Testing of Advanced On-line Meters

Transcription

1 Testing of Advanced On-line Meters Stephanie Vermande 1*, Rong Liu 1, Alex Ekster 1 1 City of San Jose San Jose/Santa Clara Water Pollution Control Plant *To whom correspondence should be addressed. stephanie.vermande@sanjoseca.gov ABSTRACT Total suspended solids (TSS) and ammonia (NH 4 ) on-line meters were evaluated at the San Jose/Santa Clara Water Pollution Control Plant in seven streams. Three TSS meters were tested in five locations. The same TSS meter was recommended to be used for the activated sludge effluent and the subnatant of the dissolved air flotation tanks, in which the meter did not measure precisely TSS under normal operating conditions but was able to capture sudden increase. Only one of the two NH 4 meters tested in the mixed liquor was recommended for measuring concentration greater than 2 mgnh 4 -N/L. An ex-situ analyzer was also evaluated for low NH 4 concentration but was found to perform not as well as claimed by the vendor. This study highlights that due to plant-specific parameters, the field evaluation of on-line sensors is a required step prior to permanent installation. KEYWORDS: wastewater, on-line sensors, total suspended solids, ammonia, raw sewage, primary effluent, waste activated sludge, dissolved air flotation. INTRODUCTION Many recent published articles and reports in the USA (e.g. WERF, 21, WERF 27, Littleton et al., 29, Walz et al., 29) show the growing interest of using on-line meters for both monitoring and process control purposes in wastewater treatment plant. The advantages of using on-line meters are multiple: (1) reduction of number of analytical tests (2) better knowledge of the water quality and of the process performances, (3) real-time control of the process. These lead to savings in terms of energy, thanks to air consumption reduction, ranging from 5.5% (Leber et al., 29) to 4% (Wall et al., 29), chemical cost reduction of up to 5% (Sunner et al., 29) and laboratory cost savings (WERF, 27). As stated in the WERF report (27), the specificity of each wastewater and the numerous technologies available for the on-line sensors made it difficult to select the best equipments; therefore, testing is a necessary step to confirm the applicability of the equipment for the chosen applications prior to permanent installation. A program to evaluate total suspended solids (TSS) and ammonia (NH 4 ) meters at the San Jose/Santa Clara Water Pollution Control Plant (SJ/SC WPCP) has been initiated. SJ/SC WPCP is a highly automated biological nutrient removal (BNR) facility, where staff is continuously looking for methods to improve quality of process monitoring and implement energy saving and cost reduction solutions. 6927

2 The methodology and the results of this program are described in this article; cost benefit analysis is also provided. METHODOLOGY SJ/SC WPCP Description The SJ/SC WPCP is a tertiary wastewater treatment plant which treats domestic, industrial and commercial wastewater. Its design flow capacity is 167 mgd (632, m 3 /day) with a peak hourly flow of 271 mgd (1,25, m 3 /day). As described in Figure 1, the treatment consists of grit removal, primary settling tanks, BNR aeration basins and clarifiers, filtration and disinfection. The waste activated sludge (WAS) is mixed with pressurized and air saturated screened primary effluent and then thickened in dissolved air flotation tanks (DAF). The thickened WAS and the primary sludge are pumped to anaerobic digesters for sludge stabilization, volatile solids reduction, and methane gas generation. Figure 1. Treatment line of the SJ/SC WPCP and Location of the on-line sensors tested. On-line Meters Tested Two types of on-line analyzers were tested to measure TSS and NH 4 at different locations of the SJ/SC WPCP. As shown in Figure 1, the sensors were tested in the following water and sludge streams: - raw sewage after grit removal (RS); - primary effluent (PE); - mixed liquor (ML); - activated sludge effluent (ASE); - filter chlorinated effluent (FCE); - subnatant of the DAF tank (DAF-Sub); and - thickened waste activated sludge (TWAS). Three TSS meters from two different vendors were tested in the five locations (Table 1). The measuring principle of these meters is based either on an absorption infra-red light or a 6º backscattered infra-red light; the specific details about each meter is provided in Table

