Prediction of biological nutrients removal in full-scale wastewater treatment plants using H2O automated machine learning and back propagation artificial neural network model: Optimization and comparison.

The effective control of total nitrogen (ETN) and total phosphorus (ETP) in effluent is challenging for wastewater treatment plants (WWTPs). In this work, automated machine learning (AutoML) (mean square error = 0.4200 âˆ¼ 3.8245, R2 = 0.5699 âˆ¼ 0.6219) and back propagation artificial neural network (BPANN) model (mean square error = 0.0012 âˆ¼ 6.9067, R2 = 0.4326 âˆ¼ 0.8908) were used to predict and analyze biological nutrients removal in full-scale WWTPs. Interestingly, BPANN model presented high prediction performance and general applicability for WWTPs with different biological treatment units. However, the AutoML candidate models were more interpretable, and the results showed that electricity carbon emission dominated the prediction. Meanwhile, increasing data volume and types of WWTP hardly affected the interpretable results, demonstrating its wide applicability. This study demonstrated the validity and the specific advantages of predicting ETN and ETP using H2O AutoML and BPANN model, which provided guidance on the prediction and improvement of biological nutrients removal in WWTPs.

The ASM2d model with two-step nitrification can better simulate biological nutrient removal systems enriched with complete ammonia oxidizing bacteria (comammox Nitrospira).

The discovery of comammox Nitrospira, a complete ammonia-oxidizing microorganism belonging to the genus Nitrospira, has brought new insights into the nitrification process in wastewater treatment plants (WWTPs). The applicability of Activated Sludge Model No. 2 d with one-step nitrification (ASM2d-OSN) or two-step nitrification (ASM2d-TSN) for the simulation of the biological nutrient removal (BNR) processes of a full-scale WWTP in the presence of comammox Nitrospira was studied. Microbial analysis and kinetic parameter measurements showed comammox Nitrospira was enriched in the BNR system operated under low dissolved oxygen (DO) and long sludge retention time (SRT). The relative abundance of Nitrospira under the conditions of stage I (DO = 0.5 mg/L, SRT = 60 d) was about twice of that under stage II conditions (DO = 4.0 mg/L, SRT = 26 d), and the copy number of the comammox amoA gene for stage I was 33 times higher than that for stage II. Compared to the ASM2d-OSN model, the ASM2d-TSN model simulated the performance of the WWTP under stage I conditions better, and the Theil inequality coefficient values of all the tested water quality parameters were lower than using ASM2d-OSN. These results indicate that an ASM2d model with two-step nitrification is a better choice for the simulation of WWTPs with the presence of comammox.

Distribution patterns of microbial community and functional characteristics in full-scale wastewater treatment plants: Focusing on the influent types.

The impacts of the influent type in wastewater treatment plants (WWTPs) on the distribution patterns of the microbial community and functional characteristics were investigated. The obtained results indicated that the influent types exhibited evident influences on the microbial distribution patterns. The diversity and richness of functional microbes in HI-WWTP (with a ratio of >30% industrial wastewater in influents) were evidently decreased compared with those in HM- (with 70-90% municipal wastewater in influents) and M-WWTPs (with >90% municipal wastewater in influents). The core functional bacteria included denitrifiers, anaerobic fermentation bacteria (AFB), organic degrading bacteria (ODB), phosphorus accumulating organisms (PAO) and nitrite oxidizing bacteria (NOB), but they exhibited distinct abundances in WWTPs receiving different categories of wastewater. The denitrifiers in HI-WWTPs was 15.6-32.5% higher than that in other WWTPs, while PAOs had higher abundances in M - and HI-WWTPs (28.9% and 39.3%, respectively) compared with HM-WWTPs. Clear co-occurrence relationships were found among the main functional microbes with similar metabolic characteristics. Moreover, information on functional genes related to carbon, nitrogen and phosphorus metabolism, which is closely associated with pollutant removal efficiency, was obtained. M-WWTPs had higher abundances of genetic expressions for organic matters degradation (i.e. amino acid (10.42%) and carbohydrate (9.86%) metabolisms). Nar, Nir and Nor showed lowest abundances in HM-WWTPs, causing the low nitrogen removal (63.04-65.79%). However, influent type had little effect on genetic expression related with phosphorus removal. This work provided new insights into the interrelationship among bacterial co-occurrence, microbial activity and pollutant removal in WWTPs with different influent types.

Integrated data-driven strategy to optimize the processes configuration for full-scale wastewater treatment plant predesign.

Wastewater treatment plants (WWTPs) play an irreplaceable role in eliminating pollutants from domestic and industrial wastewater and contribute to water recycling. Nowadays, the selection of processes configuration of WWTPs mainly depends on the local wastewater treatment standards and the experience of wastewater engineers rather than an intelligent data-driven strategy. In this study, an integrated data-driven strategy consisting of t-distributed stochastic neighbor embedding (t-SNE) and deep neural networks (DNNs) is proposed for optimizing the processes configuration of full-scale WWTP predesign. A large dataset with 14,647 samples collected from 10 full-scale WWTPs with distinct treatment processes is clustered by the t-SNE method based on the influent characteristics, and four meaningful clusters (Clusters I-IV) are identified for the subsequent development of DNN classification models. All four DNN models achieve acceptable classification accuracy (>0.8975) and the maximal testing accuracy is 0.9505. The DNN models are capable of finding the optimized processes configuration of WWTPs under target scenarios. Our results highlight the strength of combining the t-SNE and the DNN models to utilize the relationships between key parameters and processes configuration of WWTPs, and help engineers predesign WWTPs with the optimal processes configuration.

