Menniti, Adrienne

Clean Water Services ODIN Pilot Case Study

Print publication date Oct 2023

DOI10.2175/193864718825159117

Volume / Issue

Content sourceWEFTEC

Word count12

Abstract

Water Research Foundation Project 5121 is titled Development of Innovative Predictive Control Strategies for Nutrient Removal. The project is focused on developing and full-scale testing of a hybrid (machine learning + mechanistic model) nutrient management controller at four different WRRFs. The project team has named the controller ODIN: Operational Decision-making Information Network. The primary objectives of the ODIN pilot for CWS are to: Predict the expected diurnally varying plant influent orthophosphate loading over the following 24 hours. Recommend an alum dosing set point to maintain a requested primary effluent orthophosphate load. A secondary goal is the additional information provided on parameters like influent ammonia. For the CWS application, the ODIN systems appears to be a promising opportunity to improve plant operation using digital twin technology.

A hybrid (machine learning + mechanistic modelling) controller called ODIN has been developed through WRF5121 and is currently being tested at the Clean Water Services Durham WRRF. The system recommends for primary alum addition setpoints based on forecasted phosphorus loads and target bioreactor ortho-phosphate mass loading targets. The ODIN system appears to be a promising opportunity to improve plant operation using digital twin technology. This paper summarizes the status of the CWS pilot.

Presentation time

15:35:00

15:50:00

Session time

15:30:00

17:00:00

SessionCase Studies of Machine Learning in Full-Scale Nutrient Management Part II

Session locationRoom S505b - Level 5

TopicAdvanced Level, Facility Operations and Maintenance, Intelligent Water, Nutrients

Author(s)

Author(s)A. Menniti 1; A. Menniti 1 ; B. Johnson 2; K. Lesnik 3; C. Yang 4;

Author affiliation(s)Clean Water Services 1; Clean Water Services 1 ; Jacobs 2; Jacobs 3; 4;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Dance of the Machines at AlexRenew WRRF: Piloting Experience of Machine Learning Hybrid Nutrient Controller

Print publication date Oct 2023

DOI10.2175/193864718825159116

Volume / Issue

Content sourceWEFTEC

Word count8

Description: Dance of the Machines at AlexRenew WRRF: Piloting Experience of Machine Learning...

Abstract

A full-scale pilot of a hybrid (machine learning + mechanistic modelling) based nutrient controller is being implemented at Alexandria Renew Enterprises (AlexRenew), VA as part of The Water Research Foundation project 5121. The hybrid nutrient controller developed for this project has been named ODIN (Operational Decision-making Information Network). At AlexRenew, ODIN is used to recommend equalization discharge and return pumping setpoints with the goal of minimizing overall energy and chemical usage at the facility while maintaining a very low effluent total nitrogen. Recent historical and simulated nutrient removal performance is presented along with the metrics selected to evaluate the control period with ODIN suggestions being considered.

This paper was presented at WEFTEC 2023 in Chicago, IL.

SpeakerOristian, Monica

Presentation time

16:25:00

16:40:00

Session time

15:30:00

17:00:00

SessionCase Studies of Machine Learning in Full-Scale Nutrient Management Part II

Session locationRoom S505b - Level 5

TopicAdvanced Level, Facility Operations and Maintenance, Intelligent Water, Nutrients

Author(s)

Oristian, Monica

Author(s)M. Oristian 1; M. Oristian 1 ; B. Johnson ; C. Yang 3; J. Registe 5; H Stewart 6; K. Lesnik 7; A. Menniti 8;

Author affiliation(s)Alexandira Renew Enterprises (AlexRenew) 1; Alexandira Renew Enterprises (AlexRenew) 1 ; Jacobs ; Jacobs 3; Jacobs 5; Jacobs 6; Jacobs 7; Clean Water Services 8;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Print publication date Oct 2023

