Department of Civil & Environmental Engineering, Brunel University London, Uxbridge, UB8 3PH, UK.
Department of Biotechnology, University of Verona, Strada Le Grazie 15, 37134, Verona, Italy.
Water Res. 2020 Jul 1;178:115799. doi: 10.1016/j.watres.2020.115799. Epub 2020 Apr 10.
Data Analytics is being deployed to predict the dissolved nitrous oxide (NO) concentration in a full-scale sidestream sequence batch reactor (SBR) treating the anaerobic supernatant. On average, the NO emissions are equal to 7.6% of the NH-N load and can contribute up to 97% to the operational carbon footprint of the studied nitritation-denitritation and via-nitrite enhanced biological phosphorus removal process (SCENA). The analysis showed that average aerobic dissolved NO concentration could significantly vary under similar influent loads, dissolved oxygen (DO), pH and removal efficiencies. A combination of density-based clustering, support vector machine (SVM), and support vector regression (SVR) models were deployed to estimate the dissolved NO concentration and behaviour in the different phases of the SBR system. The results of the study reveal that the aerobic dissolved NO concentration is correlated with the drop of average aerobic conductivity rate (spearman correlation coefficient equal to 0.7), the DO (spearman correlation coefficient equal to -0.7) and the changes of conductivity between sequential cycles. Additionally, operational conditions resulting in low aerobic NO accumulation (<0.6 mg/L) were identified; step-feeding, control of initial NH concentrations and aeration duration can mitigate the NO peaks observed in the system. The NO emissions during aeration shows correlation with the stripping of accumulated NO from the previous anoxic cycle. The analysis shows that NO is always consumed after the depletion of NO during denitritation (after the "nitrite knee"). Based on these findings SVM classifiers were constructed to predict whether dissolved NO will be consumed during the anoxic and anaerobic phases and SVR models were trained to predict the NO concentration at the end of the anaerobic phase and the average dissolved NO concentration during aeration. The proposed approach accurately predicts the NO emissions as a latent parameter from other low-cost sensors that are traditionally deployed in biological batch processes.
数据分析被应用于预测全规模侧流序批式反应器（SBR）处理厌氧上清液中溶解的氧化亚氮（NO）浓度。平均而言，NO 排放相当于 NH-N 负荷的 7.6%，并对所研究的亚硝化-反硝化和通过亚硝酸盐增强生物除磷工艺（SCENA）的运营碳足迹贡献高达 97%。分析表明，在相似的进水负荷、溶解氧（DO）、pH 和去除效率下，平均好氧溶解 NO 浓度可能会显著变化。基于密度的聚类、支持向量机（SVM）和支持向量回归（SVR）模型的组合被用于估计 SBR 系统不同阶段的溶解 NO 浓度和行为。研究结果表明，好氧溶解 NO 浓度与平均好氧电导率下降（斯皮尔曼相关系数等于 0.7）、DO（斯皮尔曼相关系数等于-0.7）和连续周期之间的电导率变化相关。此外，确定了导致低好氧 NO 积累（<0.6mg/L）的操作条件；分步进料、控制初始 NH 浓度和曝气时间可以减轻系统中观察到的 NO 峰值。曝气过程中的 NO 排放与前缺氧周期中积累的 NO 的汽提有关。分析表明，在反硝化过程中（在“亚硝酸盐膝部”之后），NO 在消耗完后总是被消耗掉。基于这些发现，构建了 SVM 分类器来预测在缺氧和厌氧阶段溶解的 NO 是否会被消耗，并训练 SVR 模型来预测厌氧阶段结束时的 NO 浓度和曝气过程中的平均溶解 NO 浓度。该方法可以从传统上用于生物批量过程的其他低成本传感器中准确地预测作为潜在参数的 NO 排放。
