LSTM-Based Citizens’ QoL Provisioning in Smart Cities Using IoT-Enabled Environmental Sensors and Graphical Tools

Authors:
Publisher website
Google Scholar
Abstract
This study examines the role of Cyber-Physical Systems (CPS) through Internet of Things (IoT) devices, and advanced machine learning in monitoring and managing air pollution in future smart cities, with a focus on enhancing Quality of Life (QoL). CPS, integrated with IoT sensors, enables real-time data collection and analysis to understand the impact of power plant locations and operations on air quality. The study utilizes the Topographic Position Index (TPI) to assess how topography affects pollution, and regression analysis to explore the relationship between power plant output and pollution levels. Long Short-Term Memory (LSTM) networks forecast air pollution trends, while the CA-Markov chain method predicts spatial-temporal patterns. Graphical tools, such as heat maps and trend charts, visualize pollutant distribution and progression, making data more accessible for decision-makers. Findings indicate that flat areas are more susceptible to higher pollution, with February being the peak month for most pollutants. The results of this study also showed that CO and CO2 in spring have a higher correlation with power plant production (R2 =0.85) than other pollutants. The LSTM model shows high accuracy (R2 = 0.84) in predicting air quality trends within the CPS and IoT-enabled framework.
Funding Information
  • Key Research Project of Natural Science in Colleges and Universities of Anhui Provincial Department of Education (KJ2021A1163)
  • Anhui Xinhua University - Intelligent Automation Research Institute (yjs202104)
  • Major Teaching and Research Project of Quality Engineering Project of Education Department of Anhui Province (2022jyxm649)
  • Anhui Provincial Department of Education Quality Engineering Project School-Enterprise Cooperation Practice Education Base (2022xqhz028)

This publication has 35 references indexed in Scilit: