Aeration accounts for a large fraction of energy consumption in conventional water reclamation pl... more Aeration accounts for a large fraction of energy consumption in conventional water reclamation plants (WRPs). Although process operations at older WRPs can satisfy effluent permit requirements, they typically operate with excess aeration. More effective process controls at older WRPs can be challenging as operators work to balance higher energy costs and more stringent effluent limitations while managing fluctuating loads. Therefore, understandings of process resilience or ability to quickly return to original operation conditions at a WRP are important. A state-of-art WRP should maintain process resilience to deal with different kinds of perturbations even after optimization of energy demands. This work was to evaluate the applicability and feasibility of cyber-physical system (CPS) for improving operation at Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) Calumet WRP. In this work, a process model was developed and used to better understand the conditions of current Calumet WRP, with additional valuable information from two dissolved oxygen field measurements. Meanwhile, a classification system was developed to reveal the pattern of historical influent scenario based on cluster analysis and cross-tabulation analysis. Based on the results from the classification, typical process control options were investigated. To ensure the feasibility of information acquisition, the reliability and flexibility of soft sensors were assessed to typical influent conditions. Finally, the process resilience was investigated to better balance influent perturbations, energy demands, and effluent quality for long-term operations. These investigations and evaluations show that although the energy demands change as the influent conditions and process controls, in general, aeration savings could be up to 50% from the level of current consumption; with a more complex process controls, the saving could be up to 70% in relatively steady-state conditions and at least 40% in relatively challenging transient conditions. The soft sensors can provide reliable and flexible performance on target predictions. The plant can still maintain at a similar level of process resilience after 50% aeration saving, even during long-term perturbations. Overall, this work shows that it is well feasible to provide more cost-effective operations at the Calumet WRP, and meanwhile influent perturbations, effluent quality, and process resilience are well in balance.
Fox River water was supersaturated with respect to calcite; natural organic matter (NOM) might pl... more Fox River water was supersaturated with respect to calcite; natural organic matter (NOM) might play a key role in this phenomenon. Fox River NOM (FRNOM) adsorption on the calcite surface is probably an important mechanism to explain this condition. Fox River water contained moderate ultraviolet absorbance (UVA) of NOM (0.19 1/cm), high concentration of calcium (70 mg/L), suspended solids with relatively high specific surface area (SSA) (6.9 m2/g), and moderate pH value (8.4) based on historical data. To test whether the phenomenon was caused by NOM adsorption, a series of experiments was conducted to explore the interaction between NOM and calcite in conditions similar to those of the Fox River. Suwannee River NOM (SRNOM) and Nordic Reservoir NOM (NRNOM) were used as surrogate NOM. The results show that SRNOM inhibited calcite dissolution significantly after 10 min based on measuring of the decrease in the free calcium concentration. The decrease in the free calcium was not solely due to formation of NOM-calcium complexes, because these complexes made up only about 3% of the total free calcium concentration. Therefore, NOM adsorption onto calcite was probably largely responsible for the inhibited calcite activity. Experimental results also showed that NOM adsorption increased with increasing NOM concentration in the range from 2 to 14 mg NOM/L, which is a common range for river water. Higher charge density also seems to promote sorption onto calcite; relative to NRNOM, SRNOM has a higher charge density and SRNOM has a higher affinity for calcite. Other factors that promoted NOM adsorption onto calcite included higher concentration of calcium and larger SSA of calcite seed. Based on water quality characteristics, the Fox River provides a suitable environment for NOM adsorption on calcite, and it seems likely that Fox River NOM (FRNOM) adsorption on calcite can inhibit calcite precipitation. This understanding of interaction between NOM and calcite could be used by WTPs along the Fox River for better optimization and improvement in treatment and operation
Better municipal solid waste (MSW) management can help to address environmental concerns and supp... more Better municipal solid waste (MSW) management can help to address environmental concerns and supports economic and social development. Because MSW characteristics can change over time, management strategies should also evolve and be applied correspondingly. However, many previous studies have focused on MSW characterization or investigated specific management strategies for a target MSW. Few studies have assessed the spatial variations of MSW characteristics and socio-economic (SE) conditions as well as their associations. This study evaluated the feasibility of using an integrated unsupervised method (cluster analysis and cross-tabulation analysis) to explore these topics for MSW management. Results suggest that the integrated method can successfully help to reveal key information. Seven jointed MSW-SE scenarios were investigated based on 259 individual observations of Taiwan. Associations between MSW compositions and SE conditions were identified statistically significant for four MSW-SE scenarios. In general, a general SE type (SE1) is very likely to generate high food wastes and other combustible, low paper, wood, and rubber wastes. The small island SE type (SE3) is more likely to produce high paper and low wood, rubber, textile, and other noncombustible wastes. Overall, the method applied in this study could help to reveal statistical associations between MSW and SE, which can help decision-makers comprehend underlying facts and develop effective management strategies.
This study evaluated a fuzzy technique for order performance by similarity to ideal solution (TOP... more This study evaluated a fuzzy technique for order performance by similarity to ideal solution (TOPSIS) as a multicriteria decision making system that compensates for missing information with undefined weight factor criteria. The suggested Fuzzy TOPSIS was applied to ten potential dam sites in three river basins (the Han River, the Geum River, and the Nakdong River basins) in South Korea. To assess potential dam sites, the strategic environment assessment (SEA) monitored four categories: national preservation, endangered species, water quality, and toxic environment. To consider missing information, this study applied the Monte Carlo Simulation method with uniform and normal distributions. The results show that effects of missing information generation with one fuzzy set in GB1 site of the Geum River basin are not great in fuzzy positive-ideal solution (FPIS) and fuzzy negative-ideal solution (FNIS) estimations. However, the combination of two fuzzy sets considering missing informatio...
Prediction of natural gas demand can help to better manage energy demand and supply and recent de... more Prediction of natural gas demand can help to better manage energy demand and supply and recent developments of deep learning methods make it possible to improve forecast performance. This study examined the feasibility of hourly natural gas demand forecast, and compared statistical and deep learning methods to evaluate their prediction performance at five different sites in Spain. Hourly forecast can achieve an adjusted R2 ∼0.99 and MAPE lower to 2.7%. SMLR yields high prediction accuracy (MAPE: 3%–10%) in four sites, but suffers issues of missing data (5%–17%) and has relatively more extreme predictions (653 observations; ±100% away from the values). MLP has less amount of extreme predictions (517 observations) with a similar accuracy (MAPE: 3%–11%), but still suffers issue of missing data (11%–28%). LSTMs also achieves good prediction accuracy (MAPE: 3%–13%) and is able to manage most extreme values. Other methods are generally less promising but can be site-specific. Understanding their distinctive characteristics can help decision-makers to rule better decisions. Exploration of future forecasts based on LSTMs shows promising (adjusted R2: 0.90–0.99; MAPE: 11%–32%) in near future (<7 h), while model optimization can be used to further improve the performance, especially for a longer gap.
AbstractThe margin of safety (MOS) accounts for uncertainties in the total maximum daily load (TM... more AbstractThe margin of safety (MOS) accounts for uncertainties in the total maximum daily load (TMDL) development process and the variabilities involved in simulating systems, providing a complete d...
Nitrogen-containing wastewater is an important issue in optoelectronic and semiconductor industri... more Nitrogen-containing wastewater is an important issue in optoelectronic and semiconductor industries. Wastewater containing nitrogen compounds such as ammonium, monoethanolamine (MEA), and tetramethylammonium hydroxide (TMAH) must be properly treated due to concerns about health and environmental effects. MnCe-GAC (granular activated carbon) processes were developed in this study for the treatment of TMAH-contaminated wastewater in high-tech industries. The MnCe-GAC processes could effectively remove ammonium, MEA, and TMAH from aqueous solutions. The removal efficiencies of ammonium and MEA by these processes were better than observed for TMAH. Parameters affecting TMAH removal such as type of process, type of wastewater (synthetic or real), pH, salts, and t-butanol were investigated. In general, removal efficiencies of TMAH by various processes were in the following order: MnCe-GAC/O3/H2O2 > MnCe-GAC/O3 > MnCe-GAC/H2O2 > MnCe-GAC > GAC. The negative effect of sulfate an...
Soft-sensor applications for wastewater management can provide valuable information for intellige... more Soft-sensor applications for wastewater management can provide valuable information for intelligent monitoring and process control above and beyond what is available from conventional hard sensors and laboratory measurements. To realize these benefits, it is important to know how to manage gaps in the data time series, which could result from the failure of hard sensors, errors in laboratory measurements, or low-frequency monitoring schedules. A robust soft-sensor system needs to include a plan to address missing data and efficiently select variable(s) to make the most use of the available information. In this study, we developed and applied an enhanced iterated stepwise multiple linear regression (ISMLR) method through a MATLAB-based package to predict the next day's influent flowrate at the Kirie water reclamation plant (WRP). The method increased the data retention from 77% to 93% and achieved an adjusted R2 up to 0.83 by integrating with a typical artificial neural network.
This study analyzed the result of parameter optimization using the digital elevation model (DEM) ... more This study analyzed the result of parameter optimization using the digital elevation model (DEM) resolution in the TOPography-based hydrological MODEL (TOPMODEL). Also, this study investigated the sensitivity of the TOPMODEL efficiency by applying the varying resolution of the DEM grid cell size. This work applied TOPMODEL to two mountainous watersheds in South Korea: the Dongkok watershed in the Wicheon river basin and the Ieemokjung watershed in the Pyeongchang river basin. The DEM grid cell sizes were 5, 10, 20, 40, 80, 160, and 300 m. The effect of DEM grid cell size on the runoff was investigated by using the DEM grid cell size resolution to optimize the parameter sets. As the DEM grid cell size increased, the estimated peak discharge was found to increase based on different parameter sets. In addition, this study investigated the DEM grid cell size that was most reliable for use in runoff simulations with various parameter sets in the experimental watersheds. The results demon...
Ready access to comprehensive influent information can help water reclamation plant (WRP) operato... more Ready access to comprehensive influent information can help water reclamation plant (WRP) operators implement better real-time process controls, provide operational reliability and reduce energy consumption. The five-day biochemical oxygen demand (BOD5), a critical parameter for WRP process control, is expensive and difficult to measure using hard-sensors. An alternative approach based on a soft-sensor methodology shows promise, but can be problematic when used to predict high BOD5 values. Underestimating high BOD5 concentrations for process control could result in an insufficient amount of aeration, increasing the risk of an effluent violation. To address this issue, we tested a hierarchical hybrid soft-sensor approach involving multiple linear regression, artificial neural networks (ANN), and compromise programming. While this hybrid approach results in a slight decrease in overall prediction accuracy relative to the approach based on ANN only, the underestimation percentage is su...
Aeration accounts for a large fraction of energy consumption at conventional water reclamation pl... more Aeration accounts for a large fraction of energy consumption at conventional water reclamation plants (WRPs). Older plants were designed when control techniques were relatively primitive and energy consumption was less of a concern. As a result, although process operations at older WRPs can satisfy effluent permit requirements, they can operate with excess aeration. In this study, we developed a wastewater process model to evaluate possible aeration savings at the Metropolitan Water Reclamation District of Greater Chicago Calumet WRP, one of the oldest plants in Chicago. Based on subsets of influent characteristics, we identified eight steady-state scenarios. We also identified transient scenarios that included high probability perturbations and more challenging but lower probability conditions. Results indicate that the Calumet WRP frequently operates with excess aeration. Effluent dissolved oxygen is the limiting parameter with respect to aeration saving and permit requirements. I...
A water reclamation plant (WRP) needs to be resilient to successfully operate through different k... more A water reclamation plant (WRP) needs to be resilient to successfully operate through different kinds of perturbations. Perturbations such as storm events, especially long-term successive storm flows, can adversely affect operations. A better understanding of these effects can provide benefits for plant operation, in terms of effluent quality and energy efficiency. However, the concept of resilience for a WRP has not been widely studied, and we are not aware of any studies specifically related to storm flows. In this work we applied measures of resistance and recovery time to quantify resilience, and used a WRP simulation model to investigate how different storm flow characteristics (flowrate and duration) and the amount of aeration influence resilience. Not surprisingly, increasing storm flowrate leads to decreasing resilience. Although the aeration rate plays an important role in determining resilience, there is an aeration threshold (6 m(3)/s for our WRP model); higher aeration r...
AbstractMany older water reclamation plants (WRPs) are implementing model-based process control t... more AbstractMany older water reclamation plants (WRPs) are implementing model-based process control to satisfy increasingly stringent effluent requirements while they lower energy costs, but these methods require reliable process information. Soft sensors can help to provide such information by building on easily acquirable and historical data. In this investigation of a soft-sensor approach at a conventional WRP, many historical data were missing, which suggested that a large fraction of the information would be lost if the missing data were not well managed. This study applied an iterated stepwise multiple linear regression (ISMLR) approach to minimize the loss of data and to predict real-time influent ammonia and CBOD5 (5-day carbonaceous biochemical oxygen demand), and future influent flow at the Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) Calumet WRP. Relative to a simple deletion method (which retained about 45% of the daily data), the ISMLR approach successfully retained substan...
Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF... more Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF) operators and engineers with the ambition of creating a truly circular economy and assuring a sustainable future. Despite the successful application of mechanistic models in the water sector, they show some important limitations and do not fully profit from the increasing digitalisation of systems and processes. Recent advances in data-driven methods have provided options for harnessing the power of Industry 4.0, but they are often limited by the lack of interpretability and extrapolation capabilities. Hybrid modelling (HM) combines these two modelling paradigms and aims to leverage both the rapidly increasing volumes of data collected, as well as the continued pursuit of greater process understanding. Despite the potential of HM in a sector that is undergoing a significant digital and cultural transformation, the application of hybrid models remains vague. This article presents an ove...
