about measurements taken on various specimens (water, air, surfaces, sites, populations) and data about measurement protocols. The model provides software tools that support the collection and use of PHES-ODM formatted data, including performing PCR calculations and data validation, recording data into input templates, generating wide tables for analysis, and producing SQL database definitions. Fully open-source and already adopted by institutions in Canada, the European Union, and other countries, the PHES-ODM provides a path forward for creating robust, interoperable, open datasets for environmental public health surveillance for SARS-CoV-2 and beyond.

Assessing the equivalence of WRRF regulations using dynamic model simulations.

A wide diversity of regulatory practices for wastewater resource recovery plants exists throughout the world. This contribution aims to highlight the implications of choosing particular permitting structures and investigate the equivalence of effluent standards in terms of limit values and compliance assessment specifications. These factors heavily affect the true performance that a treatment plant has to attain and thus the required plant capacity and operation. The dynamic simulations executed in this work, based on a realistic case study and three selected permits from China, The Netherlands and the USA, show the impact of certain compliance specifications like sampling frequency, averaging and tolerable permit exceedances leading to differences in the required design capacity of more than 250% for the same wastewater to be treated. The results also reveal clear differences between permits in their capacity to handle excess variability. The latter is important to avoid overdesign, i.e., when further investment in treatment capacity would result only in marginal effluent quality gains, as well as to create a safe space for testing innovative technologies or ways of operation that might otherwise trigger compliance issues.

An essential tool for WRRF modelling: a realistic and complete influent generator for flow rate and water quality based on data-driven methods.

Modelling, automation, and control are widely used for water resource recovery facility (WRRF) optimization. An influent generator (IG) is a model, aiming to provide the flowrate and pollutant concentration dynamics at the inlet of a WRRF for a range of modelling applications. In this study, a new data-driven IG model is proposed, only using routine data and weather information, and without need for any additional data collection. The model is constructed by an artificial neural network (ANN) and completed with a multivariate regression to generate time series for certain pollutants. The model is able to generate flowrate and quality data (TSS, COD, and nutrients) at different time scales and resolutions (daily or hourly), depending on various user objectives. The model performance is analyzed by a series of statistical criteria. It is shown that the model can generate a very reliable dataset for different model applications.

An influent generator for WRRF design and operation based on a recurrent neural network with multi-objective optimization using a genetic algorithm.

Nowadays, modelling, automation and control are widely used for Water Resource Recovery Facilities (WRRF) upgrading and optimization. Influent generator (IG) models are used to provide relevant input time series for dynamic WRRF simulations used in these applications. Current IG models found in literature are calibrated on the basis of a single performance criterion, such as the mean percentage error or the root mean square error. This results in the IG being adequate on average but with a lack of representativeness of, for instance, the observed temporal variability of the dataset. However, adequately capturing influent variability may be important for certain types of WRRF optimization, e.g., reaction to peak loads, control system performance evaluation, etc. Therefore, in this study, a data-driven IG model is developed based on the long short-term memory (LSTM) recurrent neural network and is optimized by a multi-objective genetic algorithm for both mean percentage error and variability. Hence, the influent generator model is able to generate a time series with a probability distribution that better represents reality, thus giving a better influent description for WRRF design and operation. To further increase the variability of the generated time series and in this way approximate the true variability better, the model is extended with a random walk process.

The transition of WRRF models to digital twin applications.

Digital Twins (DTs) are on the rise as innovative, powerful technologies to harness the power of digitalisation in the WRRF sector. The lack of consensus and understanding when it comes to the definition, perceived benefits and technological needs of DTs is hampering their widespread development and application. Transitioning from traditional WRRF modelling practice into DT applications raises a number of important questions: When is a model's predictive power acceptable for a DT? Which modelling frameworks are most suited for DT applications? Which data structures are needed to efficiently feed data to a DT? How do we keep the DT up to date and relevant? Who will be the main users of DTs and how to get them involved? How do DTs push the water sector to evolve? This paper provides an overview of the state-of-the-art, challenges, good practices, development needs and transformative capacity of DTs for WRRF applications.

Mainstream short-cut N removal modelling: current status and perspectives.

This work gives an overview of the state-of-the-art in modelling of short-cut processes for nitrogen removal in mainstream wastewater treatment and presents future perspectives for directing research efforts in line with the needs of practice. The modelling status for deammonification (i.e., anammox-based) and nitrite-shunt processes is presented with its challenges and limitations. The importance of mathematical models for considering N2O emissions in the design and operation of short-cut nitrogen removal processes is considered as well. Modelling goals and potential benefits are presented and the needs for new and more advanced approaches are identified. Overall, this contribution presents how existing and future mathematical models can accelerate successful full-scale mainstream short-cut nitrogen removal applications.

