A Hybrid Spatial-Temporal Approach to Pollution Forecasting with Dynamic Updates and Time Series Analysis
A Hybrid Spatial-Temporal Approach to Pollution Forecasting with Dynamic Updates and Time Series Analysis |
||
 |
 |
|
| © 2023 by IJETT Journal | ||
| Volume-71 Issue-10 |
||
| Year of Publication : 2023 | ||
| Author : Snehlata Beriwal, John A, Kavita, Avneesh Kumar |
||
| DOI : 10.14445/22315381/IJETT-V71I10P212 | ||
How to Cite?
Snehlata Beriwal, John A, Kavita, Avneesh Kumar, "A Hybrid Spatial-Temporal Approach to Pollution Forecasting with Dynamic Updates and Time Series Analysis," International Journal of Engineering Trends and Technology, vol. 71, no. 10, pp. 133-145, 2023. Crossref, https://doi.org/10.14445/22315381/IJETT-V71I10P212
Abstract
The rapidly deteriorating air quality across the globe has increasingly become a challenge with far-reaching consequences. Hence, accurate air quality prediction, monitoring, and forecasting have become an intrinsic part of managing our living environment. Such advanced predictions and timely interventions thereof can aid in minimizing any untoward threats to our health and quality of life. The primary aim of this research is to enable effective time and location-based predictions and forecasting of air quality and pollution levels. To that end, a hybrid approach based on indexing and time series techniques has been proposed in this study. This hybrid approach is based on the D-Tree-based indexing method, SARIMA, Bidirectional LSTM, and the Pearson correlation. The D-Tree-based indexing method is used to manage current and previous data. The SARIMA is used to predict and forecast the future status of pollution particles based on current data as well as seasonal trends. The Bidirectional LSTM is utilized for Time Series Forecasting using current and past data managed by the D-tree indexing method. The Pearson correlation is used for measuring and managing the mean of two predicted outputs from inputs received concurrently from live environments. During implementation, live pollution data was concurrently collected from the different location-centric pollution sensing devices and updated using the indexing method. This current data was then appended to the previous year's data to improve accuracy further. Thus, using both past and live data, forecasts were made for the next 6, 12 hours, and 1, 2, and 3 days, respectively. Prediction accuracy was evaluated using various metrics such as accuracy, Air Quality Index (I), Mean Square Root (MSR), Mean Absolute Error (MAE), and correlation coefficient (R). The predicted results were found to produce higher accuracy (97.6%) across different time lags compared to other predominant forecasting methods. This approach, therefore, has been found to concurrently update the status of pollutant particles in dynamic environments effectively and consistently.
Keywords
Spatial and temporal data, Hybrid model, Pollution forecasting, SARIMA, LSTM, D-Tree.
References
[1] Lu Bai et al., “Air Pollution Forecasts: An Overview,” International Journal of Environmental Research and Public Health, vol. 15, no. 4, p. 780, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[2] C.H. Bosanquet, and J.L. Pearson, “The Spread of Smoke and Gases from Chimneys,” Transactions of the Faraday Society, vol. 32, pp. 1249-1263, 1936.
[CrossRef] [Google Scholar] [Publisher Link]
[3] European Union Joint Research Centre (JRC), “Features of Dispersion Models,” 2004.
[4] Nolan Atkins, Air Pollution Dispersion: Ventilation Factor, NVU-Lyndon Atmospheric Sciences, 2008.
[Online].Available:https://apollo.nvu.vsc.edu/classes/met130/notes/chapter18/dispersion_intro.html#:~:text=The%20ventilation%20factor%20gives%20us,pollutants%20are%20allowed%20to%20mix.
[5] Donald Ermak, “User's Manual for Slab: An Atmospheric Dispersion Model for Denser-than-Air Releases,” Technical Report, OSTI.GOV, UCRL-MA-105607, 1990.
