DEVELOPMENT OF A LOW-COST SENSOR NETWORK FOR COMMUNITY-MADE MEASUREMENTS OF AIR POLLUTION

摘要

The levels of pollution present in the air have been dramatically increasing over the years due to the continuous emission of greenhouse gases such as CO_2, CO, NO_x and H_2S, among others. The main source of these emissions is from burning fossil fuels for electricity, heat, and transportation. This represents a tremendous risk to the populations located near the emission sources where people get exposed to dangerous concentrations of such gases on a daily basis. The lack of open real-time monitoring tools makes people unaware of the damage these pollutants cause to their health. This research proposes the development and implementation of a low-cost independent solution to keep the members of a community informed about concentration levels of air pollution due to local emissions. This tool must be easily accessible to the users so that the data about the number of particulates per million of a specific gas within a zone of interest can be viewed at any time. The proposed solution consists of a sensor network. covering the widest possible area, with respect to the point of interest. The collected data is sent to a cloud server, which operates as storage center and in which the data can be latter accessed for subsequent analysis. The measurements are sent to the server by means of a wireless communication protocol, carried out by a General Packet Radio Service, GPRS, communication module connected to each station. In this way. the coverage of the network is not limited by issues such as the use of local area networks which at the same time facilitates the transportation and installation of the stations at any desired measurement site. Since each station can collect large amounts of data during a given period of time, it was necessary to implement techniques such as Big Data in order to extract important information and to identify patterns from the data such as the areas having the highest concentration of gases and possible correlations with other variables such as local weather conditions. This information could be used to support the making of decisions that benefit the communities impacted by air pollution, for example the early triggering of bad air quality alarms or the development of policies to regulate industry operation that can potentially impact the health of neighboring communities. A pilot case study was implemented in the city of Floridablanca, Colombia, to demonstrate the monitoring of the emissions of hydrogen sulfide of a big wastewater processing plant.