Integrated real-time control strategy in multi-tank A²O process for biological nutrient removal treating real domestic wastewater

3.1.1.1 Dynamics of online monitoring parameters during phase I

It can be seen from Fig. 3, the denitrification process began in early time of phase I due to influx of organic matter. The influx of organic matter rapidly consumed dissolved oxygen (DO) thus dissolved oxygen level was decreased quickly to less than 0.5 mg/L during 10 min and varied between (0.2 and 0.5) mg/L during the remaining period of phase I. Therefore, DO indicator cannot be used as an indication for denitrification and phosphorus release because DO level was decreased to less than 0.5 mg/L during 10 min as shown in Fig. 3. In the anaerobic process along with phosphorus release, the compounds containing phosphorus affect directly on ORP value whereas when the fermentation of organic acids (acetic, propionic, and lactic acids) are increased, the phosphoric acid is increased and this leads to a decrease in ORP value (Liu and Gao, 1995). According to that, ORP profile declined rapidly from +92 mv at 0 min to −241 mv at 70 min. Then, it declined slowly where it reached to a lower value (−246 mv) at 90 min. It can be seen from Fig. 3, the denitrification process began in the early stage of phase I because the influx of organic matter caused a rapid consumption of dissolved oxygen concentration. DO concentration has a greater impact on ORP value and additional to that, ORP value of raw wastewater itself is relatively low thus there was a sharp drop in the ORP profile at the beginning stage of phase I where ORP profile dropped rapidly from +92 mv to −54 mv during 5 min. The comparison of offline measurements of nutrient ( ${\text{NH}}_4^{+}-\text{N,}{\text{NO}}_3^{-}-\text{N}$ , TN and TP) and monitoring online parameters revealed that nitrate-N was converted into nitrogen gas by the denitrification process during 25 min, thus ORP profile decreased rapidly. After 25 min, ammonia-N concentration was increased slowly because the denitrification process was completed, thus ORP profile was decreased quickly which indicated, the anoxic process was completed and anaerobic state condition was a dominant process. Theoretically, the denitrification process is stopped when nitrate-N concentration is not consumed and thus the anaerobic state condition commences in a system. There was a large amount of generated organic acid due to anaerobic fermentation thus ORP profile was decreased quickly after 25 min the break point in the ORP profile was obvious at 25 min due to anaerobic phosphorus release therefore ORP can be used for monitoring the end of the denitrification process and beginning of anaerobic phosphorus release. ORP was declined slowly in the last stage of phase I. In the early stage of phase I, there was a rapid uptake of dissolved oxygen (DO) in tank one. ORP of influent wastewater was low, thus ORP profile was rapidly declined. After 10 min, the influent raw wastewater (low ORP) was mixed with tank one content where ORP was declined continuously and |dORP/dt| was reached to a low value at this period due to the effect of the denitrification process. At the end of denitrification, dORP/dt profile was decreased due to starting of anaerobic phosphorus release. The peak point of dORP/dt at 25 min also represented the beginning of phosphorus release. dORP/dt profile reached to −0.2 mv/min at 90 min as shown in Fig. 3. The rate of soluble phosphorus was reduced gradually along with hydrolysis of organic phosphorus thus dORP/dt absolute value reached to a lower value at 85 min. In the anaerobic process, the acidity is increased along with the increase of phosphoric acid thus, pH profile was slowly declined (downward trendline). On the other hand, the intracellular metabolism is divided into the catabolism, generating energy for the cell’s energy requirements, and the anabolism, leading to synthesis of new cell material; both processes take place simultaneously whereas anabolism and catabolism are used to complete the denitrification process; anabolism is synthesis of microbial cells while catabolism is nitrate, nitrite reduction to nitrogen gas which is the main part of the denitrification process. Theoretically; the transformation of nitrate to nitrite and then to nitrogen generates alkalinity OH⁻ so that pH would be raised but the experimental results showed, pH was decreased (downward trendline) in tank one during denitrification period because there was phenomena of simultaneous denitrification and phosphorus release. At the end of denitrification period, part of heterotrophic organism X_H started lactic acid fermentation and phosphorus accumulating organisms (PAOs) began to release phosphorus. pH pattern was suddenly and substantially declined that make pH profile appears break point which it indicates the end of denitrification process. pH profile appeared a small flat that is doesn’t have so obvious feature in pH profile as shown in Fig. 3. pH profile was declined slowly at the last stage of phase I whereas pH decline rate tends to be zero after 80 min so that pH can be used as control indicator for anaerobic phosphorus release. The dpH/dt profile was drawn as shown in Fig. 3. It was raised from negative to zero. There was an obvious peak point on dpH/dt profile at 85 min. Also, |dpH/dt| was declined to zero at 85 min. Therefore, dpH/dt profile can be used as a control parameter for anaerobic phosphorus release.