DOI10.2175/193864718825159118

Volume / Issue

Content sourceWEFTEC

Word count16

Designing for Humans

Abstract

The Water Research Foundation Project 5121 is titled Development of Innovative Predictive Control Strategies for Nutrient Removal is focused on developing and full-scale testing of a hybrid (machine learning + mechanistic model) nutrient management controller at four different WRRFs. The project team has named the controller ODIN: Operational Decision-making Information Network. The philosophy behind the ODIN User Interface is to make this tool easy to use for operations staff. To accomplish this objective, we’ve focused on two goals with the user interface design: 1. Keep the user priorities and mind set in mind at every step of the user interface design. 2. As engineers, we need to remember to keep the interface easily accessible and simple. Any new way of doing things must make the life of the operator better in some way. Without that, even the best ideas will likely fail at widespread adoption

A hybrid (machine learning + mechanistic modelling) controller developed through WRF5121 and named ODIN has been developed and is currently being tested at four different WRRFs. The philosophy behind the ODIN User Interface is to make this tool easy to use for operations staff. Every step of the user interface development process has kept user priorities and needs at the forefront. This paper reviews the ODIN approach to user interface development.

Presentation time

13:35:00

13:50:00

Session time

13:30:00

15:00:00

SessionApplications of Machine Learning in Full-Scale Nutrient Management Part I

Session locationRoom S504c - Level 5

TopicFacility Operations and Maintenance, Intelligent Water, Intermediate Level, Nutrients

Author(s)

Author(s)A. Menniti 1; B. Johnson 2 ; A. Menniti 1;

Author affiliation(s)Clean Water Services 1; Jacobs 2 ; Clean Water Services 1;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Evaluating Measurement Approach and Variability of the Apparent Fermentation Rate

Print publication date Oct 2023

DOI10.2175/193864718825159078

Volume / Issue

Content sourceWEFTEC

Word count4

Abstract

In an effort to understand options to improve EBPR stability, research into the sidestream EBPR process configurations has increased. Understanding the carbon generation rate of S2EBPR is key to determining if the process may improve EBPR performance. CWS has investigated methods to describe this apparent fermentation rate and evaluated AFR implications for fullscale performance. The results indicate that differences in measurement methodology can cause the resulting AFR to vary greatly.

This paper was presented at WEFTEC 2023 in Chicago, IL.

SpeakerSchauer, Peter

Presentation time

13:30:00

13:50:00

Session time

13:30:00

15:00:00

SessionBiomass Fermentation for Carbon Efficient Biological Phosphorus Removal

Session locationRoom S404c - Level 4

TopicFacility Operations and Maintenance, Intermediate Level, Municipal Wastewater Treatment Design, Nutrients

Author(s)

Schauer, Peter

Author(s)P. Schauer 1; P. Schauer 1 ; A. Menniti 1;

Author affiliation(s)Clean Water Services 1; Clean Water Services 1 ; Clean Water Services 1;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Genome-centric Insights into Full-scale EBPR Captured During a Biomonitoring Campaign

Print publication date Oct 2023

DOI10.2175/193864718825158989

Volume / Issue

Content sourceWEFTEC

Word count11

Description: Genome-centric Insights into Full-scale EBPR Captured During a Biomonitoring...

Abstract

Phosphorus removal from wastewater remains a core component of modern wastewater treatment strategies critical for protecting aquatic ecosystems and capturing valuable nutrients. Enhanced biological phosphorus removal (EBPR) is attractive since it reduces the need for chemical precipitating agents and facilitates phosphorus recovery. However, instability can limit the reliable application of EBPR for meeting stringent effluent phosphorus limits. As part of an ongoing biomonitoring campaign, samples were collected before and during an EBPR process stability episode and sequenced for paired metagenomes and metatranscriptomes. Analysis revealed that shifts in carbon and phosphorus loads coupled with changes in the population sizes of non-fermenting polyphosphate accumulating organisms (PAOs; Candidatus Accumulibacter spp.), as identified in the metagenomic data, were associated with poor P-uptake. Metatranscriptomic data from these PAO populations revealed they were indeed substrate limited, as hypothesized from process data, with key genes related to carbon storage appearing down regulated during the instability event with respect to the stable control despite having a greater cell load during the instability period. Accordingly, several gene families related to carbon uptake, VFA metabolism, and PHA synthesis were found to be sensitive to the underlying carbon-limited conditions of the process instability. Our findings illustrate how combined metagenomic and metatranscriptomic sequencing can facilitate interpretation of process data and aid in evaluating the root cause of instability. These data also indicate potential gene targets suitable for developing EBPR performance markers as part of longer term or more frequent biomonitoring efforts.

