Urban Monitoring of Unpleasant Odors with a Handheld Electronic Nose

Abstract

This work describes a real application of artificial olfaction where a handheld electronic nose was used as a validation tool for a chemical spillage in a southern town in Spain. The objective was to check if the palliative and precautionary measurements taken by the authorities were working effectively, removing the elevated values of phenol that were detected in a wide area of the municipality of Coria del Río (Spain). To this end, a gas distribution map of the affected neighborhoods was built with a portable electronic nose taking into consideration the likely presence of other volatile chemicals in the area. For the latter, we trained a volatile chemical classifier with a dataset of typical urban smells that we wanted to remove from the results (e.g. traffic emissions, garbage, fresh-air), as well as with a specific air-born phenol dataset. Results demonstrated that the palliative measures were in general satisfactory, but some hot-spots were located where the intensity of phenol-like smell was still higher than desired. Advice was given to the local authorities to doublecheck these locations with analytical gas-monitoring equipment.

Methodologies for smellwalks and scentwalks—a critical review

“…A few recent studies undertook novel smellwalks, engaging electronic smellwalks, hybrid human/electro smellwalks, virtual smellwalks, and retrospective smellwalks. One study in Malaga (Spain) tested the feasibility of replacing human smell detection with an e-nose—a wearable electronic nose, for the central reasons that smellmaps created through traditional means are easily outdated, and the effort to conventionally produce them is arduous ( Gongora et al 2018 ). While comparable methodologies were engaged (predetermined locations, different conditions, attribute data of smell type, intensity, and geolocation), each 1 h walk covered significant distances of around 3.5 km ( Gongora et al 2018 ).…”

“…One study in Malaga (Spain) tested the feasibility of replacing human smell detection with an e-nose—a wearable electronic nose, for the central reasons that smellmaps created through traditional means are easily outdated, and the effort to conventionally produce them is arduous ( Gongora et al 2018 ). While comparable methodologies were engaged (predetermined locations, different conditions, attribute data of smell type, intensity, and geolocation), each 1 h walk covered significant distances of around 3.5 km ( Gongora et al 2018 ). While recognized by the researchers that the e-nose smell detection rate was lower than that of a human (including false positive readings), it was suggested that future e-nose research could be improved with an increased data-size and fine-tuning of classifiers ( Gongora et al 2018 ).…”

Multi-Classifier Tree With Transient Features for Drift Compensation in Electronic Nose

“…RTIFICIAL Olfaction (AO) refers to identifying and discriminating different odors using electronic instruments like an Electronic Nose System (ENS). The ENS has a broad range of applications including the food industry [1]- [3], health sector [4], [5], environment monitoring [6], oil and gas industry [7], and security purposes [8], [9], among others. An ENS is comprised of two units, a sensing unit and a recognition unit.…”

OdorSense: measuring, assessment and alerting the health effects of odor pollution

“…In this research study, an embedded electronic device is designed to measure the concentration of Odor pollution at selected areas. The design of artificial olfactory electronic nose (Gongora et al, 2019;Erisman, 2001) (EPA, 2021) and then providing the alert information to end users. Finally, the Data processor (Atmega328) coordinates and controls all the peripherals connected to the system, processes the sensor data and sends the sensed information to the Cloud.…”