Enter Note Done

Media

Decrease font sizeIncrease font size

Carbon and energy management in high-rate contact-stabilization through dynamic oxygen uptake rate control


Abstract number: 1195

Primary Author
Khoa Nam Ngo, DC Water

Co-Author
Patrexia Tampon, DC Water
Haydee De Clippelier, D. C. Water
Sudhir Murthy, DC Water
Belinda Sturm, University of Kansas
Tim Van Winckel, DC Water
Charles Bott, HRSD
Bernhard Wett, ARAconsult
Ahmed Al-Omari, Brown and Caldwell
Siegfried Vlaeminck, University of Antwerp
Arash Massoudieh, Catholic University of America

Speaker
Patrexia Tampon, DC Water

Purpose:
Introduction Combined with carbon-efficient nutrient removal systems), high-rate activated sludge systems (HRAS) provide a major pathway towards energy neutrality as they allow for the recovery of 50-60% of wastewater COD through anaerobic digestion into energy. The A-stage of the AB-process is a HRAS process because the extremely short SRT (0.3-0.5 days) combined with high loading (> 2 kg COD kg VSS-1 d-1) minimizes oxidation and maximizes sorption onto sludge (Böhnke et al., 1998; De Graaff & Roest, 2012). However, these operational conditions can lead to a very dynamic process with a high uncertainty on the clarifier performance and thus amount of net sludge captured. Bioflocculation has known to deteriorate at shorter SRT and when the food to microbe (F/M) rate drops below an (unknown) critical threshold (Rahman et al., 2016). Contact-stabilization was able to mitigate this issue by imposing a feast-famine regime with contact stabilization (CS), thus triggering an effective degree of bioflocculation (Rahman et al., 2016). However, the ideal operational SRT for contact-stabilization was higher than an A-stage, sacrificing potential redirection. Finding and controlling on a balance between the maximization of carbon redirection and preservation of stable carbon capture is imperative for the efficacy of the high-rate process. Although the SRT target of 0.2 days was often suggested for A-stage systems (De Graaff & Roest, 2012), carbon redirection and carbon capture resulted in high fluctuations of energy demand within daily and seasonal variability. The use of OUR as a control variable for management of sludge wasting rather than SRT or MLSS targets was shown to result in more direct control of energy input and thus COD oxidation management, which further increases the energy efficiency of plants (Van Winckel et al., 2018). When the OUR setpoint was chosen correctly and bioflocculation was managed, controlling on OUR resulted in stable COD redirection and capture. However, given that the loading rate and environmental conditions will dictate the biomass inventory and degree of bioflocculation, these parameters will fluctuate dramatically even at stable OUR. Thus, the optimal OUR setpoint will differ in practise and will depend on the broader operational conditions. Bioflocculation limitation has been observed as a result of (i) decreased food to microbe (F/M) rates, leading to decreased EPS production rates (Rahman et al., 2016) and (ii) limited MLSS concentration, decreasing the successful formation of flocs in the clarifier (Mancell-Egala et al., 2017). This study aimed to develop bioflocculation limitation indicators based on online OUR measurements in contactor and stabilizer zones. The bioflocculation limitation indicators were developed based on 600 days of HRAS pilot runs under different OUR and bioflocculation conditions. The dynamic OUR control proposed in this study mitigates risks of bio-flocculation limitations automatically and will provide stable effluent composition suitable for downstream. Materials and Methods A high-rate CS pilot was fed with effluent from a full-scale chemically enhanced primary treatment, and operated under diurnal concentration patters and temperature changes. The OUR was measured ex situ based on the slope of declining DO levels and fed into dynamic OUR control using MATLAB Simulink and PID control. The OUR ratio between online OUR measurement of contactor and stabilizer was used to identify the loading limitation (lower limit) and MLSS limitation (higher limit). The wasting pump rate was controlled to maintain an OUR setpoint in the contactor (Fig. 3). Results & Discussion The OUR control was initially operated with a fixed setpoint range between 22-24 mg O2/L/h. The tuned control was able to achieve a stable COD redirection (56 ± 10%) and COD capture (42 ± 8%) (Figure 1A). However, the MLSS dropped from 350 to 145 mg TSS/L leading clarifier TSS capture from 92 to 57 %. This ultimately led to a loss in bioflocculation, losing all carbon capture. To avoid this, constraints must be put on the controller. A feast-famine regime has been proposed as the main driver for EPS production when the feed COD concentration is low (figure 1D) (Rahman et al., 2017). Therefore, the ratio of OUR in the contactor (feast-zone) to the OUR in the stabilizer (famine-zone) (figure 1B) was identified as an indicator for EPS response, and thus by extension for carbon capture. To detect bioflocculation limitations caused by low inventory, the MLSS in the contactor was compared against the OUR ratio (Figure 2A). The minimum value of measured TOF (describe collision efficiency) in the HRAS system at Blue Plains was around 300 mg TSS/L (Ngo al et., 2019. in preparation). Based on 340 datapoints used for this analysis, when the OUR ratio exceeded 1.2, the probability of achieving good MLSS went down from 62 to 35% (Figure 2 C & 2D). That indicated that the biomass would be too limited to achieve efficient TSS capture. Therefore, we can conclude that the OUR setpoint in Figure 1A was too low, leading to an unavoidable decline in MLSS in the reactor and loss of bioflocculation when the environmental conditions changed. Operation under increased OUR setpoints would be needed when the OUR ratio increases above 1.2 to compensate for MLSS-induced bioflocculation limitations. Loading limitation commonly caused by rain events and leads to a decrease in F/M-rate, resulting in a decreased EPS production. Given that 1.7 kg COD/m3/d was a quarter of volumetric loading rate (Figure 2E) of the entire data set, a lower number could indicate a rain event. A low OUR ratio can be attributed to a decrease in feast-famine, caused by a decrease in load. Indeed, the probability of measuring an volumetric loading rate higher than 1.7 kg COD/m3/d increased significantly from 64 to 83% when the OUR ratios were above 1.1 (Figure 2F & 2G) which would be the lower OUR ratio limit to detect rain events and thus increased risk of bioflocculation loss due to loading limitations. The OUR setpoint in the contactor will dynamically change depending on the OUR ratio measured in the reactor (Figure 3). When the OUR ratio is above the MLSS limitation limit (1.2), the MLSS limit is subtracted from the current OUR ratio, leading to a positive delta OUR. Therefore, the OUR setpoint will be pushed up to decrease wasting thus accommodating for more growth and oxidation, which will result in an increase in MLSS concentration. A similar increase in OUR setpoint will occur when the OUR ratio drops below 1, indicating loading limitations. When the OUR ratio is between 1 and 1.2, and bioflocculation is optimal, OUR setpoints will be pushed down to minimize energy use and maximize carbon capture (target = 42% C capture). This control strategy has been running for the past 4 weeks, and detailed steady-state data will be provided in the full paper.