3.1.1.2 Dynamics of online monitoring parameters during phase II, III and IV

It was observed from Fig. 3; DO profile was increased slowly after the beginning of aerobic phase (90 min). Then, it was entered in a plateau period at 185 min. It was flocculated between (0.2 and 0.5) mg/L at the beginning stage of aeration period. This happened because dissolved oxygen of water was consumed by microbial cell for degrading organic matter. Organic matter was exhausted after 120 min and the remaining was un-biodegradable organic matter which could not be removed by nitrification process. During nitrification and phosphorus uptake processes, autotrophic organisms (X_A) oxidized ${\text{NH}}_4^{+}-\text{N}$ while phosphorus accumulating organisms (PAOs) degraded PHB at low dissolved oxygen requirement whereas the oxygen consumption rate is less than oxygen supply rate thus the remaining dissolved oxygen of tank one was raised slowly but it was raised quickly at 160 min. There was a jump in dissolved oxygen profile at 180 min due to degradation of a small part of biodegradable matter where a little of organic matter remained. Heterotrophic organism (X_H) could not intake much organic matter because substrate concentration was reduced. Thus, the rate of oxygen consumption was less than oxygen supply rate and according to that, nitrification process was completed. Then, the dissolved oxygen remained basically stable and its values were maintained at about 2.51 mg/L in spite of continuous aeration. It can be seen from Fig. 3, ORP profile was increased suddenly and significantly. It was raised from −246 mv at 90 min to +18 mv at 130 min. This happened because tank one was operated under aerobic state condition and thus there was a large amount of dissolved oxygen in water. According to relationship between ORP and DO, ORP pattern is so sensitive to dissolved oxygen variation where a small amount of oxygen could cause a significant increase in ORP pattern. Correspondingly, there was a balance between oxygen consumption rate of microbial cell and oxygen supply rate at early stage of aeration period. ORP profile was basically stable after 180 min because nitrification process was completed. It was increased slowly corresponding to nitrate-N variation so jump points did not appear similar to DO profile. There are two reasons for that, either DO (absolute value) was high and DO pattern was changed slowly so that it could not cause a significant variation in ORP pattern or nitrification process continued to oxidize ${\text{NH}}_4^{+}-\text{N}$ and uptake phosphorus by phosphorus accumulating organisms (PAO_s). From the above rules; nitrification and phosphorus uptake are in response to ORP thus oxidation reduction potential basically remained unchanged. It can be seen from Fig. 3, pH profile was increased, decreased, and increased again as general a phenomenon; this was similar to SBR system during the aerobic phase (Cui et al., 2009). pH profile was increased with time along with degradation of organic matter. This happened because heterotrophic organisms (X_H) degraded organic matter and released OH⁻ thus the alkalinity was raised. At the same time, there was aerobic degradation of produced organic acid which caused rising of pH profile. On the other hand; nitrification process produces H⁺ ion and consumes OH⁻ ion and although this phase was aerobic phosphorus uptake phosphoric acid reduced alkalinity thus pH profile was gradually decreased. On the other hand, the generation rate of acidity is more than alkalinity because HNO₃ is a strong acid thus pH profile was decreased as a whole. At the end of nitrification process, the consumption of OH⁻ ion was reduced because nitrification process was basically completed which caused increasing of pH profile. pH profile was contained a fold point at 190 min which can be used as an indicator for completing nitrification process. From the above analysis, it can be seen that pH and DO profiles can be used to indicate the end of nitrification and phosphorus uptake.