3 Three ammonia meters were tested: two of them in the mixed liquor and one in the activated sludge and filter chlorinated effluent (Table 2). Table 1. Total suspended solids meter characteristics. Vendors Endress-Hauser (EH) Sensors Name Turbimax W Turbimax W CUS65-A CUS65-C ViSolids 7IQ Parameter TSS TSS TSS Measuring Principle Absorption light, based on four beam alternating light 6º backscattered infra-red light, based on one source and one detector Cleaning System No cleaning system by default Ultrasound cleaning Range -12 g/l -5 g/l -4 mg/l to -1, g/l Channel of PE Location tested Channel of RS Well of Well ASE TWAS Well of DAF-Sub Well of TWAS Table 2. Ammonia meter characteristics. Vendors Hach ATI Sensors Name Varion Plus 7 IQ NH4Dsc Q45-FN Parameter NH 4 NH 4 NH 4 / NH 2 Cl Measuring ISE ISE Amperometric Principle Measurement Type Cleaning System Range In-situ Compressed Air Cleaning.1-1 mg/l to 1-1, mg/l In-situ Compressed Air Cleaning Ex-situ Wet chemistry with 2 reagents (1) N/A.2-1, mg/l -2. mg/l ASE Location tested ML ML FCE (1) Reagent A is sodium hypochlorite with buffer and stabilizer. Reagent B is hydrogen peroxide. The meters tested in the mixed liquor are in-situ sensors; their measuring principle is based on ion-selective electrode (ISE). A detailed description of the ISE method is given the WERF report (27). The accuracy claimed by the vendors is ±5% or ±.2 mg/l in standard solution. A wet chemistry analyzer was also tested for lower NH 4 concentration in the ASE and FCE. This ex-situ on-line meter measures both ammonia and monochloramine (NH 2 Cl). It consists of two amperometric sensors; the first one analyzes NH 2 Cl in the raw sample, the second sensor analyzes NH 2 Cl in the sample after addition of the reagent which converts NH 4 into NH 2 Cl. The difference in NH 2 Cl concentration given by the two sensors corresponds to the concentration of NH 4 in the raw sample. A filter upstream of the sensors removes excessive particles that could be present in the sample. The accuracy specified by the vendor is ±.1 mg/l. 6929

4 All the on-line meters (TSS and NH 4 ) were installed on temporary set-up as shown on Figure 2; the controllers were connected to data-loggers to record meter readings every minute. Raw Sewage EH - Turbimax W CUS 65-A Figure 2. Meter installation. Activated Sludge Effluent - ViSolids 7IQ Activated Sludge Effluent ATI - Q 45-FN Sample Collection In order to evaluate the accuracy of the meters, grab samples were taken and brought immediately to the SJ/SC WPCP s laboratory for analyses in triplicate. The analyses were done following the Standard Methods for Examination of Water and Wastewater (25). The average of the laboratory results was compared to the reading provided by the meter. Cost Analysis A simple cost analysis was performed assuming the equipment cost, a weekly analysis for validation of the meter, and a daily analysis before installation of the meter. The installation and maintenance cost were not considered. RESULTS and DISCUSSIONS This section describes and discusses the results obtained for the different meters at the various locations of the SJ/SC WPCP; first with the TSS meters and secondly with the NH 4 meters. The last paragraph presents a cost analysis with a payback period for the recommended on-line meters for purchase. TSS Meters Table 3 summarizes the results obtained at five locations with three TSS meters. 693

5 Table 3. TSS meters results in the various locations tested. Location RS PE ASE DAF-Sub TWAS (1) Brand - Sensor EH - CUS 65-A - ViSolids 7IQ 73 (2) Number of Spike only: 33 Average 273 (±76) 84 (±26) Concentration (mg/l) (standard deviation) Meter 623 (±368) 178 (±42) 8.2 (±2.3) 8. (±2.2) Average mg/l Difference (3) % -13% -125% 1.6% 4 59 (2) (±2) Spike only: 178 (±24) 4 (2) (±11) Spike Only: 166 (±21) 19. (2) Spike: % (2) Spike: 7% 4,5 (±5,5) 41,5 (±9,9) 1, Conclusion Not Recom. Not Recom. Recom. Recom. (4) Not Recom. 2.% (1) (2) (3) (4) Endress-Hauser CUS 65-C was also tested; results are not presented in this table see discussions. Excluding artificial spike test. Difference (mg/l) is between laboratory measurements and meter readings (mg/l). Difference (%) is the ratio between the difference (mg/l) and the laboratory measurements. TSS meter recommended for installation to detect spike, DAF process issues (increase in TSS). Recom. =Recommended for installation. By coupling two TSS on-line meters, one the raw sewage and one in the primary effluent, realtime primary clarifier performance could be obtained. To reach this goal, the TSS meter CUS 65-A from Endress Hauser was installed in the RS channel. The TSS measurements by the laboratory were on average 273 (±76) mg/l; the meter values were 623 (±368) mg/l. An average difference of 35 mg/l (13%) was observed between the two methods (Table 3), which is clearly unacceptable. As shown on Figure 3, the meter signal rapidly drifted just after 6 days in the stream. Loss of signals and rag accumulation around the probe were other operating issues encountered with this sensor. This TSS meter does not include a default cleaning system; an external pneumatic cleaning system was added in order to reduce rag accumulation and improve accuracy. Unfortunately this modification did not resolve the problems. The ViSolids 7IQ was not tested in raw sewage due to poor performance in the primary effluent (see below). 6931