As part of an ongoing biomonitoring campaign, samples were collected before and during an EBPR process stability episode and sequenced for paired metagenomes and metatranscriptomes. Analysis revealed that shifts in carbon and phosphorus loads coupled with changes in the population sizes and gene transcription of PAOs were associated with poor P-uptake. Our findings illustrate how sequence data can facilitate interpretation of process data and aid in evaluating the root cause of instability.

SpeakerLindner, Blake

Presentation time

14:00:00

14:30:00

Session time

13:30:00

15:00:00

SessionConnecting Molecular Biology Sensors to EBPR Performance

Session locationRoom S502 - Level 5

TopicAdvanced Level, Facility Operations and Maintenance, Municipal Wastewater Treatment Design, Nutrients

Author(s)

Lindner, Blake

Author(s)B. Lindner 1; B. Layton 2 ; B. Lindner 1; B. Layton 2; R. Golda 2; K. Lesnik 3; K. Konstantinidis 4; A. Menniti 2; P. Schauer 2;

Author affiliation(s)School of Civil and Environmental Engineering, Georgia Institute of Technology 1; Clean Water Services 2 ; School of Civil and Environmental Engineering, Georgia Institute of Technology 1; Clean Water Services 2; Clean Water Services 2; Maia Analytica 3; School of Civil and Environmental Engineering, Georgia Institute of Technology 4; Clean Water Services 2; Clean Water Services 2;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Heating Primary Sludge for Increased Fermenter Volatile Fatty Acid Production

Print publication date Oct 2023

DOI10.2175/193864718825159085

Volume / Issue

Content sourceWEFTEC

Word count11

Abstract

The Durham Advanced Treatment Facility in Tigard, OR operates a fermentation complex that utilizes a Unified Fermentation and Thickening (UFAT) process to ferment primary sludge. Volatile fatty acid (VFA) produced through fermentation are used to support biological phosphorus removal (BPR). During cold weather months, solids retention time (SRT) limitations result in reduced VFA production, leaving BPR somewhat unstable especially in late spring when the plant effluent total phosphorus limit becomes more stringent. A sludge heating system was installed upstream from the fermentation complex to heat primary sludge, overcoming SRT limitations of the current system to increase cold weather VFA production. This improves BPR stability in the late spring and changes future fermenter design criteria, reducing the required footprint for future fermentation system expansion.

A. MennitiA. MennitiR. GoldaJ. Abrook

Presentation time

10:50:00

11:10:00

Session time

10:30:00

11:10:00

SessionAddressing Low Influent Carbon for EBPR

Session number517

TopicMunicipal Wastewater Treatment Design, Nutrients, Research and Innovation

Author(s)

Author(s)A. Menniti1; A. Menniti1; R. Golda2; J. Abrook2;

Author affiliation(s)Clean Water Services, OR1; Clean Water Services, OR, OR2

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Living Capital Planning Tools