Many wastewater utilities have discharge permits directly tied with the receiving river flow, so ... more Many wastewater utilities have discharge permits directly tied with the receiving river flow, so it is critical to have accurate prediction of the hydraulic throughput to ensure safe operation and environment protection. Current empirical knowledge-based operation faces many challenges, so in this study we developed and assessed daily-adaptive, probabilistic soft sensor prediction models to forecast the next month's average receiving river flowrate and guide the utility operations. By comparing 11 machine-learning methods, extra trees regression exhibits desired deterministic prediction accuracy at day 0 (overall accuracy index: 3.9 × 10−3 1/cms2) (cms: cubic meter per second), which also increases steadily over the course of the month (e.g., MAPE and RMSE decrease from 41.46% and 23.31 cms to 3.31% and 2.81 cms, respectively). The overall classification accuracy of three river flow classes reaches 0.79 at the beginning and increases to about 0.97 over the course of the predicted month. To manage the uncertainty caused by potential false negative classification as overestimations, a probabilistic assessment on the predictions based on 95% lower PI is developed and successfully reduces the false negative classification from 17% to nearly zero with a slight sacrifice of overall classification accuracy.
Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF... more Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF) operators and engineers with the ambition of creating a truly circular economy and assuring a sustainable future. Despite the successful application of mechanistic models in the water sector, they show some important limitations and do not fully profit from the increasing digitalisation of systems and processes. Recent advances in data-driven methods have provided options for harnessing the power of Industry 4.0, but they are often limited by the lack of interpretability and extrapolation capabilities. Hybrid modelling (HM) combines these two modelling paradigms and aims to leverage both the rapidly increasing volumes of data collected, as well as the continued pursuit of greater process understanding. Despite the potential of HM in a sector that is undergoing a significant digital and cultural transformation, the application of hybrid models remains vague. This article presents an overview of HM methodologies applied to WRRFs and aims to stimulate the wider adoption and development of HM. We also highlight challenges and research needs for HM design and architecture, good modelling practice, data assurance, and software compatibility. HM is a paradigm for WRRF modelling to transition towards a more resource-efficient, resilient, and sustainable future.
Prediction of natural gas demand can help to better manage energy demand and supply and recent de... more Prediction of natural gas demand can help to better manage energy demand and supply and recent developments of deep learning methods make it possible to improve forecast performance. This study examined the feasibility of hourly natural gas demand forecast, and compared statistical and deep learning methods to evaluate their prediction performance at five different sites in Spain. Hourly forecast can achieve an adjusted R2 ∼0.99 and MAPE lower to 2.7%. SMLR yields high prediction accuracy (MAPE: 3%–10%) in four sites, but suffers issues of missing data (5%–17%) and has relatively more extreme predictions (653 observations; ±100% away from the values). MLP has less amount of extreme predictions (517 observations) with a similar accuracy (MAPE: 3%–11%), but still suffers issue of missing data (11%–28%). LSTMs also achieves good prediction accuracy (MAPE: 3%–13%) and is able to manage most extreme values. Other methods are generally less promising but can be site-specific. Understanding their distinctive characteristics can help decision-makers to rule better decisions. Exploration of future forecasts based on LSTMs shows promising (adjusted R2: 0.90–0.99; MAPE: 11%–32%) in near future (<7 h), while model optimization can be used to further improve the performance, especially for a longer gap.
Aeration accounts for a large fraction of energy consumption in conventional water reclamation pl... more Aeration accounts for a large fraction of energy consumption in conventional water reclamation plants (WRPs). Although process operations at older WRPs can satisfy effluent permit requirements, they typically operate with excess aeration. More effective process controls at older WRPs can be challenging as operators work to balance higher energy costs and more stringent effluent limitations while managing fluctuating loads. Therefore, understandings of process resilience or ability to quickly return to original operation conditions at a WRP are important. A state-of-art WRP should maintain process resilience to deal with different kinds of perturbations even after optimization of energy demands. This work was to evaluate the applicability and feasibility of cyber-physical system (CPS) for improving operation at Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) Calumet WRP. In this work, a process model was developed and used to better understand the conditions of current Calumet WRP, with additional valuable information from two dissolved oxygen field measurements. Meanwhile, a classification system was developed to reveal the pattern of historical influent scenario based on cluster analysis and cross-tabulation analysis. Based on the results from the classification, typical process control options were investigated. To ensure the feasibility of information acquisition, the reliability and flexibility of soft sensors were assessed to typical influent conditions. Finally, the process resilience was investigated to better balance influent perturbations, energy demands, and effluent quality for long-term operations. These investigations and evaluations show that although the energy demands change as the influent conditions and process controls, in general, aeration savings could be up to 50% from the level of current consumption; with a more complex process controls, the saving could be up to 70% in relatively steady-state conditions and at least 40% in relatively challenging transient conditions. The soft sensors can provide reliable and flexible performance on target predictions. The plant can still maintain at a similar level of process resilience after 50% aeration saving, even during long-term perturbations. Overall, this work shows that it is well feasible to provide more cost-effective operations at the Calumet WRP, and meanwhile influent perturbations, effluent quality, and process resilience are well in balance.