6 14 12 Comparison of TSS Meter (EH - Turbimax CUS 65-A) and oratory Results in the Raw Sewage EH Period of 6 days before drift 1 TSS (mg/l) Period of 6 days before drift 2 Cleaning Figure 3. TSS meter readings in the raw sewage. The TSS meter ViSolids 7IQ from was tested in the PE channel. On average TSS measured by the meter was 94 mg/l (125%) higher than the laboratory measurements (178±42 mg/l versus 84±26 mg/l Table 3). Moreover TSS profiles obtained by laboratory results and meter readings were not in agreement as plotted on Figure 4; it can be clearly seen that the meter did not properly reflect TSS variations. 25 Comparison of TSS Meter () and oratory Results in the Primary Effluent over a 24 hour Period 2 TSS (mg/l) : 12: 15: 18: 21: : 3: 6: 9: Figure 4. TSS meter readings in the primary effluent. 12: According to the vendors, the dark color of the SJ/SC WPCP RS and PE absorb most the infrared light emitted by the sensor resulting in inaccurate readings. Another hypothesis for poor performance of the TSS meters in these streams is continuously changing particle distribution that, probably, affects optical properties of water. The ViSolids 7IQ meter was also tested in the activated sludge effluent. By better knowing the TSS in this stream, operators can respond more quickly to TSS increase and consequently adjust operating conditions of the BNR clarifiers and/or the downstream gravity filters. After calibration, the difference between the meter readings and laboratory analyses was on average.2 mg/l (1.6%) for a TSS ranging from 4 to 12 mg/l (Table 3). TSS profiles demonstrated that this probe accurately captures TSS variations (Figure 5). Both methods 6932

7 showed an increase of TSS during the evening and night from 5 to 1 mg/l. The origin of this fluctuation has not yet been identified and is still under investigation. Recommendations have been made to permanently install this meter to monitor TSS at this location Comparison of TSS Meter () and oratory Results in the Activated Sludge Effluent TSS (mg/l) /2/9 12: 7/21/9 12: 7/22/9 12: 7/23/9 12: 7/24/9 12: 7/25/9 12: 7/26/9 12: 7/27/9 12: Figure 5. TSS meter readings in the activated sludge effluent. The TSS on the subnatant of DAF tank has been tested with the ViSolids 7IQ meter. Under regular operating conditions, i.e. TSS < 1-12 mg/l, the meter results showed high discrepancies compared to the laboratory measurements. Indeed a difference of 28% (19 mg/l) is observed between the laboratory measurements (average of 59±2 mg/l) and the meter readings (average of 4±11 mg/l) (Table 3). As shown on Figure 6 (a and b), the TSS meter could not properly capture the TSS variations of the subnatant. However when the subnatant quality degrades (TSS > 1-12 mg/l) the meter readings agreed with the laboratory results. The difference between laboratory and meter values was within 7%; value acceptable for this application. As described in the SJ/SC WPCP s description section, the DAF tanks receive WAS mixed with air saturated primary effluent; the subnatant of the DAF is therefore mostly made of PE. As described above, this TSS meter did not perform well in primary effluent, so it is not surprising that it could not perform correctly when subnatant TSS values are in the same range as in primary effluent. However when performance of DAF tanks deteriorated and WAS solids migrated to the subnatant (> 1-12 mg/l TSS) meter performance has drastically improved. This on-line TSS meter is recommended for installation to detect DAF process issues. 6933