Print publication date Oct 2020

DOI10.2175/193864718825157520

Volume / Issue

Content sourceWEFTEC

Word count11

Abstract

INTRODUCTION Clean Water Services (CWS) operates four WRRFs that are expanded and improved through a ~$40 million yearly capital budget. The goal of the CWS capital planning approach is to have a level capital outlay year to year. Capital expenditures are reprioritized yearly to achieve this goal and to ensure that processes are expanded only as they reach capacity. Typical master planning approaches are updated only about every five years. A 'living' capital plan allows CWS to adapt to changes in critical assumptions that drive planning timelines and proactively manage capital expenditures. This paper presents the CWS approach to creating and managing a living capital plan and summarizes the benefits. OVERVIEW The first integrated capital planning tool was developed for the Durham WRRF in 2020 with Carollo and Jacobs, the consultants for the most recent Durham master plan project (Figure 1). As part of the ongoing West Basin master plan, work is underway with Carollo to develop the tool for the three interconnected West Basin facilities, which include Rock Creek, Hillsboro and Forest Grove (Figure 1). The initial pieces of the tool are developed in collaboration with consultants but the final product is owned and managed by CWS. The goal of the living capital planning tool is that it's updated every year and should improve and evolve over time. OUTCOMES FLOW AND LOAD PROJECTIONS Flow and load projections are developed from base assumptions in the flow and load spreadsheet. The process of building the spreadsheet provided CWS with a comprehensive understanding of all the critical assumptions used to generate the projections, making it easier to evaluate those assumptions over time. Alternate planning scenarios were also created based on differing growth assumptions for the Durham WRRF, and flow diversion alternatives for the West Basin WRRFs. Historic flows and loads are evaluated annually and compared to projections. Tracking observations against planning numbers makes it evident when revisions are necessary (Figures 2 and 3). While the intention is not to revise the projections every year, the spreadsheet tool and yearly evaluation process is meant to make the effort for revising projections more efficient when needed. PROCESS PERFORMANCE AND DESIGN CRITERIA Design criteria, which set the capacity of the unit processes (e.g. solids loading rate), and process performance criteria, which are assumptions based on operational observations (e.g. percent removal, solids capture, sludge volume index) are also evaluated yearly. The data validation and process capacity spreadsheets gather process performance assumptions and design criteria from the master plan and compare them to historical observations. This comparison determines when unit processes are approaching capacity triggers and identifies when unit processes should be optimized due to declining performance. One example of a process performance parameter (capture in the primary solids thickening process, Figure 4) that drove the capital planning timeline for secondary treatment (Figure 5) is illustrated here. Lower solids thickening capture sends additional solids loads to the secondary process, decreasing its treatment capacity. Just this single assumption changed the secondary capacity trigger year by 4 years (Figure 5). With projects at four WRRFs to phase, improving this 4-year discrepancy in secondary capacity allows capital outlay to be phased more effectively. MASS BALANCE SPREADSHEET VERUS PROCESS SIMULATOR CWS uses both a process simulator and a spreadsheet mass balance model to create solids projections for its West Basin WRRFs. The mass balance spreadsheet can be used to quickly evaluate scenarios that require a lower level of sensitivity and to identify scenarios that could have a major impact to timelines. This simpler, quicker approach is beneficial with the three interconnected West Basin treatment facilities as running a process simulator with all three WRRFs connected is complex and prohibitively slow. However, the process simulator generates more detailed predictions within each individual facility that are beneficial for informing comprehensive planning evaluations. Table 1 compares CWS experiences using these tools. CWS aims to extend both approaches to its East Basin capital planning system to take advantage of their complementary strengths. LESSONS LEARNED The living capital planning approach means that planning is not confined to periodic, consultant-lead master planning efforts. Tracking of flows and loads and other critical assumptions yearly against master planning values has provided valuable insights into when capacity and capital planning shifts are needed. It also daylights areas of conservatism or risk as capacity triggers are reached. Even if a utility is not running models or doing the projections themselves, tracking assumptions against reality is beneficial and achievable for any utility. Other benefits: Having ownership over the data and tools gives CWS the independence to quickly evaluate new scenarios (e.g. impact of a new large industrial user or shifts in population growth). Daylighting and tracking assumptions also aids in documenting the information that drives capital planning decisions. The integrated tool makes capital planning information more accessible to a wider range of CWS stakeholders from executive managers to operators.

This presentation presents the Clean Water Services approach to creating and managing a living capital plan and summarizes the benefits. The goal of our approach is to daylight areas of conservatism and risk as capacity triggers are reached and to routinely review unit process performance and plant influent flows and loads to catch and address deviations as they arise.