3.1.2.1 Dynamics of online monitoring parameters during phase I and II

It can be seen from Fig. 4, dissolved oxygen level was decreased below 0.5 mg/L during 40 min. Dissolved oxygen level was maintained between (0.2 and 0.5) mg/L which met the requirement of the denitrification process. Tank two was operated for a long period of aeration where it got 68.75% of total aeration time. Thus, if dissolved oxygen level was reached to a high value in the final stage of aeration period then the denitrification process would be faced with difficulty for reaching the required dissolved oxygen level. Dissolved oxygen level was reached above 5 mg/L at the last stage of phase VIII thus anoxic process needed more than 50 min to reduce dissolved oxygen to the required level of denitrification. Also, the high level of dissolved oxygen concentration increases power consumption therefore it is important to control aeration flux for reaching an appropriated level. The above analysis depicted that SND via DO is undesirable phenomena in the system because it increases power consumption in addition to that; it makes the denitrification process inefficient for a long time. Generally, ORP pattern is declined during the denitrification process. ORP profile was decreased from 121 mv at the beginning of phase I to −179 mv at the end of the denitrification process as shown in Fig. 4. It was dropped quickly in the beginning of phase I and then the rate was decreased slowly because the denitrification process and also organic phosphorus were reduced. The absolute value of dORP/dt was decreased below 0.4 mv/min after 130 min as shown in Fig. 4. Alkalinity/pH increases during the denitrification process because denitrification process released OH⁻ ion but pH of tank two was decreased most time of denitrification except the period of 115–130 min. pH was decreased due to low alkalinity of excess water from tank one. It was increased during the period of 115–130 min and then it was decreased because the denitrification process is complementary to alkalinity and denitrification process was weak during the last stage of phase II. It can be seen from Fig. 4, there was a turning point in pH profile at 130 min. This turning point can be used as a signal or indication for distinguishing the end of the denitrification process.

3.1.2.2 Dynamics of online monitoring parameters during phase III and IV

Dissolved oxygen level was increased immediately during aeration phases as shown in Fig. 4. DO profile of tank two at the early stage of phase III was not similar to dissolved oxygen profile of tank one. Dissolved oxygen profile of tank one was maintained in a plateau form for a certain time period. This happened because sludge concentration of tank two was less than sludge concentration of tank one. In addition; volume of tank two is equal to 0.67 of tank one thus tank two contained a few microorganisms compared to tank one which causes low oxygen consumption compared with tank one where the pollutant degradation was less than oxygen supply rate. Consequently, the amount of oxygen consumption in tank two was less than the amount of oxygen supply. Thus, DO profile was raised at the beginning of phase I till 165 min. Then, it was increased slowly as shown in Fig. 4. DO level increased along with continued aeration where it reached about 3.51 mg/L at the end of the nitrification process. ORP was increased with time during phase III and phase IV as shown in Fig. 4. Fig. 4 revealed, that dORP/dt was increased along with dDO/dt at the early stage of phase I where dDO/dt and dORP/dt profiles appeared at a peak point at 165 min and 170 min respectively. ORP was increased due to the interaction between nitrification and phosphorus uptake. ORP was relatively stable between 205 and 240 min. DO level was reached to a higher value at the end of a half cycle. DO concentration was varied with little amplitude so that it could not cause a significant variation in ORP profile. This happened because the effects of nitrification reaction and phosphorus uptake have finished therefore ORP was increased slowly. Then, it was slightly downward. The ORP growth rate dropped suddenly from positive to negative after 225 min. During phase III and phase IV, PH profile was increased (upward trend line) as a whole as shown in Fig. 4. Generally, pH profile was decreased and then increased. It was decreased due to aerobic degradation of anaerobic VFA which was the reason for reducing organic acids. Then, the rate of pH began to increase although the nitrification process of tank two produced H⁺ continuously. This happened because pH of raw wastewater has high acidity where the influent raw wastewater was continued through tank two as a first source point during phase III and IV. However, the biomass of tank two was little so that the nitrification process could not cause a drop in pH pattern. Sometimes, pH pattern was decreased due to a combined effect of influent raw wastewater and generated acidity by nitrification.

3.1.3.1 Dynamics of online monitoring parameters during phase I and II

It can be seen from Fig. 5, dissolved oxygen (DO) pattern was raised during phase I but it was declined gradually after the beginning of phase II. It was increased from 0.37 mg/L to 3.51 mg/L during phase I where it reached about 2 mg/L after 35 min of aeration period which met with the requirements of active nitrification process. Dissolved oxygen profile was raised quickly because the sludge concentration of tank three is relatively low compared to tank one and tank five. The mixed liquor suspended solid of tank three was about 2000 mg/L thus its microorganisms were few and according to that, the activity of the nitrification process for degrading organic matter and nutrient removal was weak. Thus, the oxygen consumption rate was less than oxygen supply rate. During phase II, tank one, tank three and tank four were operated under aeration state condition thus aeration quantity of tank three was decreased. DO level of tank three was decreased from 3.51 mg/L to 1.39 mg/L during phase II.