Influence of MBBR carrier geometrical properties and biofilm thickness restraint on biofilm properties, effluent particle size distribution, settling velocity distribution, and settling behaviour.

The relatively poor settling characteristics of particles produced in moving bed biofilm reactor (MBBR) outline the importance of developing a fundamental understanding of the characterization and settleability of MBBR-produced solids. The influence of carrier geometric properties and different levels of biofilm thickness on biofilm characteristics, solids production, particle size distribution (PSD), and particle settling velocity distribution (PSVD) is evaluated in this study. The analytical ViCAs method is applied to the MBBR effluent to assess the distribution of particle settling velocities. This method is combined with microscopy imaging to relate particle size distribution to settling velocity. Three conventionally loaded MBBR systems are studied at a similar loading rate of 6.0 g/(m2 •day) and with different carrier types. The AnoxK™ K5 carrier, a commonly used carrier, is compared to so-called thickness-restraint carriers, AnoxK™ Z-carriers that are newly designed carriers to limit the biofilm thickness. Moreover, two levels of biofilm thickness, 200 µm and 400 µm, are studied using AnoxK™ Z-200 and Z-400 carriers. Statistical analysis confirms that K5 carriers demonstrated a significantly different biofilm mass, thickness, and density, in addition to distinct trends in PSD and PSVD in comparison with Z-carriers. However, in comparison of thickness-restraint carriers, Z-200 carrier results did not vary significantly compared to the Z-400 carrier. The K5 carriers showed the lowest production of suspended solids (0.7 ± 0.3 g-TSS/day), thickest biofilm (281.1 ± 8.7 µm) and lowest biofilm density (65.0 ± 1.5 kg/m3). The K5 effluent solids also showed enhanced settling behaviour, consisting of larger particles with faster settling velocities.

Dynamic grit chamber modelling: dealing with particle settling velocity distributions.

Grit chambers are meant to reduce the impact of inorganic particles on equipment and processes downstream. Despite their important role, characterization and modelling studies of these process units are scarce, leading to a lack of knowledge and suboptimal operation. Thus, this study presents the first dynamic model, based on mass balances and particle settling velocity distributions, for use in a water resource recovery facility (WRRF) simulator for design and optimization of grit removal units.

A critical review of the data pipeline: how wastewater system operation flows from data to intelligence.

Faced with an unprecedented amount of data coming from evermore ubiquitous sensors, the wastewater treatment community has been hard at work to develop new monitoring systems, models and controllers to bridge the gap between current practice and data-driven, smart water systems. For additional sensor data and models to have an appreciable impact, however, they must be relevant enough to be looked at by busy water professionals; be clear enough to be understood; be reliable enough to be believed and be convincing enough to be acted upon. Failure to attain any one of those aspects can be a fatal blow to the adoption of even the most promising new measurement technology. This review paper examines the state-of-the-art in the transformation of raw data into actionable insight, specifically for water resource recovery facility (WRRF) operation. Sources of difficulties found along the way are pinpointed, while also exploring possible paths towards improving the value of collected data for all stakeholders, i.e., all personnel that have a stake in the good and efficient operation of a WRRF.

How Urban Storm- and Wastewater Management Prepares for Emerging Opportunities and Threats: Digital Transformation, Ubiquitous Sensing, New Data Sources, and Beyond - A Horizon Scan.

Ubiquitous sensing will create many opportunities and threats for urban water management, which are only poorly understood today. To identify the most relevant trends, we conducted a horizon scan regarding how ubiquitous sensing will shape the future of urban drainage and wastewater management. Our survey of the international urban water community received an active response from both the academics and the professionals from the water industry. The analysis of the responses demonstrates that emerging topics for urban water will often involve experts from different communities, including aquatic ecologists, urban water system engineers and managers, as well as information and communications technology professionals and computer scientists. Activities in topics that are identified as novel will either require (i) cross-disciplinary training, such as importing new developments from the IT sector, or (ii) research in new areas for urban water specialists, for example, to help solve open questions in aquatic ecology. These results are, therefore, a call for interdisciplinary research beyond our own discipline. They also demonstrate that the water management community is not yet prepared for the digital transformation, where we will experience a data demand, i.e. a "pull" of urban water data into external services. The results suggest that a lot remains to be done to harvest the upcoming opportunities. Horizon scanning should be repeated on a routine basis, under the umbrella of an experienced polling organization.

Tanks in series versus compartmental model configuration: considering hydrodynamics helps in parameter estimation for an N2O model.