TopicEnergy Production, Conservation, and Management, Facility Operations and Maintenance, Intelligent Water, Intermediate Level, Nutrients

Presentation time

16:30:00

16:50:00

Session time

15:30:00

17:00:00

SessionLeveraging Automation and Analytics for Better Situational Awareness and Optimization: Part II

Session number224

Session locationRoom 350

Author(s)

Menniti, Adrienne, Sosa-Hernandez, Ornella, Barker, Leila, Schauer, Peter

Author(s)A. Menniti1, O. Sosa-Hernandez1, L.K. Barker1, P. Schauer1

Author affiliation(s)1Clean Water Services, OR

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Online Instrumentation for Dewatering Centrifuge Optimization

Print publication date Oct 2024

DOI10.2175/193864718825159537

Volume / Issue

Content sourceWEFTEC

Word count5

Abstract

Dewatering is a major resource recovery facility operating cost. Therefore, dewatering optimization could provide a significant cost saving opportunity. At CWS, the historical approach for dewatering centrifuge operation is to set targets for dewatered cake dryness and solids capture. This approach encouraged increased polymer use with limited incentive to critically evaluate operational set points. Because of the costs associated with dewatering, CWS has moved toward cost-based dewatering centrifuge operation using a suite of analyzers from Valmet that allows cost-based dewatering optimization information to be available in real time. The first of four new dewatering centrifuges installed at the Durham and Rock Creek facilities was commissioned in May 2019. Valmet instruments measuring centrifuge feed total solids (Valmet TS analyzer), dewatered cake total solids (Valmet DS analyzer) and centrate total suspended solids (Valmet LS analyzer) were also commissioned in May 2019 with a pilot agreement to allow their performance to be evaluated for 60 days before investing fully. Based on trial period performance, CWS has chosen to retain the Valmet instrumentation to support dewatering centrifuge operation and will install instrumentation on the remaining new dewatering centrifuges.

M. GatesM. GatesA. MennitiA. MennitiM. Idehara

Presentation time

11:10:00

11:30:00

Session time

10:30:00

11:30:00

SessionOvercoming Dewatering Difficulties

Session number520

TopicBiosolids and Residuals, Facility Operations and Maintenance

Author(s)

Author(s)M. Gates1; M. Gates1; A. Menniti2; A. Menniti2; M. Idehara1;

Author affiliation(s)Clean Water Services, OR, OR1; Clean Water Services, OR2

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Online Phosphorus Stability Analyzer: Quantifying The Risk Of BPR Upset

