Application of Machine Learning for Fenceline Monitoring of Odor Classes and Concentrations at a Wastewater Treatment Plant

Implementation of an electronic nose for real -time identification of odour emission peaks at a wastewater treatment plant

This paper proposes a novel approach for the real-time monitoring of odour emissions from a WasteWater Treatment Plant (WWTP) using an Instrumental Odour Monitoring System (IOMS). The plant is characterized by unpredictable odour peaks at its arrival tank (AT), generating nuisance and complaints in the population living nearby the plant. Odour peaks are most likely due to the conferment of non-identified and malodorous wastewaters coming from various industrial activities. Due to the high variability of sources collecting their wastewaters to the WWTP, a new methodology to train the IOMS, based on the use of a one-class classifier (OCC), has been exploited. The OCC enables to detect deviations from a "Normal Operating Region" (NOR), defined as to include odour concentrations levels unlikely to cause nuisance in the citizenship. Such deviations from the NOR thus should be representative of the odour peaks. The results obtained prove that the IOMS is able to detect real-time alterations of odour emissions from the AT with an accuracy on independent validation data of about 90% (CI95% 55-100%). This ability of detecting anomalous conditions at the AT of the WWTP allowed the targeted withdrawal of liquid and gas samples in correspondence of the odour peaks, then subjected to further analyses that in turn enabled to investigate their origin and take proper counteractions to mitigate the WWTP odour impact.

A critical review on odor measurement and prediction

Odor pollution has become a global environmental issue of increasing concern in recent years. Odor measurements are the basis of assessing and solving odor problems. Olfactory and chemical analysis can be used for odor and odorant measurements. Olfactory analysis reflects the subjective perception of human, and chemical analysis reveals the chemical composition of odors. As an alternative to olfactory analysis, odor prediction methods have been developed based on chemical and olfactory analysis results. The combination of olfactory and chemical analysis is the best way to control odor pollution, evaluate the performances of the technologies, and predict odor. However, there are still some limitations and obstacles for each method, their combination, and the prediction. Here, we present an overview of odor measurement and prediction. Different olfactory analysis methods (namely, the dynamic olfactometry method and the triangle odor bag method) are compared in detail, the latest revisions of the standard olfactometry methods are summarized, and the uncertainties of olfactory measurement results (i.e., the odor thresholds) are analyzed. The researches, applications, and limitations of chemical analysis and odor prediction are introduced and discussed. Finally, the development and application of odor databases and algorithms for optimizing odor measurement and prediction methods are prospected, and a preliminary framework for an odor database is proposed. This review is expected to provide insights into odor measurement and prediction.

Development of an electronic nose to characterize water quality parameters and odor concentration of wastewater emitted from different phases in a wastewater treatment plant

For public health consideration, it is important to ensure the wastewater discharged from wastewater treatment plant is within the regulatory limits. This problem can be effectively solved by improving the accuracy and rapid characterization of water quality parameters and odor concentration of wastewater. In this paper, we proposed a novel solution to realize the precisive analysis of water quality parameters and odor concentration of wastewater by the electronic nose device. The main work of this paper was divided into three steps: 1) recognizing wastewater samples qualitatively from different sampling points, 2) analyzing the correlation between electronic nose response signals and water quality parameters and odor concentration, and 3) predicting the odor concentration and water quality parameters quantitatively. Combined with different feature extraction methods, support vector machine and linear discriminant analysis were applied as classifiers to recognize samples at different sampling points, which reported the best recognition rate of 98.83%. Partial least squares regression was applied to complete the second step, and R² was reaching 0.992. As for the third step, ridge regression was used to predict water quality parameters and odor concentration with the RMSE less than 0.9476. Thus, electronic noses can be applied to determine water quality parameters and odor concentrations in the effluent discharged from wastewater plants.

Livestock and poultry farm wastewater treatment and its valorization for generating value-added products: Recent updates and way forward

Livestock and poultry wastewater poses a potent risk factor for environmental pollution accelerating disease load and premature deaths. It is characterized by high chemical oxygen demand, biological oxygen demand, suspended solids, heavy metals, pathogens, and antibiotics, among other contaminants. These contaminants have a negative impact on the quality of soil, groundwater, and air, and is a potential hazard to human health. Depending on the specific characteristics of wastewater, such as the type and concentration of pollutants present; several physical, chemical and biological strategies have been developed for wastewater treatment. This review aims at providing comprehensive overview of the profiling of livestock wastewater from the dairy, swine and poultry sub-sectors along with the biological (annamox and genetically modified bacteria) and physico-chemical treatment methodologies, and valorisation for the generation of value-added products such as bioplastics, biofertilizers, biohydrogen and microalgal-microbial fuel cells. Additionally, future perspectives for efficient and sustainable wastewater treatment are contemplated.