The choice of the spatial submodel of a water resource recovery facility (WRRF) model should be one of the primary concerns in WRRF modelling. However, currently used mechanistic models are limited by an over-simplified representation of local conditions. This is illustrated by the general difficulties in calibrating the latest N2O models and the large variability in parameter values reported in the literature. The use of compartmental model (CM) developed on the basis of accurate hydrodynamic studies using computational fluid dynamics (CFD) can take into account local conditions and recirculation patterns in the activated sludge tanks that are important with respect to the modelling objective. The conventional tanks in series (TIS) configuration does not allow this. The aim of the present work is to compare the capabilities of two model layouts (CM and TIS) in defining a realistic domain of parameter values representing the same full-scale plant. A model performance evaluation method is proposed to identify the good operational domain of each parameter in the two layouts. Already when evaluating for steady state, the CM was found to provide better defined parameter ranges than TIS. Dynamic simulations further confirmed the CM's capability to work in a more realistic parameter domain, avoiding unnecessary calibration to compensate for flaws in the spatial submodel.

The future of WRRF modelling - outlook and challenges.

The wastewater industry is currently facing dramatic changes, shifting away from energy-intensive wastewater treatment towards low-energy, sustainable technologies capable of achieving energy positive operation and resource recovery. The latter will shift the focus of the wastewater industry to how one could manage and extract resources from the wastewater, as opposed to the conventional paradigm of treatment. Debatable questions arise: can the more complex models be calibrated, or will additional unknowns be introduced? After almost 30 years using well-known International Water Association (IWA) models, should the community move to other components, processes, or model structures like 'black box' models, computational fluid dynamics techniques, etc.? Can new data sources - e.g. on-line sensor data, chemical and molecular analyses, new analytical techniques, off-gas analysis - keep up with the increasing process complexity? Are different methods for data management, data reconciliation, and fault detection mature enough for coping with such a large amount of information? Are the available calibration techniques able to cope with such complex models? This paper describes the thoughts and opinions collected during the closing session of the 6th IWA/WEF Water Resource Recovery Modelling Seminar 2018. It presents a concerted and collective effort by individuals from many different sectors of the wastewater industry to offer past and present insights, as well as an outlook into the future of wastewater modelling.

A framework for good biofilm reactor modeling practice (GBRMP).

A researcher or practitioner can employ a biofilm model to gain insight into what controls the performance of a biofilm process and for optimizing its performance. While a wide range of biofilm-modeling platforms is available, a good strategy is to choose the simplest model that includes sufficient components and processes to address the modeling goal. In most cases, a one-dimensional biofilm model provides the best balance, and good choices can range from hand-calculation analytical solutions, simple spreadsheets, and numerical-method platforms. What is missing today is clear guidance on how to apply a biofilm model to obtain accurate and meaningful results. Here, we present a five-step framework for good biofilm reactor modeling practice (GBRMP). The first four steps are (1) obtain information on the biofilm reactor system, (2) characterize the influent, (3) choose the plant and biofilm model, and (4) define the conversion processes. Each step demands that the model user understands the important components and processes in the system, one of the main benefits of doing biofilm modeling. The fifth step is to calibrate and validate the model: System-specific model parameters are adjusted within reasonable ranges so that model outputs match actual system performance. Calibration is not a simple 'by the numbers' process, and it requires that the modeler follows a logical hierarchy of steps. Calibration requires that the adjusted parameters remain within realistic ranges and that the calibration process be carried out in an iterative manner. Once each of steps 1 through 5 is completed satisfactorily, the calibrated model can be used for its intended purpose, such as optimizing performance, trouble-shooting poor performance, or gaining deeper understanding of what controls process performance.

Dynamic modelling of solids in a full-scale activated sludge plant preceded by CEPT as a preliminary step for micropollutant removal modelling.

The presence of micropollutants in the environment has triggered research on quantifying and predicting their fate in wastewater treatment plants (WWTPs). Since the removal of micropollutants is highly related to conventional pollutant removal and affected by hydraulics, aeration, biomass composition and solids concentration, the fate of these conventional pollutants and characteristics must be well predicted before tackling models to predict the fate of micropollutants. In light of this, the current paper presents the dynamic modelling of conventional pollutants undergoing activated sludge treatment using a limited set of additional daily composite data besides the routine data collected at a WWTP over one year. Results showed that as a basis for modelling, the removal of micropollutants, the Bürger-Diehl settler model was found to capture the actual effluent total suspended solids (TSS) concentrations more efficiently than the Takács model by explicitly modelling the overflow boundary. Results also demonstrated that particular attention must be given to characterizing incoming TSS to obtain a representative solids balance in the presence of a chemically enhanced primary treatment, which is key to predict the fate of micropollutants.

Chemically enhancing primary clarifiers: model-based development of a dosing controller and full-scale implementation.