Print publication date Oct 2020

DOI10.2175/193864718825157544

Volume / Issue

Content sourceWEFTEC

Word count7

Abstract

INTRODUCTION
Biological phosphorus removal (BPR) has been a cornerstone of the Clean Water Services (CWS) effluent phosphorus permit compliance strategy for decades. The BPR process has historically been observed to operate stably over long periods of time, only to become upset during critical times of the phosphorus permit compliance season. For CWS, an upset condition is defined as the secondary effluent orthophosphate (OP) daily average concentration > 0.1 mg/L P. To reliably meet a stringent effluent phosphorus permit with limited chemical use, it is useful to have an early warning metric for detection of upsets in the secondary effluent. We have demonstrated the effectiveness of an online orthophosphate analyzer coined the 'Stability Analyzer' to better understand BPR Stability (Menniti et al, 2016, Bushee et al, 2018). The analyzer has been operating on aeration basin one (AB1) at the Durham Wastewater Resource Recovery Facility (WRRF) since 2017. Its location was intentionally determined at about halfway through the aerobic zone, which we define as the stability point, where generally all phosphorus uptake has occurred by this point during stable operation. The purpose of this paper is to quantify the risk of an upset event using historical stability analyzer data. A new aeration basin (AB5) recently came online at the Durham facility. This paper will also demonstrate the process of strategically locating a stability analyzer on AB5, which has a different physical geometry to AB1 (Figure 1, Table 1).
UPSET RISK QUANTIFICATION RESULTS
Through empirical observation, we found that BPR operates stably when the Stability Analyzer reads 'zero,' which we define as a daily average concentration < 0.2 mg/L P, and that the BPR process is moving toward an upset event when the analyzer becomes non-zero (Figure 2). Not all instances of non-zero daily average Stability Analyzer measurements result in upset events. Therefore, we have quantified the risk of upset using historical data. Through data analysis from summers 2018-2020, we have quantified the risk of near-future upset events. All analysis is based on daily average analyzer concentrations and excludes time periods when alum was applied to the secondary clarifiers, isolating periods of phosphorus removal solely by biological means. We observe two types of conditions where the Stability Analyzer offers early warning of upset events: A. The daily average Stability Analyzer OP concentration has read zero for at least three days followed by an increase into non-zero daily average territory (Figure 3A, 3B). B. The daily average Stability Analyzer OP concentration is already non-zero and the daily average concentration increases the next day (Figure 4). There are 18 instances of Condition A and 139 instances of Condition B. Of the 18 Condition A instances, 9 resulted in a secondary effluent upset event within the next 2 to 6 days. This equates to a 50% risk of upset. Within these 9 scenarios, the average early warning of upset is 2.8 days. Of the 139 Condition B instances, 107 resulted in a secondary effluent upset event within 3 days. This equates to a 77% risk of upset. To divert an upset, the operations analysts can re-direct VFA flow from fermentate, and if necessary, add alum in the primary and/or secondary clarifiers. The Stability Analyzer has informed process decision making on a day to day basis and has proven useful in predicting BPR stability.
FINDING THE STABILITY POINT
After success with the Stability Analyzer on AB1, CWS wanted to demonstrate that the process could be replicated for basins with different geometries. AB5 is configured with a larger anaerobic and anoxic volume than AB1 (Figure 1, Table 1). In order to find the stability point in AB1, basin OP profiling efforts began in 2013 on all four identical aeration basins at the Durham WRRF (Figure 5). The profiles measure OP concentrations in multiple locations throughout the aerobic zone, with the goal of locating the stability point (Menniti et al, 2016). The stability point for AB1 was found 58% down the aerobic zone, referred to as Cell 7B (Figure 5). However, the stability analyzer was installed in the middle of Cell 7B, 52% down the aerobic zone, rather than where the sampling point was located due to inaccessibility for an analyzer to be placed. The profiling procedure was repeated for AB5 during its first summer of operation (2021) to determine the stability point, therefore informing stability analyzer installation location (Figure 6). The sampling points focused on for AB5 were accessible and appropriate locations for installing an analyzer. It is imperative to collect the profile samples at the same time of day due to diurnal patterns observed in carbon and phosphorus loading (Figure 7). Throughout summer 2021, AB5 was profiled once a week between 9-9:30am for 3.5 months, through stable times and during upset events. Determining the stability point and therefore the location for a stability analyzer is informed by where uptake is complete during time periods when no secondary effluent upset is observed. The OP Basin Profiling efforts on AB5 suggest the ideal location for predicting stability is at Cell 6C, where 45% of the aerated volume is completed (Figure 6). Figure 8 includes the profile results from the anaerobic and anoxic zones in addition to the aerobic zone. These profiles demonstrate that there is limited correlation between anaerobic zone release and corresponding strong aerobic zone uptake. In fact, the anaerobic release is highest during a secondary effluent upset event. We will be installing the stability analyzer on AB5 in January 2022.
CONCLUSIONS
The Stability Analyzer on AB1 provides an indication of BPR stability, and has enabled us to calculate for risk of secondary effluent upset given two prominent conditions, which have been empirically observed to provide early warning of upset events. Operations analysts have used the stability analyzer data in combination with other bioassay data to determine the need to re-distribute fermentate flow and decide on alum dosing approaches to divert a major upset event during stringent phosphorus permit limits. Despite the difference in anaerobic and anoxic volumes, we conclude that the stability points for the two basins with different geometries to be about halfway through the aerobic zone, with AB1 stability analyzer located at 52% and the AB5 stability point found at 45% down the aerobic zone of the basins.
<bF>UTURE WORK
We are currently performing OP Basin Profiling on two step feed basins at our Rock Creek WRRF. The results of this effort will be included in the full paper. Operational experience with the AB5 stability analyzer will also be included from its first 9 months in service, specifically focusing on summer operation when upsets are likely to occur while under strict phosphorus permit limits.

CWS has demonstrated the effectiveness of an on-line orthophosphate analyzer installed halfway through the aerobic zone to predict impending BPR instability. The Stability Analyzer provides an indirect measure of phosphorus uptake and offers 3 days early warning for BPR upsets. This paper quantifies the risk of upset events using 5 years of historical Stability Analyzer data and describes the process of strategically locating the stability point on an operating aeration basin.