[Google Scholar] [Publisher Link]
[6] Pritthijit Nath et al., “Spatio-Temporal Pollution Forecasting using Hybrid Networks,” Research Square, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Yanlin Qi et al., “A Hybrid Model for Spatiotemporal Forecasting of PM2.5 based on Graph Convolutional Neural Network and Long Short-Term Memory,” Science of the Total Environment, vol. 664, pp. 1-10, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Dewen Seng et al., “Spatiotemporal Prediction of Air Quality Based on LSTM Neural Network,” Alexandria Engineering Journal, vol. 60, no. 2, pp. 2021-2032, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Pritthijit Nath et al., “Long-Term Time-Series Pollution Forecast using Statistical and Deep Learning Methods,” Neural Computing and Applications, vol. 33, pp. 12551-12570, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Fang Zhao et al., “Research on PM2.5 Spatiotemporal Forecasting Model Based on LSTM Neural Network,” Computational Intelligence and Neuroscience, vol. 2021, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Shurui Fan et al., “A Hybrid Model for Air Quality Prediction Based on Data Decomposition,” Information, vol. 12, no. 5, p. 210, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Wenjing Mao et al., “A Hybrid Integrated Deep Learning Model for Predicting Various Air Pollutants,” GIScience & Remote Sensing, vol. 58, no. 8, pp. 1395-1412, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Shengdong Du et al., “Deep Air Quality Forecasting using Hybrid Deep Learning Framework,” IEEE Transactions on Knowledge and Data Engineering, vol. 33, no. 6, pp. 2412-2424, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Yang Han et al., “A Domain-Specific Bayesian Deep-learning Approach for Air Pollution Forecast,” IEEE Transactions on Big Data, vol. 8, no. 4, pp. 1034-1046, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Van-Duc Le, “Spatiotemporal Graph Convolutional Recurrent Neural Network Model for Citywide Air Pollution Forecasting,” arXiv, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Atakan Kurt et al., “An Online Air Pollution Forecasting System using Neural Networks,” Environment International, vol. 34, no. 5, pp. 592-598, 2008.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Vlado Spiridonov et al., “Development of Air Quality Forecasting System in Macedonia, based on WRF-Chem Model,” Air Quality, Atmosphere & Health, vol. 12, pp. 825-836, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Sharnil Pandya et al., “Pollution Weather Prediction System: Smart Outdoor Pollution Monitoring and Prediction for Healthy Breathing and Living,” Sensors, vol. 20, no. 18, pp. 1-25, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Jian Wei Koo et al., “Prediction of Air Pollution Index in Kuala Lumpur using Fuzzy Time Series and Statistical Models,” Air Quality, Atmosphere & Health, vol. 13, pp. 77-88, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Shuixia Chen, Jian-qiang Wang, and Hong-yu Zhang, “A Hybrid PSO-SVM Model Based on Clustering Algorithm for Short-Term Atmospheric Pollutant Concentration Forecasting,” Technological Forecasting and Social Change, vol. 146, pp. 41-54, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Hufang Yang et al., “A Novel Combined Forecasting System for Air Pollutants Concentration based on Fuzzy Theory and Optimization of Aggregation Weight,” Applied Soft Computing, vol. 87, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Angel Cujia et al., “Forecast of PM10 Time-Series Data: A Study Case in Caribbean Cities,” Atmospheric Pollution Research, vol. 10, no. 6, pp. 2053-2062, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Snezhana Georgieva Gocheva-Ilieva et al., “Regression Trees Modeling of Time Series for Air Pollution Analysis and Forecasting,” Neural Computing and Applications, vol. 31, pp. 9023-9039, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[24] Zena A. Aziz Aziz, and Siddeeq Y. Ameen Ameen, “Air Pollution Monitoring using Wireless Sensor Networks,” Journal of Information Technology and Informatics, vol. 1, no. 1, pp. 20-25, 2021.
[Google Scholar] [Publisher Link]
[25] Yang Yurong et al., “A Study on Water Quality Prediction by a Hybrid CNN-LSTM Model with Attention Mechanism,” Environmental Science and Pollution Research, vol. 28, pp. 55129-55139, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[26] Gaurav Anand, Sharda Kumari, and Ravi Pulle, “Fractional-Iterative BiLSTM Classifier: A Novel Approach to Predicting Student Attrition in Digital Academia,” SSRG International Journal of Computer Science and Engineering, vol. 10, no. 5, pp. 1-9, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[27] Yue-Shan Chang et al., “An LSTM-based Aggregated Model for Air Pollution Forecasting,” Atmospheric Pollution Research, vol. 11, no. 8, pp. 1451-1463, 2020.
[CrossRef] [Google Scholar] [Publisher Link]