Induction Motor Fault Diagnosis Using Support Vector Machine, Neural Networks, and Boosting Methods

Induction motors are robust and cost effective; thus, they are commonly used as power sources in various industrial applications. However, due to the characteristics of induction motors, industrial processes can stop when motor failures occur. Thus, research is required to realize the quick and accurate diagnosis of faults in induction motors. In this study, we constructed an induction motor simulator with normal, rotor failure, and bearing failure states. Using this simulator, 1240 vibration datasets comprising 1024 data samples were obtained for each state. Then, failure diagnosis was performed on the acquired data using support vector machine, multilayer neural network, convolutional neural network, gradient boosting machine, and XGBoost machine learning models. The diagnostic accuracies and calculation speeds of these models were verified via stratified K-fold cross validation. In addition, a graphical user interface was designed and implemented for the proposed fault diagnosis technique. The experimental results demonstrate that the proposed fault diagnosis technique is suitable for diagnosing faults in induction motors.

Smart instrumental Odour Monitoring Station for the efficient odour emission management and control in wastewater treatment plants

Odour emission assessment in wastewater treatment plants (WWTP) is a key aspect that needs to be improved in the plant management to avoid complaints and guarantee a sustainable environment. The research presents a smart instrumental odour monitoring station (SiOMS) composed of an advanced instrumental odour monitoring system (IOMS) integrated with other measurement units, for the continuous characterization and measurement of the odour emissions, with the aim of managing the potential odour annoyance causes in real time, in order to avoid negative effects. The application and on-site validation procedure of the trained IOMS is discussed. Experimental studies have been conducted at a large-scale WWTP. Fingerprint analysis has been applied to analyze and identify the principal gaseous compounds responsible for the odour annoyance. The artificial neural network has been adopted to elaborate and dynamically update the odour monitoring classification and quantification models (OMMs) of the IOMS. The results highlight the usefulness of a real-time measurement and control system to provide continuous and different information to the plant operators, thus allowing the identification of the odour sources and the most appropriate mitigation actions to be implemented. The paper provides important information for WWTP operators, as well as for the regulating bodies, authorities, manufacturers and end-users of odour monitoring systems involved in environmental odour impact management.

Combined machine learning and biomolecular analysis for stability assessment of anaerobic ammonium oxidation under salt stress

In this study, the stability of the total nitrogen removal efficiency (TNRE) was modeled using an artificial neural network (ANN)-based binary classification model for the anaerobic ammonium oxidation (AMX) process under saline conditions. The TNRE was stabilized to 80.2 ± 11.4% at the final phase under the salinity of 1.0 ± 0.02%. The results of terminal restriction fragment length polymorphism (T-RFLP) analysis showed the predominance of Candidatus Jettenia genus. Real-time quantitative PCR analysis revealed the average abundance of Ca. Jettenia and Kuenenia spp. increased in 3.2 ± 5.4 × 10⁸ and 2.0 ± 2.2 × 10⁵ copies/mL, respectively. The prediction accuracy using operational parameters with data augmentation was 88.2%. However, integration with T-RFLP and real-time qPCR signals improved the prediction accuracy by 97.1%. This study revealed the feasible application of machine learning and biomolecular signals to the stability prediction of the AMX process under increased salinity.

Demonstration of VOC Fenceline Sensors and Canister Grab Sampling near Chemical Facilities in Louisville, Kentucky

Experimental fenceline sensor pods (SPods) fitted with 30 s duration canister grab sampling (CGS) systems were deployed at a site near chemical facilities in Louisville, KY, from 4 June 2018 to 5 January 2020. The objective of the study was to better understand lower cost 10.6 eV photoionization detector (PID)-based volatile organic compound (VOC) sensors and investigate their utility for near-source emissions detection applications. Prototype SPods containing PID sensor elements from two different manufacturers yielded between 78% and 86% valid data over the study, producing a dataset of over 120,000 collocated pair fenceline measurements averaged into 5-min datapoints. Ten-second time-resolved SPod data from an elevated fenceline sensor signal day are presented, illustrating source emission detections from the direction of a facility 500 m west of the monitoring site. An SPod-triggered CGS acquired in the emission plume on this day contained elevated concentrations of 1,3-butadiene and cyclohexane (36 parts per billion by volume (ppbv) and 637 ppbv, respectively), compounds known to be emitted by this facility. Elevated concentrations of these compounds were observed in a subset of the 61 manual and triggered CGS grab samples acquired during the study, with winds from the west. Using novel wind-resolved visualization and normalization approaches described herein, the collocated pair SPod datasets exhibited similarity in emission source signature. With winds from the west, approximately 50% of SPod readings were above our defined theoretical detection limit indicating persistent measurable VOC signal at this site. Overall, this 19-month study demonstrated reasonable prototype SPod operational performance indicating that improved commercial forms of lower cost PID sensors could be useful for select VOC fenceline monitoring applications.