Results:
Böhnke, B., Schulze-Rettner, R., Zuckut, S.W. 1998. Cost-effective reduction of high-strength wastewater by adsorption-based activated sludge technology. Water engineering & Management, 145(12), 31-34. De Graaff, M., Roest, K. 2012. Inventarisatie van AB-systemen - optimale procescondities in de A-trap. KWR. Mancell-Egala, W., De Clippeleir, H., Su, C., Takacs, I., Novak, J.T., Murthy, S.N. 2017. Novel Stokesian Metrics that Quantify Collision Efficiency, Floc Strength, and Discrete Settling Behavior. Water Environ Res, 89(7), 586-597. Rahman, A., Meerburg, F.A., Ravadagundhi, S., Wett, B., Jimenez, J., Bott, C., Al-Omari, A., Riffat, R., Murthy, S., De Clippeleir, H. 2016. Bioflocculation management through high-rate contact-stabilization: A promising technology to recover organic carbon from low-strength wastewater. Water Res, 104, 485-496. Van Winckel, T., Olagunju, O., Sturm, B., Vlaeminck, S.E., Bott, C., Wett, B., Al-Omari, A., Murthy, S., De Clippeleir, H. 2018. Oxygen uptake rate control in high-rate contact stabilization for effective carbon management. in: IWA Nutrient Removal and Recovery Conference. Brisbane, Australia.

Keywords
Carbon (BOD) Removal
Energy Conservation/Management
Research and Development
Resource Recovery
Treatment: Secondary (Does not include Nutrients)
Water, Food, Energy Nexus