8 TSS (mg/l) Comparison of the TSS meter () and oratory Results in the Subnatant of the DAF Artificial spike (a) Figure 6. TSS meter results in the DAF subnatant. Meter Values (mg/l) oratory Results versus Meter () Results in Subnatant of the DAF (normal operating conditions) Measurements (mg/l) (b) Finally two TSS meters (Endress Hauser Turbimax CUS65-C and ViSolids 7IQ) were tested in the thickened waste activated sludge with the objective to better characterize the anaerobic digester feeding. The meters were installed in the well where TWAS is pumped to the anaerobic digesters. According to the laboratory analyses the total solids (TS) concentration ranged from 2,9 to 46,9 mg/l (Figure 7) with an average of 4,5 (±5,5) mg/l (Table 3). The laboratory measured total dissolved solids around 77 mg/l, value negligible compared to TS (>2, mg/l), therefore TSS can be approximated by TS. Loss of signal was experienced with the Endress-Hauser meter which made it impossible for the calibration step. Adding an air cleaning system did not improve the results. The ViSolids meter was also tested. On average 2% difference was observed between the meter readings and the laboratory results (Table 3); this small difference does not reflect, however, the fluctuations observed. Indeed as indicated on Figure 7 a TS difference of up to 23,7 mg/l has been observed between the laboratory and meter values (41,5 versus 17,8 mg/l). Recalibrations the meter did not improve the accuracy. Potential causes of the problem may include the set-up of the system and/or inefficient cleaning system. The probe was installed in the well where TWAS was discharged from the DAF tank and pumped to the anaerobic digester. Very little turbulence was observed in this well and sludge tended to adhere on the probe despite the ultrasonic cleaning system. In order to confirm the hypotheses, two tests should be scheduled: (1) keep the probe at the same location and add a different cleaning system (e.g. pneumatic air); (2) install the probe directly in the feeding line to the digesters where sludge flows continuously. None of these tests have yet been performed. 6934

9 6, Comparison of TSS Meter () and oratory Results in the Thickened Waste Activated Sludge 5, 4, TS (mg/l) 3, 2, 1, Figure 7. TSS meter results in the thickened waste activated sludge. Ammonia On-Line Meters in the Mixed Liquor Two on-line ammonia meters, Varion Plus 7IQ by and NH4D SC by HACH, based on the same measuring principle, had been tested in the mixed liquor of the aerated basins. The ultimate goal of this testing is to implement control strategies to control aeration in the BNR area to avoid ammonia breakthrough and reduce air consumption. The Varion Plus 7IQ was tested in four different longitudinal locations in the basin; at each location, the average NH 4 concentration was (Table 4): Location 1: 7. (±1.5) mg NH 4 -N/L; Location 2: 4.7 (±1.5) mg NH 4 -N/L; Location 3: 1.9 (±.8) mg NH 4 -N/L; Location 4:.2 (±.1) mg NH 4 -N/L. Table 4. Ammonia meters results in the mixed liquor. Brand - Sensor - Varion Plus 7 IQ Hach - NH4Dsc Location Nber of samples Av.. 7. (±1.5) 4.7 (±1.5) 1.9 (±.8).2 (±.1) 2.2 (±2.7) 5.9 (±2.) Conc. (1) (mg/l) Meter 7.3 (±1.5) 4.5 (±1.5) 1.9 (±.9).4 (±.1) 2.9 (±6.9) 7.6 (±3.6) Difference (2) -.29 mg/l -5.%.2 mg/l 3.2% -.2 mg/l -5.%.23 mg/l 53.1% -.65 mg/l -2.4% mg/l -55.7% Conclusion Recom. Recom. Recom. Not Recom. Not Recom. Not Recom. (1) Concentration in terms of NH 4 -N - Standard deviation is specified in bracket. (2) Difference (mg/l) is between laboratory measurements and meter readings (mg/l). Difference (%) is the ratio between the difference (mg/l) and the laboratory measurements. Av. Conc. = Average Concentration. Recom.= Recommended for installation. This meter provided accurate readings within 5% when the NH 4 concentration was greater than 2 mgnh 4 -N/L; the difference that is in the precision range claimed by the vendor. As shown on Figure 7 a, b and c, the meter values followed closely the laboratory measurements. 6935

