Determination of Odor Intensity of Binary Gas Mixtures Using Perceptual Models and an Electronic Nose Combined with Fuzzy Logic

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.

Utilisation of QSPR ODT modelling and odour vector modelling to predict Cannabis sativa odour

Cannabis flower odour is an important aspect of product quality as it impacts the sensory experience when administered, which can affect therapeutic outcomes in paediatric patient populations who may reject unpalatable products. However, the cannabis industry has a reputation for having products with inconsistent odour descriptions and misattributed strain names due to the costly and laborious nature of sensory testing. Herein, we evaluate the potential of using odour vector modelling for predicting the odour intensity of cannabis products. Odour vector modelling is proposed as a process for transforming routinely produced volatile profiles into odour intensity (OI) profiles which are hypothesised to be more informative to the overall product odour (sensory descriptor; SD). However, the calculation of OI requires compound odour detection thresholds (ODT), which are not available for many of the compounds present in natural volatile profiles. Accordingly, to apply the odour vector modelling process to cannabis, a QSPR statistical model was first produced to predict ODT from physicochemical properties. The model presented herein was produced by polynomial regression with 10-fold cross-validation from 1,274 median ODT values to produce a model with R2 = 0.6892 and a 10-fold R2 = 0.6484. This model was then applied to terpenes which lacked experimentally determined ODT values to facilitate vector modelling of cannabis OI profiles. Logistic regression and k-means unsupervised cluster analysis was applied to both the raw terpene data and the transformed OI profiles to predict the SD of 265 cannabis samples and the accuracy of the predictions across the two datasets was compared. Out of the 13 SD categories modelled, OI profiles performed equally well or better than the volatile profiles for 11 of the SD, and across all SD the OI data was on average 21.9% more accurate (p = 0.031). The work herein is the first example of the application of odour vector modelling to complex volatile profiles of natural products and demonstrates the utility of OI profiles for the prediction of cannabis odour. These findings advance both the understanding of the odour modelling process which has previously only been applied to simple mixtures, and the cannabis industry which can utilise this process for more accurate prediction of cannabis odour and thereby reduce unpleasant patient experiences.

Volatile Organic Compound Monitoring during Extreme Wildfires: Assessing the Potential of Sensors Based on LbL and Sputtering Films

A new theory suggests that flammable gases generated by heated vegetation, in particular the volatile organic compounds (VOC) common to Mediterranean plants, may, under certain topographic and wind conditions, accumulate in locations where, after the arrival of the ignition source, they rapidly burst into flames as explosions. Hence, there is a need for the development of a system that can monitor the development of these compounds. In this work, a sensor's array is proposed as a method for monitoring the amount of eucalyptol and α-pinene, the major VOC compounds of the Eucalyptus and Pine trees. The detection of the target compounds was assessed using the impedance spectroscopy response of thin films. Combinations of layers of polyelectrolytes, such as poly(allylamine hydrochloride) (PAH), polyethyleneimine (PEI), poly(sodium 4-sytrenesulfonate) (PSS) graphene oxide (GO), and non/functionalized multiwall nanotubes (MWCNT-COOH or MWCNT), namely, PAH/GO, PEI/PSS, PEI/GO, PAH/MWCNT, PAH/MWCNT-COOH, films, and TiO2 and ZnO sputtered films, were deposited onto ceramic supports coated with gold interdigitated electrodes. The results showed that concentrations of the target VOCs, within the range of 68 to 999 ppmv, can be easily distinguished by analyzing the impedance spectra, particularly in the case of the ZnO- and PAH/GO-film-based sensors, which showed the best results in the detection of the target compounds. Through principal component analysis (PCA), the best set of features attained for the ZnO and PAH/GO based sensor devices revealed a linear trend of the PCA's first principal component with the concentration within the range 109 and 807 ppmv. Thus, the values of sensitivity to eucalyptol and α-pinene concentrations, which were (2.2 ± 0.3) × 10-4 and (5.0 ± 0.7) × 10-5 per decade, respectively, as well as resolutions of 118 and 136 ppbv, respectively, were identified.

Artificial neural network (ANN) modeling for the prediction of odor emission rates from landfill working surface

Landfills release significant odorous compounds from the working surface, and their emission rates are crucial for odor and health risk assessment. A total of 99 valid datasets of odor emissions from a landfill working surface were obtained from in situ monitoring for 9 months. Meteorological parameters (temperature, humidity, atmospheric pressure) and waste properties (contents of protein, lipid, carbohydrate, ash, and moisture) were used to construct artificial neural network (ANN) models for the emission rate prediction of typical compounds. The optimal structures and performance of the ANN models were determined by comparing and training with different structural configurations. The ANN models with genetic algorithm (GA) optimization show better performance than those without GA. With the data distribution of input parameters, the ranges of the emission rates of typical compounds were predicted by combining the established ANN models and the Monte Carlo approach. The sensitivity and uncertainty analyses revealed that temperature, atmospheric pressure, protein and lipid contents are parameters sensitive to emission rates, and meteorological parameters have significant impacts on the uncertainty. The established ANN models for the prediction of emission rates can provide scientific evidence and an approach to assess and control the odor and health risk in waste sectors.

Visual Analysis of Odor Interaction Based on Support Vector Regression Method

The complex odor interaction between odorants makes it difficult to predict the odor intensity of their mixtures. The analysis method is currently one of the factors limiting our understanding of the odor interaction laws. We used a support vector regression algorithm to establish odor intensity prediction models for binary esters, aldehydes, and aromatic hydrocarbon mixtures, respectively. The prediction accuracy to both training samples and test samples demonstrated the high prediction capacity of the support vector regression model. Then the optimized model was used to generate extra odor data by predicting the odor intensity of more simulated samples with various mixing ratios and concentration levels. Based on these olfactory measured and model predicted data, the odor interaction was analyzed in the form of contour maps. This intuitive method showed more details about the odor interaction pattern in the binary mixture. We found that that the antagonism effect was commonly observed in these binary mixtures and the interaction degree was more intense when the components’ mixing ratio was close. Meanwhile, the odor intensity level of the odor mixture barely influenced the interaction degree. The machine learning algorithms were considered promising tools in odor researches.