Fate of typical endocrine active compounds in full-scale wastewater treatment plants: Distribution, removal efficiency and potential risks.

In this study, the distribution, removal efficiency, and potential risks of 9 typical endocrine active compounds (EACs) in two full-scale wastewater treatment plants (WWTPs) were investigated. The EAC concentrations ranged from 0.2 to 7394.2 ng/L in influents. The source of influents was a critical factor in determining the EAC levels. EACs were primarily removed in the secondary biological processing units, with removal efficiencies fluctuating from 13.7% to 98%. The biological treatment processes and operating parameters (i.e., HRT and SRT) influenced the EAC removal efficiency. Bisphenol A (BPA) and estriol were mainly removed by biodegradation, while antidepressants were primarily eliminated by sludge adsorption according to the distribution patterns and mass flow of EACs in WWTPs. Novosphingobium, Saprospiraceae, etc. were the core functional bacteria for EAC biodegradation. In addition, sertraline in effluents and dewatered sludge may pose medium environmental risks, while the other EACs pose low environmental risks.

Bacteria from the Genus Arcobacter Are Abundant in Effluent from Wastewater Treatment Plants.

Pathogenic bacteria in wastewater are generally considered to be efficiently removed in biological wastewater treatment plants. This understanding is almost solely based on culture-based control measures, and here we show, by applying culture-independent methods, that the removal of species in the genus Arcobacter was less effective than for many other abundant genera in the influent wastewater. Arcobacter was one of the most abundant genera in influent wastewater at 14 municipal wastewater treatment plants and was also abundant in the "clean" effluent from all the plants, reaching up to 30% of all bacteria as analyzed by 16S rRNA gene amplicon sequencing. Metagenomic analyses, culturing, genome sequencing of Arcobacter isolates, and visualization by fluorescent in situ hybridization (FISH) confirmed the presence of the human-pathogenic Arcobacter cryaerophilus and A. butzleri in both influent and effluent. The main reason for the high relative abundance in the effluent was probably that Arcobacter cells, compared to those of other abundant genera in the influent, did not flocculate and attach well to the activated sludge flocs, leaving a relatively large fraction dispersed in the water phase. The study shows there is an urgent need for new standardized culture-independent measurements of pathogens in effluent wastewaters, e.g., amplicon sequencing, and an investigation of the problem on a global scale to quantify the risk for humans and livestock.IMPORTANCE The genus Arcobacter was unexpectedly abundant in the effluent from 14 Danish wastewater treatment plants treating municipal wastewater, and the species included the human-pathogenic A. cryaerophilus and A. butzleri Recent studies have shown that Arcobacter is common in wastewater worldwide, so the study indicates that discharge of members of the genus Arcobacter may be a global problem, and further studies are needed to quantify the risk and potentially minimize the discharge. The study also shows that culture-based analyses are insufficient for proper effluent quality control, and new standardized culture-independent measurements of effluent quality encompassing most pathogens should be considered.

Correlations of nitrogen removal and core functional genera in full-scale wastewater treatment plants: Influences of different treatment processes and influent characteristics.

The denitrification process is crucial for biological nitrogen removal in wastewater treatment plants (WWTPs). In this study, the nitrogen removal efficiency in full-scale WWTPs with different treatment processes and influent characteristics was investigated. The results indicated that the average total nitrogen removal rate (NRR) and denitrification rate in the A/O or A2/O systems were 67.5% and 2.08 mg N h-1 gMLVSS-1, respectively. However, cyclic activated sludge systems (CASSs) showed more efficient nitrogen removal with an average NRR and denitrification rate of 79.6% and 9.89 mg N h-1 gMLVSS-1, respectively. The microbial communities in WWTPs with similar influent compositions were similar and mainly shaped by BOD5. Candidatus Competibacter, Caldilineaceae and Anaerolineaceae were the functional genera closely associated with nitrogen removal based on high-throughput sequencing and correlation analysis. This study provides new insights into the regulation and amelioration of full-scale WWTPs to meet the increasingly stringent nitrogen discharge standard.

Insight into mature biofilm quorum sensing in full-scale wastewater treatment plants.

Quorum sensing (QS) has been thoroughly investigated during initial biofilm formation stages, while the role of QS in mature biofilms has received little research attention. This study assessed QS in 22 biofilm samples from full-scale wastewater treatment plants in China. Results showed that the concentration of acyl-homoserine lactones (AHLs) in various biofilm bound forms, ranged from 15.63 to 609.76 ng/g. The highest concentration of AHLs was found in the tightly bound biofilm fraction, while the lowest concentrations were observed in the surface biofilm fraction. Environmental variables, C/N ratio and temperature, were found to be significant factors influencing biofilm AHL distribution (p < 0.01). Higher C/N ratios (ranging from 3 to 12) and low temperatures contributed to the higher concentration of AHLs in biofilms. Dominant AHLs (C10-HSL and C12-HSL) were significantly associated with biofilm activity (R2 = 0.98/0.97, p < 0.05), with the tightly bound biofilm fraction (TB-biofilm) presenting the highest activity (ATP concentration). Biofilm aging and re-formation processes were more active in the surface biofilm layer (S-biofilm), while the stable structure of the TB-biofilm layer which is attached to the surface of bio-carriers ensures high biofilm activity. This study furthers our understanding of the roles of AHLs in the regulation of mature biofilm activities.