ORP profile was similar to dissolved oxygen profile because ORP profile was also raised during phase I while it was declining during phase II as shown in Fig. 5. ORP was increased from 23 mv at the early stage of phase I to 208 mv at the end of phase I. ORP profile began to decline after 90 min where it was reduced from 208 mv to 87 mv during phase II.

Owing to the fact that the nitrification process produces H⁺ ion and consumes OH⁻ ion, the overall pH profile is declined during the nitrification process. From Fig. 5, pH profile was decreased from 7.38 at the early stage of phase I to 7.22 at the end of phase I while it was varying between 7.21 and 7.23 during phase II. The activity of nitrification process was reduced after 90 min due to a decrease of oxygen supply rate thus alkalinity consumption rate was reduced relatively meanwhile pH profile appeared in plateau period during phase II.

3.1.3.2 Dynamics of online monitoring parameters during phase III and IV

It can be seen from Fig. 5, at the early stage of phase III; dissolved oxygen level was dropped from 1.39 mg/L to less than 0.53 mg/L at 175 min which met with the requirement of the efficient denitrification process. DO level fluctuated between 0.33 mg/L and 0.43 mg/L after 180 min. It can be seen from Fig. 5, ORP as a whole was decreased during the denitrification process. It was decreased along with dissolved oxygen decreasing. It was decreased from 87 mv at the early stage of phase III to −16 at 210 min. Then, it was increased to 19 mv at the end of phase IV. It can be indicated from ORP break point at 210 min, that denitrification process was basically completed. pH was increased from 7.22 at the early stage of phase III to 7.37 at the end of phase IV as shown in Fig. 5 because the denitrification process increased alkalinity.

3.1.4.1 Dynamics of online monitoring parameters during phase I, II, III and IV

It can be seen from Fig. 6, the variations of online monitoring parameters were so obvious. DO level was increased from 3.43 mg/L to 4.83 mg/L during phase I. During Phase II and III, it was decreased to 2.24 mg/L at 175 min and varied between 2.24 mg/L and 2.26 mg/L till the end of phase III. DO level was increased from 2.26 mg/L to 2.49 mg/L during phase IV. The variations of DO profile were due to the amount of oxygen supply that varied between phases according to numbers and volume of aeration tanks whereas there are two tanks that were operated under aerobic state condition during phase I, three tanks were operated under aerobic state condition during phase II and III and two tanks were operated under aerobic state condition during phase IV. Hence, the quantity of oxygen supply during phase II and III was less than the quantity of oxygen supply during phase I and IV. It is important to control dissolved oxygen concentration below 5 mg/L because the experimental investigation revealed, tank four needs more than 60 min during a second half cycle to reach denitrification requirement when DO level is above 5 mg/L at the end of phase IV where the nitrification rate is equal to denitrification rate in SBRs when DO level was around 0.5 mg/L (Elisabeth et al., 1996).

It can be seen from Fig. 6, ORP profile was increased quickly during the first 60 min. It remained without obvious variation from 90 min through the end of a first half cycle. pH profile increased rapidly from 7.35 to 7.48 during phase I. Then, it was stable and fluctuated between 7.48 and 7.51.

3.1.5.1 Dynamics of online monitoring parameters during phase I, II, III and IV

It can be seen from Fig. 7, dissolved oxygen level was decreased from 2.39 mg/L at the beginning of first half cycle to 1.2 mg/L at 60 min. Then, it was stabilized at around 1.2 mg/L. ORP profile was declined rapidly such as dissolved oxygen profile from 183 mv at the beginning of first half cycle to 99 mv at 70 min. Then, ORP profile was basically stable at about 96 mv. PH profile was overall increased from 7.25 at the beginning of first half cycle to 7.38 at 240 min because the denitrification process increased the alkalinity.

3.2.1.1 Conversion of phase I to phase II

The main difference between phase I and II, all tanks were operated under constant state condition except tank one. Tank one was changed from anaerobic state condition CSTR during phase I to aerobic state condition SBR during phase II. Therefore, it is considered as a critical path of phase I. The main functions of tank one during phase I were denitrification and phosphorus release where these main functions can be controlled using ORP and pH parameter as online monitoring parameters in tank one.