Fox River water was supersaturated with respect to calcite; natural organic matter (NOM) might pl... more Fox River water was supersaturated with respect to calcite; natural organic matter (NOM) might play a key role in this phenomenon. Fox River NOM (FRNOM) adsorption on the calcite surface is probably an important mechanism to explain this condition. Fox River water contained moderate ultraviolet absorbance (UVA) of NOM (0.19 1/cm), high concentration of calcium (70 mg/L), suspended solids with relatively high specific surface area (SSA) (6.9 m2/g), and moderate pH value (8.4) based on historical data. To test whether the phenomenon was caused by NOM adsorption, a series of experiments was conducted to explore the interaction between NOM and calcite in conditions similar to those of the Fox River. Suwannee River NOM (SRNOM) and Nordic Reservoir NOM (NRNOM) were used as surrogate NOM. The results show that SRNOM inhibited calcite dissolution significantly after 10 min based on measuring of the decrease in the free calcium concentration. The decrease in the free calcium was not solely due to formation of NOM-calcium complexes, because these complexes made up only about 3% of the total free calcium concentration. Therefore, NOM adsorption onto calcite was probably largely responsible for the inhibited calcite activity. Experimental results also showed that NOM adsorption increased with increasing NOM concentration in the range from 2 to 14 mg NOM/L, which is a common range for river water. Higher charge density also seems to promote sorption onto calcite; relative to NRNOM, SRNOM has a higher charge density and SRNOM has a higher affinity for calcite. Other factors that promoted NOM adsorption onto calcite included higher concentration of calcium and larger SSA of calcite seed. Based on water quality characteristics, the Fox River provides a suitable environment for NOM adsorption on calcite, and it seems likely that Fox River NOM (FRNOM) adsorption on calcite can inhibit calcite precipitation. This understanding of interaction between NOM and calcite could be used by WTPs along the Fox River for better optimization and improvement in treatment and operation
Better municipal solid waste (MSW) management can help to address environmental concerns and supp... more Better municipal solid waste (MSW) management can help to address environmental concerns and supports economic and social development. Because MSW characteristics can change over time, management strategies should also evolve and be applied correspondingly. However, many previous studies have focused on MSW characterization or investigated specific management strategies for a target MSW. Few studies have assessed the spatial variations of MSW characteristics and socio-economic (SE) conditions as well as their associations. This study evaluated the feasibility of using an integrated unsupervised method (cluster analysis and cross-tabulation analysis) to explore these topics for MSW management. Results suggest that the integrated method can successfully help to reveal key information. Seven jointed MSW-SE scenarios were investigated based on 259 individual observations of Taiwan. Associations between MSW compositions and SE conditions were identified statistically significant for four MSW-SE scenarios. In general, a general SE type (SE1) is very likely to generate high food wastes and other combustible, low paper, wood, and rubber wastes. The small island SE type (SE3) is more likely to produce high paper and low wood, rubber, textile, and other noncombustible wastes. Overall, the method applied in this study could help to reveal statistical associations between MSW and SE, which can help decision-makers comprehend underlying facts and develop effective management strategies.
This study evaluated a fuzzy technique for order performance by similarity to ideal solution (TOP... more This study evaluated a fuzzy technique for order performance by similarity to ideal solution (TOPSIS) as a multicriteria decision making system that compensates for missing information with undefined weight factor criteria. The suggested Fuzzy TOPSIS was applied to ten potential dam sites in three river basins (the Han River, the Geum River, and the Nakdong River basins) in South Korea. To assess potential dam sites, the strategic environment assessment (SEA) monitored four categories: national preservation, endangered species, water quality, and toxic environment. To consider missing information, this study applied the Monte Carlo Simulation method with uniform and normal distributions. The results show that effects of missing information generation with one fuzzy set in GB1 site of the Geum River basin are not great in fuzzy positive-ideal solution (FPIS) and fuzzy negative-ideal solution (FNIS) estimations. However, the combination of two fuzzy sets considering missing informatio...
