Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Annals of Computer Science and Information Systems

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Development of software tool for pptimization and evaluation of cycling routes by characterizing cyclist exposure to air pollution

Zhivkov Petar, Simidchiev Alexander

Annals of Computer Science and Information Systems

2022

Vol. 32

105--112

In modern cities, poor air quality has contributed to replacing motorized cars with active modes of transportation such as cycling. However, when designing and building bike infrastructure, officials neglect to consider air quality concerns connected to cyclists, and most cycling lanes are developed next to heavy-traffic roadways. This poses additional health risks to cyclists, due to their increased ventilation rate. To preserve a sustainable quality of life for a city's residents, it's critical to understand how to detect and quantify PM exposure, especially in potentially hazardous locations. This study offers a software tool based on experimental data to optimize and evaluate cycling routes by calculating the overall amount of particulate matter intake in terms of the physiological response of cyclists.

Optimization and Evaluation of Calibration for Low-cost Air Quality Sensors: Supervised and Unsupervised Machine Learning Models

Zhivkov Petar

Annals of Computer Science and Information Systems

2021

Vol. 25

255--258

With the advancement of air pollution management, low-cost sensors are increasingly being used in air quality monitoring, but the data quality of these sensors is still a major source of concern. In this paper, data from five air monitoring stations in Sofia were compared to data from fixed low-cost PM sensors. The values of atmospheric pressure from low-cost sensors and the effects of relative humidity were investigated. A two-step model was created to refine the calibration process for low-cost PM sensors. At first, we calibrated the sensors with five separate supervised machine learning models and then the ANNf inal model with anomaly detection completed the results. The ANN-final model improved the R2 values of the PM10 determined by low-cost sensors from 0.62 to 0.95 as compared to standard instruments. In conclusion, the two-step calibration model proved to be a positive solution to addressing low-cost sensor efficiency issues.