Chemically enhanced primary treatment (CEPT) can be used to mitigate the adverse effect of wet weather flow on wastewater treatment processes. In particular, it can reduce the particulate pollution load to subsequent secondary unit processes, such as biofiltration, which may suffer from clogging by an overload of particulate matter. In this paper, a simple primary clarifier model able to take into account the effect of the addition of chemicals on particle settling is presented. Control strategies that optimize the treatment process by chemical addition were designed and tested by running simulations with this CEPT model. The most adequate control strategy in terms of treatment performance, chemicals saving, and maintenance effort was selected. Full-scale implementation of the controller was performed during the autumn of 2015, and the results obtained confirmed the behaviour of the controlled system. Practical issues related to the implementation are presented.

Concentration-driven models revisited: towards a unified framework to model settling tanks in water resource recovery facilities.

A new perspective on the modelling of settling behaviour in water resource recovery facilities is introduced. The ultimate goal is to describe in a unified way the processes taking place both in primary settling tanks (PSTs) and secondary settling tanks (SSTs) for a more detailed operation and control. First, experimental evidence is provided, pointing out distributed particle properties (such as size, shape, density, porosity, and flocculation state) as an important common source of distributed settling behaviour in different settling unit processes and throughout different settling regimes (discrete, hindered and compression settling). Subsequently, a unified model framework that considers several particle classes is proposed in order to describe distributions in settling behaviour as well as the effect of variations in particle properties on the settling process. The result is a set of partial differential equations (PDEs) that are valid from dilute concentrations, where they correspond to discrete settling, to concentrated suspensions, where they correspond to compression settling. Consequently, these PDEs model both PSTs and SSTs.

Influence of Different Sewer Biofilms on Transformation Rates of Drugs.

To estimate drug consumption more reliably, wastewater-based epidemiology would benefit from a better understanding of drug residue stability during in-sewer transport. We conducted batch experiments with real, fresh wastewater and sewer biofilms. Experimental conditions mimic small to medium-sized gravity sewers with a relevant ratio of biofilm surface area to wastewater volume (33 m2 m-3). The influences of biological, chemical, and physical processes on the transformation of 30 illicit drug and pharmaceutical residues were quantified. Rates varied among locations and over time. Three substances were not stable-that is, >20% transformation, mainly due to biological processes-at least for one type of tested biofilm for a residence time ≤2 h: amphetamine, 6-acetylcodeine, and 6-monoacetylmorphine. Cocaine, ecgonine methyl ester, norcocaine, cocaethylene, and mephedrone were mainly transformed by chemical hydrolysis and, hence, also unstable in sewers. In contrast, ketamine, norketamine, O-desmethyltramadol, diclofenac, carbamazepine, and methoxetamine were not substantially affected by in-sewer processes under all tested conditions and residence times up to 12 h. Our transformation rates include careful quantification of uncertainty and can be used to identify situations in which specific compounds are not stable. This will improve accuracy and uncertainty estimates of drug consumption when applied to the back-calculation.

Evaluation of different nitrous oxide production models with four continuous long-term wastewater treatment process data series.

Five activated sludge models describing N2O production by ammonium oxidising bacteria (AOB) were compared to four different long-term process data sets. Each model considers one of the two known N2O production pathways by AOB, namely the AOB denitrification pathway and the hydroxylamine oxidation pathway, with specific kinetic expressions. Satisfactory calibration could be obtained in most cases, but none of the models was able to describe all the N2O data obtained in the different systems with a similar parameter set. Variability of the parameters can be related to difficulties related to undescribed local concentration heterogeneities, physiological adaptation of micro-organisms, a microbial population switch, or regulation between multiple AOB pathways. This variability could be due to a dependence of the N2O production pathways on the nitrite (or free nitrous acid-FNA) concentrations and other operational conditions in different systems. This work gives an overview of the potentialities and limits of single AOB pathway models. Indicating in which condition each single pathway model is likely to explain the experimental observations, this work will also facilitate future work on models in which the two main N2O pathways active in AOB are represented together.

Population balance models: a useful complementary modelling framework for future WWTP modelling.

Population balance models (PBMs) represent a powerful modelling framework for the description of the dynamics of properties that are characterised by distributions. This distribution of properties under transient conditions has been demonstrated in many chemical engineering applications. Modelling efforts of several current and future unit processes in wastewater treatment plants could potentially benefit from this framework, especially when distributed dynamics have a significant impact on the overall unit process performance. In these cases, current models that rely on average properties cannot sufficiently capture the true behaviour and even lead to completely wrong conclusions. Examples of distributed properties are bubble size, floc size, crystal size or granule size. In these cases, PBMs can be used to develop new knowledge that can be embedded in our current models to improve their predictive capability. Hence, PBMs should be regarded as a complementary modelling framework to biokinetic models. This paper provides an overview of current applications, future potential and limitations of PBMs in the field of wastewater treatment modelling, thereby looking over the fence to other scientific disciplines.