Prediction of natural gas demand can help to better manage energy demand and supply and recent de... more Prediction of natural gas demand can help to better manage energy demand and supply and recent developments of deep learning methods make it possible to improve forecast performance. This study examined the feasibility of hourly natural gas demand forecast, and compared statistical and deep learning methods to evaluate their prediction performance at five different sites in Spain. Hourly forecast can achieve an adjusted R2 ∼0.99 and MAPE lower to 2.7%. SMLR yields high prediction accuracy (MAPE: 3%–10%) in four sites, but suffers issues of missing data (5%–17%) and has relatively more extreme predictions (653 observations; ±100% away from the values). MLP has less amount of extreme predictions (517 observations) with a similar accuracy (MAPE: 3%–11%), but still suffers issue of missing data (11%–28%). LSTMs also achieves good prediction accuracy (MAPE: 3%–13%) and is able to manage most extreme values. Other methods are generally less promising but can be site-specific. Understanding their distinctive characteristics can help decision-makers to rule better decisions. Exploration of future forecasts based on LSTMs shows promising (adjusted R2: 0.90–0.99; MAPE: 11%–32%) in near future (&lt;7 h), while model optimization can be used to further improve the performance, especially for a longer gap.
AbstractThe margin of safety (MOS) accounts for uncertainties in the total maximum daily load (TM... more AbstractThe margin of safety (MOS) accounts for uncertainties in the total maximum daily load (TMDL) development process and the variabilities involved in simulating systems, providing a complete d...
Nitrogen-containing wastewater is an important issue in optoelectronic and semiconductor industri... more Nitrogen-containing wastewater is an important issue in optoelectronic and semiconductor industries. Wastewater containing nitrogen compounds such as ammonium, monoethanolamine (MEA), and tetramethylammonium hydroxide (TMAH) must be properly treated due to concerns about health and environmental effects. MnCe-GAC (granular activated carbon) processes were developed in this study for the treatment of TMAH-contaminated wastewater in high-tech industries. The MnCe-GAC processes could effectively remove ammonium, MEA, and TMAH from aqueous solutions. The removal efficiencies of ammonium and MEA by these processes were better than observed for TMAH. Parameters affecting TMAH removal such as type of process, type of wastewater (synthetic or real), pH, salts, and t-butanol were investigated. In general, removal efficiencies of TMAH by various processes were in the following order: MnCe-GAC/O3/H2O2 > MnCe-GAC/O3 > MnCe-GAC/H2O2 > MnCe-GAC > GAC. The negative effect of sulfate an...
Soft-sensor applications for wastewater management can provide valuable information for intellige... more Soft-sensor applications for wastewater management can provide valuable information for intelligent monitoring and process control above and beyond what is available from conventional hard sensors and laboratory measurements. To realize these benefits, it is important to know how to manage gaps in the data time series, which could result from the failure of hard sensors, errors in laboratory measurements, or low-frequency monitoring schedules. A robust soft-sensor system needs to include a plan to address missing data and efficiently select variable(s) to make the most use of the available information. In this study, we developed and applied an enhanced iterated stepwise multiple linear regression (ISMLR) method through a MATLAB-based package to predict the next day's influent flowrate at the Kirie water reclamation plant (WRP). The method increased the data retention from 77% to 93% and achieved an adjusted R2 up to 0.83 by integrating with a typical artificial neural network.
This study analyzed the result of parameter optimization using the digital elevation model (DEM) ... more This study analyzed the result of parameter optimization using the digital elevation model (DEM) resolution in the TOPography-based hydrological MODEL (TOPMODEL). Also, this study investigated the sensitivity of the TOPMODEL efficiency by applying the varying resolution of the DEM grid cell size. This work applied TOPMODEL to two mountainous watersheds in South Korea: the Dongkok watershed in the Wicheon river basin and the Ieemokjung watershed in the Pyeongchang river basin. The DEM grid cell sizes were 5, 10, 20, 40, 80, 160, and 300 m. The effect of DEM grid cell size on the runoff was investigated by using the DEM grid cell size resolution to optimize the parameter sets. As the DEM grid cell size increased, the estimated peak discharge was found to increase based on different parameter sets. In addition, this study investigated the DEM grid cell size that was most reliable for use in runoff simulations with various parameter sets in the experimental watersheds. The results demon...
Ready access to comprehensive influent information can help water reclamation plant (WRP) operato... more Ready access to comprehensive influent information can help water reclamation plant (WRP) operators implement better real-time process controls, provide operational reliability and reduce energy consumption. The five-day biochemical oxygen demand (BOD5), a critical parameter for WRP process control, is expensive and difficult to measure using hard-sensors. An alternative approach based on a soft-sensor methodology shows promise, but can be problematic when used to predict high BOD5 values. Underestimating high BOD5 concentrations for process control could result in an insufficient amount of aeration, increasing the risk of an effluent violation. To address this issue, we tested a hierarchical hybrid soft-sensor approach involving multiple linear regression, artificial neural networks (ANN), and compromise programming. While this hybrid approach results in a slight decrease in overall prediction accuracy relative to the approach based on ANN only, the underestimation percentage is su...
