Modeling of two-stage anaerobic onsite wastewater sanitation system to predict effluent soluble chemical oxygen demand through machine learning.

The present research aims to predict effluent soluble chemical oxygen demand (SCOD) in anaerobic digestion (AD) process using machine-learning based approach. Anaerobic digestion is a highly sensitive process and depends upon several environmental and operational factors, such as temperature, flow, and load. Therefore, predicting output characteristics using modeling is important not only for process monitoring and control, but also to reduce the operating cost of the treatment plant. It is difficult to predict COD in a real time mode, so it is better to use Complex Mathematical Modeling (CMM) for simulating AD process and forecasting output parameters. Therefore, different Machine Learning algorithms, such as Linear Regression, Decision Tree, Random Forest and Artificial Neural Networks, have been used for predicting effluent SCOD using data acquired from in situ anaerobic wastewater treatment system. The result of the predicted data using different algorithms were compared with experimental data of anaerobic system. It was observed that the Artificial Neural Networks is the most effective simulation technique that correlated with the experimental data with the mean absolute percentage error of 10.63 and R2 score of 0.96. This research proposes an efficient and reliable integrated modeling method for early prediction of the water quality in wastewater treatment.

Reduced sludge growth at high bulk liquor dissolved oxygen induced by increased secondary cell maintenance.

Sludge reduction by physico-chemical methods results in the buildup of chemicals, which may require further treatment. Owing these reasons various biologically sustainable methods of sludge reduction including the application of high oxygenation have been successfully tested. Experiments on actual sewage in two lab-scale sequencing batch reactors (SBRs) were conducted under normal (1.5-2.5 mgDO/L) and high dissolved oxygen (DO) (HDO: 3-6.5 mgDO/L) regimes. It was observed that microorganism allocated substrate between maintenance and growth in the form of maintenance coefficient. Which could be induced by endogenous respiration owing to high solids retention time (SRT), predation on bacteria, chemical toxicity, adverse environment, and viral attack on bacteria. The wastewater treatment process may experience one or more maintenance inducing factors; nevertheless, high SRT and prevailing environmental conditions are imminent and thus considered as primary maintenance (mp), while remaining are classified as secondary maintenance (ms). Average yield coefficient reduction at HDO was 32.7% and 28.2% compared to stoichiometric and at normal DO, respectively. The observed primary and secondary maintenance was 0.11gCOD/gVSS.d (±0.01) at an SRT of 25.2 d (±2.0) and 0.096 g 0.1 gCOD/gVSS.d (±0.045) at an SRT of 24.2 d (±3.6d), respectively. The results obtained under the study are not as precise as on pure culture and defined substrate, nevertheless, it gives an idea that how stress factors inducing maintenance need to be addressed more seriously and objectively while managing our efforts on sludge reduction.

Effect of physical property of supporting media and variable hydraulic loading on hydraulic characteristics of advanced onsite wastewater treatment system.

A laboratory-scale study was carried out to investigate the effects of physical properties of the supporting media and variable hydraulic shock loads on the hydraulic characteristics of an advanced onsite wastewater treatment system. The system consisted of two upflow anaerobic reactors (a septic tank and an anaerobic filter) accommodated within a single unit. The study was divided into three phases on the basis of three different supporting media (Aqwise carriers, corrugated ring and baked clay) used in the anaerobic filter. Hydraulic loadings were based on peak flow factor (PFF), varying from one to six, to simulate the actual conditions during onsite wastewater treatment. Hydraulic characteristics of the system were identified on the basis of residence time distribution analyses. The system showed a very good hydraulic efficiency, between 0.86 and 0.93, with the media of highest porosity at the hydraulic loading of PFF≤4. At the higher hydraulic loading of PFF 6 also, an appreciable hydraulic efficiency of 0.74 was observed. The system also showed good chemical oxygen demand and total suspended solids removal efficiency of 80.5% and 82.3%, respectively at the higher hydraulic loading of PFF 6. Plug-flow dispersion model was found to be the most appropriate one to describe the mixing pattern of the system, with different supporting media at variable loading, during the tracer study.

Modified septic tank-anaerobic filter unit as a two-stage onsite domestic wastewater treatment system.

This study demonstrates the performance evaluation of a uniquely designed two-stage system for onsite treatment of domestic wastewater. The system consisted of two upflow anaerobic bioreactors, a modified septic tank followed by an upflow anaerobic filter, accommodated within a single cylindrical unit. The system was started up without inoculation at 24 h hydraulic retention time (HRT). It achieved a steady-state condition after 120 days. The system was observed to be remarkably efficient in removing pollutants during steady-state condition with the average removal efficiency of 88.6 +/- 3.7% for chemical oxygen demand, 86.3 +/- 4.9% for biochemical oxygen demand and 91.2 +/- 9.7% for total suspended solids. The microbial analysis revealed a high reduction (>90%) capacity of the system for indicator organism and pathogens. It also showed a very good endurance against imposed hydraulic shock load. Tracer study showed that the flow pattern was close to plug flow reactor. Mean HRT was also found to be close to the designed value.