10 When the NH 4 concentration is below 1.9(±.8) mgnh 4 -N/L (Location 4), discrepancy of 53% (.23 mg/l) was observed. The meter is not accurate enough to detect precisely low ammonia concentration (Figure 7-d); the limit of this on-line meter was reached. The ammonia meter based on ISE measuring principle is only recommended for application with NH 4 concentration in the mixed liquor above 2 mg/l. 12 Comparison of the Ammonia Meter () and oratory Results in Location 1 8 Comparison of the Ammonia Meter () and oratory Results in Location 2 NH4-N (mg/l) (a) NH4-N (mg/l) (b) Comparison of the Ammonia Meter () and oratory Results in Location Comparison of the Ammonia Meter () and oratory Results in Location 4 NH4-N (mg/l) NH4-N (mg/l) (c) (d) Figure 8. Ammonia meter in the mixed liquor. The NH4D SC meter was tested in two different longitudinal locations in the basin; at each location, the average NH 4 concentration was (Table 4): Location 5: 2.2 (±2.7) mg NH 4 -N/L; Location 6: 5.9 (±2.) mg NH 4 -N/L. By looking only at the average difference between the lab results and the meter readings for the Location 5, it appears that the sensor was accurate within 2.4% (.65 mg/l Table 4) and provided precise information about ammonia concentration. However when looking more closely at the results, the NH 4 concentration given by the meter did not follow the laboratory results; most of the meter readings were more than 1% different compared with the laboratory results (Figure 9-a). At the Location 6, the discrepancy between laboratory and meter values averaged at 56% (1.7 mg/l). Moreover when the probe was immerged in laboratory standard solutions, the ammonia concentration drifted over time. Several unsuccessful attempts to improve accuracy (recalibration, changing cartridge, troubleshooting with Hach R&D center) did 6936

11 not solve the problems; the vendor agreed to take the system back and reimbursed the SJ/SC WPCP for the purchase. 35 Comparison of the Ammonia Meter (Hach) and oratory Results in Location 5 16 Comparison of the Ammonia Meter (Hach) and oratory Results in Location NH4-N (mg/l) Hach NH4-N (mg/l) Hach Sample (a) (b) Figure 9. Ammonia meter (Hach) in the mixed liquor. Ammonia On-Line Meters in the Activated Sludge Effluent and Filter Chlorinated Effluent As ISE probe could not measure low NH 4 concentration, an ex-situ NH 4 analyzer based on amperometric probes has been evaluated. The goal is to continuously monitor ammonia to capture breakthrough that can cause National Pollutant Discharge Elimination System (NPDES) permit violation. A wet chemical analyzer (ATI Q 45-FN) was tested in the activated sludge effluent and the filter chlorinated effluent for ammonia and monochloramine concentrations. The results obtained are depicted in the Table 5. In the ASE, no monochloramine was present. The NH 4 concentration ranged from.2 to.33 mgnh 4 -N/L with an average of.15 (±.9) mg NH 4 -N/L (Figure 1). Consistently the analyzer indicated NH 4 concentration below.1 mgnh 4 -N/L which appears to be the limit of detection of the meter. Several attempts to recalibrate the sensor, changing equipment parts and reagents were done but did not improve the accuracy. Table 5. Ammonia meter results in the activated sludge and filter chlorinated effluent. Location Activated Sludge Effluent Filter Chlorinated Effluent Parameters NH 4 -N NH 4 -N (free) NH 2 Cl-Cl 2 Number of samples Av. Conc. (1) (mg/l).15 (±.9).46 (±.8).72 (±.4) Meter <.1 (n/a).66 (±.1) 1.51 (±.22) Difference (2) -.2 mg/l -.79 mg/l n/a -49% -112% Conclusion Not Recom. Not Recom. Not Recom. (1) Standard deviation is specified in bracket. (2) Difference (mg/l) is between laboratory measurements and meter readings (mg/l). Difference (%) is the ratio between the difference (mg/l) and the laboratory measurements. Av. Conc. = Average Concentration. n/a = not applicable. Recom. = Recommended for installation. 6937

12 Comparison of the Ammonia Meter (ATI) and oratory Results in the Activated Sludge Effluent ATI NH4-N (mg/l) Figure 1. Ammonia meter (ATI) in the activated sludge effluent. The analyzer was also installed in the filter chlorinated effluent. At this location, the free NH 4 concentration averaged at.46 (±.8) mgnh 4 -N/L and the NH 2 Cl at.72 (±.4) mgnh 4 -N/L (Table 5). The analyzer results were not in agreement with the laboratory measurements; a difference of 49% and 112% was obtained respectively for free NH 4 and NH 2 Cl concentration (Figure 11). Free NH4-N (mg/l) Comparison of the ATI Meter and oratory Results in Filter Chlorinated Effluent ATI NH2Cl-Cl2 (mg/l) Comparison of the ATI Meter and oratory Results in Filter Chlorinated Effluent ATI (a) Free Ammonia (b) Dichloramine Figure 11. Ammonia meter (ATI) in the filter chlorinated effluent. One of the hypotheses suggested by the vendor is the presence of dichloramine in the FCE, which may interfere and false the meter readings. oratory analyses showed an equal amount of dichloramine and monochloramine in the SJ/SC WPCP FCE. Other issues observed in both locations (SE and FCE) with this analyzer were: - variations of the signal while pumping the same sample; - drift of the signal over time; and - slow response time of the meter. This ex-situ on-line ammonia analyzer was not recommended for installation at neither of the tested locations at the SJ/SC WPCP. 6938