Aeration accounts for a large fraction of energy consumption at conventional water reclamation pl... more Aeration accounts for a large fraction of energy consumption at conventional water reclamation plants (WRPs). Older plants were designed when control techniques were relatively primitive and energy consumption was less of a concern. As a result, although process operations at older WRPs can satisfy effluent permit requirements, they can operate with excess aeration. In this study, we developed a wastewater process model to evaluate possible aeration savings at the Metropolitan Water Reclamation District of Greater Chicago Calumet WRP, one of the oldest plants in Chicago. Based on subsets of influent characteristics, we identified eight steady-state scenarios. We also identified transient scenarios that included high probability perturbations and more challenging but lower probability conditions. Results indicate that the Calumet WRP frequently operates with excess aeration. Effluent dissolved oxygen is the limiting parameter with respect to aeration saving and permit requirements. I...
A water reclamation plant (WRP) needs to be resilient to successfully operate through different k... more A water reclamation plant (WRP) needs to be resilient to successfully operate through different kinds of perturbations. Perturbations such as storm events, especially long-term successive storm flows, can adversely affect operations. A better understanding of these effects can provide benefits for plant operation, in terms of effluent quality and energy efficiency. However, the concept of resilience for a WRP has not been widely studied, and we are not aware of any studies specifically related to storm flows. In this work we applied measures of resistance and recovery time to quantify resilience, and used a WRP simulation model to investigate how different storm flow characteristics (flowrate and duration) and the amount of aeration influence resilience. Not surprisingly, increasing storm flowrate leads to decreasing resilience. Although the aeration rate plays an important role in determining resilience, there is an aeration threshold (6 m(3)/s for our WRP model); higher aeration r...
AbstractMany older water reclamation plants (WRPs) are implementing model-based process control t... more AbstractMany older water reclamation plants (WRPs) are implementing model-based process control to satisfy increasingly stringent effluent requirements while they lower energy costs, but these methods require reliable process information. Soft sensors can help to provide such information by building on easily acquirable and historical data. In this investigation of a soft-sensor approach at a conventional WRP, many historical data were missing, which suggested that a large fraction of the information would be lost if the missing data were not well managed. This study applied an iterated stepwise multiple linear regression (ISMLR) approach to minimize the loss of data and to predict real-time influent ammonia and CBOD5 (5-day carbonaceous biochemical oxygen demand), and future influent flow at the Metropolitan Water Reclamation District of Greater Chicago (MWRDGC) Calumet WRP. Relative to a simple deletion method (which retained about 45% of the daily data), the ISMLR approach successfully retained substan...
Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF... more Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF) operators and engineers with the ambition of creating a truly circular economy and assuring a sustainable future. Despite the successful application of mechanistic models in the water sector, they show some important limitations and do not fully profit from the increasing digitalisation of systems and processes. Recent advances in data-driven methods have provided options for harnessing the power of Industry 4.0, but they are often limited by the lack of interpretability and extrapolation capabilities. Hybrid modelling (HM) combines these two modelling paradigms and aims to leverage both the rapidly increasing volumes of data collected, as well as the continued pursuit of greater process understanding. Despite the potential of HM in a sector that is undergoing a significant digital and cultural transformation, the application of hybrid models remains vague. This article presents an ove...
Many wastewater utilities have discharge permits directly tied with the receiving river flow, so ... more Many wastewater utilities have discharge permits directly tied with the receiving river flow, so it is critical to have accurate prediction of the hydraulic throughput to ensure safe operation and environment protection. Current empirical knowledge-based operation faces many challenges, so in this study we developed and assessed daily-adaptive, probabilistic soft sensor prediction models to forecast the next month's average receiving river flowrate and guide the utility operations. By comparing 11 machine-learning methods, extra trees regression exhibits desired deterministic prediction accuracy at day 0 (overall accuracy index: 3.9 × 10−3 1/cms2) (cms: cubic meter per second), which also increases steadily over the course of the month (e.g., MAPE and RMSE decrease from 41.46% and 23.31 cms to 3.31% and 2.81 cms, respectively). The overall classification accuracy of three river flow classes reaches 0.79 at the beginning and increases to about 0.97 over the course of the predicted month. To manage the uncertainty caused by potential false negative classification as overestimations, a probabilistic assessment on the predictions based on 95% lower PI is developed and successfully reduces the false negative classification from 17% to nearly zero with a slight sacrifice of overall classification accuracy.
Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF... more Mathematical modelling is an indispensable tool to support water resource recovery facility (WRRF) operators and engineers with the ambition of creating a truly circular economy and assuring a sustainable future. Despite the successful application of mechanistic models in the water sector, they show some important limitations and do not fully profit from the increasing digitalisation of systems and processes. Recent advances in data-driven methods have provided options for harnessing the power of Industry 4.0, but they are often limited by the lack of interpretability and extrapolation capabilities. Hybrid modelling (HM) combines these two modelling paradigms and aims to leverage both the rapidly increasing volumes of data collected, as well as the continued pursuit of greater process understanding. Despite the potential of HM in a sector that is undergoing a significant digital and cultural transformation, the application of hybrid models remains vague. This article presents an overview of HM methodologies applied to WRRFs and aims to stimulate the wider adoption and development of HM. We also highlight challenges and research needs for HM design and architecture, good modelling practice, data assurance, and software compatibility. HM is a paradigm for WRRF modelling to transition towards a more resource-efficient, resilient, and sustainable future.
Prediction of natural gas demand can help to better manage energy demand and supply and recent de... more Prediction of natural gas demand can help to better manage energy demand and supply and recent developments of deep learning methods make it possible to improve forecast performance. This study examined the feasibility of hourly natural gas demand forecast, and compared statistical and deep learning methods to evaluate their prediction performance at five different sites in Spain. Hourly forecast can achieve an adjusted R2 ∼0.99 and MAPE lower to 2.7%. SMLR yields high prediction accuracy (MAPE: 3%–10%) in four sites, but suffers issues of missing data (5%–17%) and has relatively more extreme predictions (653 observations; ±100% away from the values). MLP has less amount of extreme predictions (517 observations) with a similar accuracy (MAPE: 3%–11%), but still suffers issue of missing data (11%–28%). LSTMs also achieves good prediction accuracy (MAPE: 3%–13%) and is able to manage most extreme values. Other methods are generally less promising but can be site-specific. Understanding their distinctive characteristics can help decision-makers to rule better decisions. Exploration of future forecasts based on LSTMs shows promising (adjusted R2: 0.90–0.99; MAPE: 11%–32%) in near future (<7 h), while model optimization can be used to further improve the performance, especially for a longer gap.
It is difficult to identify inorganic aerosol (IA) (primary and secondary), the main component of... more It is difficult to identify inorganic aerosol (IA) (primary and secondary), the main component of PM2.5, without the significant tracers for sources. We are not aware of any studies specifically related to the IA’s local contribution to PM2.5. To effectively reduce the IA load, however, the contribution of local IA sources needs to be identified. In this work, we developed a hybrid methodology and applied online measurement of PM2.5 and the associated compounds to (1) classify local and long-range transport PM2.5, (2) identify sources of local PM2.5 using PMF model, and (3) quantify local source contribution to IA in PM2.5 using regression analysis. Coal combustion and iron ore and steel industry contributed the most amount of IA (~42.7%) in the study area (City of Taichung), followed by 32.9% contribution from oil combustion, 8.9% from traffic-related emission, 4.6% from the interactions between agrochemical applications and combustion sources (traffic-related emissions and biomass burning), and 2.3% from biomass burning. The methodology developed in this study is an important preliminary step for setting up future control policies and regulations, which can also be applied to any other places with serious local air pollution.
The rapid increase in both the quantity and complexity of data that are being generated daily in ... more The rapid increase in both the quantity and complexity of data that are being generated daily in the field of environmental science and engineering (ESE) demands accompanied advancement in data analytics. Advanced data analysis approaches, such as machine learning (ML), have become indispensable tools for revealing hidden patterns or deducing correlations for which conventional analytical methods face limitations or challenges. However, ML concepts and practices have not been widely utilized by researchers in ESE. This feature explores the potential of ML to revolutionize data analysis and modeling in the ESE field, and covers the essential knowledge needed for such applications. First, we use five examples to illustrate how ML addresses complex ESE problems. We then summarize four major types of applications of ML in ESE: making predictions; extracting feature importance; detecting anomalies; and discovering new materials or chemicals. Next, we introduce the essential knowledge required and current shortcomings in ML applications in ESE, with a focus on three important but often overlooked components when applying ML: correct model development, proper model interpretation, and sound applicability analysis. Finally, we discuss challenges and future opportunities in the application of ML tools in ESE to highlight the potential of ML in this field.
The iterated stepwise multiple linear regression (ISMLR) application, a MATLAB based graphical us... more The iterated stepwise multiple linear regression (ISMLR) application, a MATLAB based graphical user interface tool, is designed to process a large time-series dataset for providing a primary prediction, minimizing missing data, selecting subset of important regressors, and/or serving as a screening step when a better algorithm is available. The application version 1.0 was published on August 01, 2017; the latest version 1.2 was published on July 11, 2018.
Uploads
Thesis Chapters by Junjie Zhu
Papers by Junjie Zhu