3.2.1.2 Conversion of phase II to phase III

The main difference between phase II and III, all tanks were operated under constant state condition except tank two and tank three. Tank three was not a limiting factor of aerobic function while tank two was a limiting factor of denitrification and phosphorus uptake by DNPAOs .Therefore, tank two was a critical path of phase II. Denitrification and phosphorus uptake can be controlled using ORP, pH, and DO parameter as online monitoring parameters in tank two.

3.2.1.3 Conversion of phase III to phase IV

The main difference between phase III and IV, all tanks were operated under constant state condition except tank one. Tank one was operated under static settling state condition during phase IV. Therefore, it is important to ensure that the main functions of nitrification, phosphorus uptake, and organic matter degradation were completed in tank one. These main functions can be controlled using ORP, pH, and DO parameter as online monitoring parameters in tank one.

3.2.1.4 Conversion of phase IV to phase V

The main task of phase IV was to prepare tank one as a settling discharged tank during a second half cycle where there is no alternative way to use settling tank instead of it. Therefore the time of phase IV should be determined based on sludge settling characteristics. The settling time is extended in order to avoid the loss of sludge with effluent water. Settling column was used for determining the settling time by drawing interface height with settling time. Then, three tangent lines and one perpendicular line were drawn. This procedure showed that 22 min is enough time for settling but it is extended to 30 min to avoid the loss of sludge with effluent water. Therefore, the static setting time 0.5 h can meet Chinese discharged requirements safely.

3.2.2.1 Performance of the IMT–A²O system under RTC strategy

The online monitoring parameters (DO, ORP and pH parameter) are used to implement automatic control strategy for phase transformations in the IMT–A²O process and offline measurements of nutrient removal were taken in the influent and effluent. The used raw wastewater was collected from the residential area of Wuxi campus, Southeast University. In this pilot process, the concentration ranges of pollutants in raw wastewater were COD of (175.4–343.9) mg/L, TP of (3.47–4.59) mg/L, ${\text{NH}}_4^{+}-\text{N}$ of (34.8–45.4) mg/L and TN of (39.2–51.2) mg/L.

It can be seen readily from Fig. 9; COD was always below 50 mg/L. The average removal rate of COD was more than 81% (Fig. 9a). The removal efficiency of TP in the systems was relatively stable and influent TP concentration was between 3.47 mg/L and 4.59 mg/L and the effluent TP concentration was below 0.5 mg/L. The average removal rate of TP was reached to more than 89.47% (Fig. 9b). The removal rate of ${\text{NH}}_4^{+}-\text{N}$ in the system was very stable. The influent ${\text{NH}}_4^{+}-\text{N}$ concentration was fluctuated between 34.8 mg/L and 45.4 mg/L while the effluent ${\text{NH}}_4^{+}-\text{N}$ concentration was below 5 mg/L (Fig. 9c). The average removal rate of ${\text{NH}}_4^{+}-\text{N}$ was more than 89.28% (Fig. 9c). The effluent concentration of TN was less than 15 mg/L. It was not affected by variations of influent TN. The effluent TN concentration was always below 9.97 mg/L (Fig. 9d).

Close

Figure 9

Pollutant removal rate in IMT–A²O under the real time control strategy (RTC_s).

Export to PPT

The above analysis showed, the IMT–A²O process runs stable and has a better removal rate of organic matter, TP, ${\text{NH}}_4^{+}-\text{N}$ and TN under the real time control strategy. The average removal efficiency of COD, ${\text{NH}}_4^{+}-\text{N}$ ,TN and TP was above 86%, 91%, 80% and 93%, respectively, using the integrated real-time control strategy as shown in Fig. 9. The effluent organic matter, TP, ${\text{NH}}_4^{+}-\text{N}$ and TN were met with Chinese discharged standard requirements of municipal wastewater treatment plant (GB18918-2002) level-A.

3.2.2.2 Frequency of real time control parameters

To determine the actual state of online monitoring parameters according to the frequency of online monitoring parameters which is used in phase conversion, the situation of phase transformations was observed in the IMT–A²O process with the real time control technology for ten operation cycles. The frequencies of used online monitoring parameters in each phase are shown in Table 3.