Presentation time

10:30:00

10:55:00

Session time

10:30:00

12:00:00

TopicAdvanced Level, Intelligent Water, Nutrients, Research and Innovation

Author(s)

Author(s)Adrienne Menniti1; Peter Schauer1; Skylar Watnick1;

Author affiliation(s)Clean Water Services, Tigard, OR1

SourceProceedings of the Water Environment Federation

Document typeConference Paper

Use of Bio-electrochemical Sensors to Inform Enhanced Biological Phosphorus Removal Operation Under RAS Fermentation and A2O Conditions

Print publication date Oct 2022

DOI10.2175/193864718825158551

Volume / Issue

Content sourceWEFTEC

Word count11

Description: Use of Bio-electrochemical Sensors to Inform Enhanced Biological Phosphorus Removal...

Abstract

This paper was presented at WEFTEC 2023 in Chicago, IL.

This paper summarizes the performance bio-electrochemical sensors deployed at the Clean Water Services Durham WRRF with the following goals to support EBPR operation: 1. Guide RAS-fermentation operation to determine the optimal setpoints to create stable EBPR. 2. Compare anaerobic zone performance during A2O operation between two aeration basins where one basin has an anaerobic zone that is twice as large as the other. 3. Inform plant operations of process influent carbon loadings.

Presentation time

13:30:00

14:00:00

Session time

13:30:00

15:00:00

SessionConnecting Molecular Biology Sensors to EBPR Performance

Session locationRoom S502 - Level 5

TopicAdvanced Level, Facility Operations and Maintenance, Municipal Wastewater Treatment Design, Nutrients

Author(s)

Author(s)A. Menniti 1; J. Ambler 2 ; J. Mejia Franco 3; R. Cusick 4; A. Menniti 1;

Author affiliation(s)Chemical Engineering, Université Laval, 1065 Avenue de la Médecine, Québec, QC, Canada 1; SENTRY Water Technologies 2 ; SENTRY Water Technologies 3; University of Illinois at Urbana-Champaign 4; Clean Water Services 1;

SourceProceedings of the Water Environment Federation

Document typeConference Paper

WRF 5121 Final Report Overview: Long-Term Evaluation of Hybrid Digital Twins at Three Full-Scale WRRFs

Print publication date Oct 2023

DOI10.2175/193864718825159084

Volume / Issue

Content sourceWEFTEC

Word count18

Abstract

INTRODUCTION The Water Research Foundation (WRF) project 5121, Development of Innovative Predictive Control Strategies for Nutrient Removal, aims at developing full-scale hybrid nutrient management controllers (named here as Hybrid Optimizer, HO) by employing a hybrid approach that combines both machine learning and mechanistic models. Three full-scale pilots are deployed and evaluated for this project to demonstrate long-term controller abilities. The concept, design and preliminary results have been shared previously [1-7], which triggered wide interest and heated discussion. Over the past year, refinements were made to improve performance and evaluation approaches were standardized. Currently, all three pilots are fully deployed and operational. This paper provides an overview of three case studies. The control for three pilots is advisory only with no direct control functions due to cybersecurity concerns. Instead, operators receive daily email notifications outlining recommended operations and/or access a web-based user interface (UI), guiding their actions. The full-scale pilots are: 1. Clean Water Services Durham WRRF (CWS): Primary clarifier alum addition setpoints once a day, aiming to stabilize orthophosphate load to the downstream biological treatment system. 2. Agua Nueva WRF (AN): Dissolved oxygen setpoints twice daily for the bioreactor systems, aiming to minimize overall aeration energy consumption. 3.AlexRenew WRF (AR): HO recommends three equalization setpoints once a day to equalize the influent loadings to the bioreactor to minimize energy, chemicals, and blower staging. METHODOLOGY As shown in Figure 1, Hybrid Optimizer consists of seven major components. Descriptions of each component can be found in [1-3]. In this paper, the performance of Soft Sensor, Influent Forecaster and emulator/optimizer is mainly presented and discussed. Four different types of standard graphs are used to evaluate the performance of the HO. (1)Control Chart. Directly conveys the validity of the model, based on which warnings and alarms are derived. It integrates the measured and the modelled values into a dimensionless metric (t_score) and allows its comparison with predefined bounds. (2)Process Plot. The process plot presents the measured and the modelled values. (3)Unscaled Error Plot. The Error plot presents the residuals between the modelled and the measured. (4)XY Plot. The XY Plot presents the Modelled v. s. the Measured and the prediction intervals. (not shown here for brevity) RESULTS AND DISCUSSION The components and timelines of the three pilots are listed in Table 1. The Graphic User Interface (GUI) for two pilots are in progress. Evaluation results are available for all three, however this abstract presents only partial results. The soft sensor can be considered a full digital twin of the modelled components. Thus, it provides dynamic 15-minute frequency data on all wastewater components in the model scope. For CWS the primary influent and effluent ortho-phosphate data was used to support primary alum addition recommendations, without an actual sensor. For AN the focus was ammonia. The soft sensor demonstrates alignment with measured data (Figure 2).