13 Cost Analysis As previously discussed only three on-line meters were recommended for permanent installations: - ViSolids 7IQ TSS meter for the activated sludge effluent; - ViSolids 7IQ TSS meter for the subnatant of the DAF; and - Varion Plus 7 IQ ammonia meter in the mixed liquor. As described in Table 6, these meters have a short payback period, i.e. less than one year. This quick return on investment (ROI) makes these on-line sensors attractive. Moreover if aeration control strategy is developed, the ROI on the ammonia meter could be even shorter. Table 6. Cost analysis. Meters Location Equipment Cost oratory Cost Payback Period TSS Meter Activated Sludge Effluent ~ $6, $27 9 months TSS Meter Subnatant of the DAF ~$6, $37 9 months Ammonia Meter Mixed Liquor ~$9, $37 9 months CONCLUSIONS Out of the seven locations tested only at a few of them meter results were similar to laboratory measurements: - in the activated sludge effluent, ViSolids 7IQ TSS meter readings were within 2% of laboratory results; - in the subnatant of the DAF, ViSolids 7IQ TSS meter readings were within 7% of laboratory results at concentration above 15 mg/l; and - in the mixed liquor, Varion NH 4 meter readings were within 5% of laboratory results at concentration above 2 mgnh 4 -N/L. These meters were recommended for installation at these three SJ/SC WPCP locations. Return on investment on purchase of recommended on-line meters is less than a year when only the reduction of laboratory analyzes is considered. The return on investment could be even shorter for the ammonia meter if aeration control strategies are developed. As demonstrated by this on-line meter evaluation program at the SJ/SC WPCP, results obtained are often site specific due to different wastewater characteristics and specificities of the treatment line. Prior to investing money and time in purchasing and installing meters and developing control strategies, testing is highly recommended to maximize the chance of success. Research and testing will continue at the SJ/SC WPCP to find on-line meters that perform correctly in raw sewage, primary effluent, thickened sludge and in activated sludge effluent for TSS or NH 4 concentration. ACKNOWLEDGMENTS Authors thank SJ/SC WPCP Operation, Instrumentation and oratory staff for providing support during this project. Authors would also like to thank Dr. Issayas Lemma and Process Engineering interns. 6939

14 REFERENCES American Public Health Association, American Water Works Association and Water Environment Federation (25) Standard Methods for the Examination of Water and Wastewater. 21st edition. Biological Nutrient Removal Task Force, Alexandria, VA, USA. Leber R., Kestel S., Gray M., Duffy G. (29) Utilizing Automatic Dissolved Oxygen and Internal Recycle Set- Point Control at Abington WWTP in Southeastern Pennsylvania. Weftec. Littleton H., Daigger G., Amad S., Strom P.(29) Develop Control Strategy to Maximize Nitrogen Removal and Minimize Operation Cost in Wastewater Treatment by Online Analyzer. Weftec. Sunner N., Haeck M., Thornton A. (29) Evaluation of a Full-Scale Activated Sludge Real Time Control System. Weftec. Wall A., Hart M., Mack J., O Brien M. (29) The Use of Advanced Process Monitoring and Control to Optimize Energy Use on an Activated Sludge Plant. Weftec. Walz T., J.R. Coughenour J.R., Williams K., Jacobs J., Shone L., Stahl T., Kestel S., Palmer D. (29) Energy Savings at Phoenix 23rd Avenue Wastewater Treatment Plant Using Feed-Forward Process Control. Weftec. Water Environment Research Foundation (21) Thickening and dewatering processes: how to evaluate and implement automation package. Project 98-REM-3. Water Environment Research Foundation (27) On-Line Nitrogen Monitoring and Control Strategies. 3-CST