Search for: All records

High-flow events that significantly impact Water Resource Recovery Facility (WRRF) operations are rare, but accurately predicting these flows could improve treatment operations. Data-driven modeling approaches could be used; however, high flow events that impact operation are an infrequent occurrence, providing limited data from which to learn meaningful patterns. The performance of a statistical model (logistic regression) and two machine learning (ML) models (support vector machine and random forest) were evaluated to predict high flow events one-day-ahead to two plants located in different parts of the United States, Northern Virginia and the Gulf Coast of Texas, with combined and separate sewers, respectively. We compared baseline models (no synthetic data added) to models trained with synthetic data added from two different sampling techniques (SMOTE and ADASYN) that increased the representation of rare events in the training data. Both techniques enhanced the sample size of the very high-flow class, but ADASYN, which focused on generating synthetic samples near decision boundaries, led to greater improvements in model performance (reduced misclassification rates). Random forest combined with ADASYN achieved the best overall performance for both plants, demonstrating its robustness in identifying one-day-ahead extreme flow events to treatment plants. These results suggest that combining sampling techniques with ML has the potential to significantly improve the modeling of high-flow events at treatment plants. Our work will prove useful in building reliable predictive models that can inform management decisions needed for the better control of treatment operations. 

Membrane‐aerated biofilm reactors (MABRs) are being increasingly being implemented at full‐scale for domestic wastewater treatment and effective biofilm control is critical to their performance. This study investigated the impact of three biofilm scouring strategies on nitrogen removal performance of a pilot‐scale MABR operated in Houston, TX: (1) regular air scouring, (2) high intensity air scouring, and (3) high liquid flow scouring. Normal and high intensity air scouring regimes and a high liquid flow scour (10× baseline flow) were each tested sequentially. High NH₄⁺‐N removal efficiency of 52% in flow‐through mode was observed post‐high liquid flow scouring, which was comparable to the performance during the intense scouring regime. The absolute abundance ofamoAgene for ammonia oxidizing bacteria (AOB) increased significantly by over 200%, between pre‐ and post‐high liquid flow scouring. The energy consumption was 43% lower for the combination of high liquid flow scouring with regular air scouring as compared to the intense air scouring. This study showed that high liquid flows may be utilized as an energy‐efficient biofilm control strategy in nitrifying MABR systems. Practitioner PointsPilot‐scale MABR reactors were operated with different scouring settings: regular aeration, intense aeration, and high liquid flow.High liquid flow scouring improved nitrification efficiency, comparable to intense scouring.High liquid flow scouring selected for nitrifiers as seen by an increase in AOB quantified asamoAgene abundance.Using high liquid flow with regular aeration scouring reduces electrical energy consumption by 43% as compared to intense aeration scouring.High liquid flows may be used as an energy‐efficient biofilm control strategy to improve nitrification performance in MABR systems.