*In the Control Chart, the t score falls within the warning and alarm bounds (set at 95% and 99% confidence intervals), which indicates soft sensor performance. The t-score is the normalized metric of a given set of measurements.

*The Process and Residual Plots are more intuitive for process engineers to evaluate. The modelled and measured values align closely, and their parallel trends indicate good soft sensor performance. Figure 3 shows the AN performance for primary influent and effluent ammonia. The top graph shows a good match between the composite samples. However, the primary effluent in the middle and bottom graphs shows a poor match. It was found that high influent sulfide was biasing the measurements high. The bottom graph showing the dynamic concentration variation, other than the bias, shows a good variability match. Flow forecasting is critical to predictive control. The Forecaster uses the available weather forecast and other variables to estimate plant influent flow. Figure 4 compares influent flow (forecasted vs actual) as well as the forecasted versus actual rainfall. A close examination shows that much of the influent forecast error can be attributed to the error in the precipitation forecast. The AR optimizer (Figure 5) aims to equalize ammonia load, which helps stabilize the aeration demand and optimization of aeration and chemicals. The flow and loading prediction from the HO allows the optimizer to recommend three equalization setpoints for the next 24 hours. In the top graph, the green line is the target daily average loading, red and blue show the load with and without the recommendations. CONCLUSIONS This abstract presents abbreviated results of the long-term performance of HO during three full-scale pilot studies. The HO appears to be a promising opportunity to improve plant operation using digital twin technology by giving both optimized operational recommendations and increased insight into the facility's status. The ongoing assessment of the HO performance is anticipated to yield further valuable practical experiences and insights, including the benefits and costs of such applications.

This paper was presented at the WEFTEC 2024 conference in New Orleans, LA October 5-9.

SpeakerYang, Cheng

Presentation time

13:30:00

13:50:00

Session time

13:30:00

15:00:00

SessionLeveraging Automation and Analytics for Better Situational Awareness and Optimization: Part I

Session number205

Session locationRoom 350

TopicEnergy Production, Conservation, and Management, Facility Operations and Maintenance, Intelligent Water, Intermediate Level, Nutrients

Author(s)

Yang, Cheng, Johnson, Bruce, Lesnik, Keaton, Registe, Joshua, Rieger, Leiv, Stewart, Heather, Miletic, Ivan, Menniti, Adrienne, Oristian, Monica, Pienta, Drew, Mason, Tim

Author(s)C. Yang1, B.R. Johnson1, K. Lesnik2, J. Registe3, L.P. Rieger4, H.A. Stewart5, I. Miletic6, A. Menniti7, M.K. Oristian8, D.J. Pienta9, T. Mason10

Author affiliation(s)1Jacobs, CO, 2Maia Analytica, OR, 3Jacobs Engineering, NJ, 4Jacobs, NL, 5Jacobs, PA, 6inCTRL Solutions Inc., ON, 7Clean Water Services, OR, 8Alexandria Renew Enterprise, VA, 9Jacobs, 10Jacobs Eng, AZ

SourceProceedings of the Water Environment Federation

Document typeConference Paper