In order to maintain the air quality, continuous monitoring and analysis of the air pollution data is necessary; especially in areas where industrial and vehicular emissions contribute more to poor air quality. Inhalation of high concentration of fine particulate matter (PM2.5) causes lung, heart and various other diseases, which increase hospital visits and mortalities each day. The focus of this paper is to analyse the historical pollution data and corresponding meteorological data from selected areas, and to forecast PM2.5 over the next 48 hours by using multiple neural networks. In this proposed model, experiment is conducted by including pollutant and meteorology data recorded for every hour from 14 places, which includes northern, southern, western and eastern parts of India. Spatial temporal relations and terrain impact are then extracted. The proposed system applies multiple neural networks including convolutional neural network, artificial neural network, long short-term memory and adaptive neuro-fuzzy inference system to predict the air quality. The proposed model - ANFIS prediction performance - is better than the existing ANN model.

Convolutional neural network; LSTM; adaptive neuro-fuzzy inference system; air quality forecast; dynamic time warping; Euclidean distance

[1] Hao Guo, Sri Harsha Kota, Shovan Kumar Sahu, Jianlin Hu, Qi Ying, Aifang Gao, Hongliang Zhang, “Source apportionment of PM2.5 in North India using source-oriented air quality models”, Environmental Pollution 231 (2017) pp.426-436.
[2] The impact of air pollution on deaths, disease burden, and life expectancy across the states of India: The Global Burden of Disease Study 2017, India State-Level Disease Burden Initiative Air Pollution Collaborators, Lancet Planet Health 2019; 3: e26–39.
[3] Ping-Wei Soh, Jia-Wei Chang, and Jen-Wei Huang, “Adaptive Deep Learning-Based Air Quality Prediction Model Using the Most Relevant Spatial-Temporal Relations”,. Ieee Access,Vol-6, 2018, pp. 38186-38199
[4] Chiou-Jye Huang and Ping-Huan Kuo, “A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities”, Sensors 2018, 18, 2220;
[5] Athira V, Geetha P, Vinayakumar R, Soman K P, “Deep AirNet: Applying Recurrent networks for Air Quality Prediction”, Procedia Computer Science 132(2018) pp-1394-1403.
[6] Guyu Zhao, Guoyan Huang, Hongdou He, And Qian Wang, “Innovative Spatial-Temporal Network Modeling and Analysis Method of Air Quality”, Ieee Access, vol.7,2019, pp-26241-26254.
[7] Nazif1 • N. I. Mohammed1 • A. Malakahmad1 • M. S. Abualqumboz1, “Multivariate analysis of monsoon seasonal variation and prediction of particulate matter episode using regression and hybrid models”, International Journal of Environmental Science and Technology, June 2019, vol.16(6), pp-2587-2600.
[8] Li-Yu Hu, Min-Wei Huang, Shih-Wen Ke, and Chih-Fong Tsai, “ The distance function effect on k-nearest neighbor classification for medical datasets”, Springerplus. 2016; 5(1): 1304.
[9] Duarte Folgado, Marília Barandas, Ricardo Matias, Rodrigo Martins, Miguel Carvalho, Hugo Gamboa, “Time Alignment Measurement for Time Series”, Pattern Recognition, Volume 81, September 2018, pp- 268-279.
[10] Sakshi Indolia, Anil Kumar Goswami, S. P. Mishra, Pooja Asopa, “Conceptual Understanding of Convolutional Neural Network- A Deep Learning Approach”, International Conference on Computational Intelligence and Data Science (ICCIDS 2018), Procedia Computer Science 132 (2018) pp-679–688
[11] Jianfeng Zhang, Yan Zhu, Xiaoping Zhang, Ming Ye, Jinzhong Yang, “Developing a Long Short-Term Memory (LSTM) based Model for Predicting Water Table Depth in Agricultural Areas”, Journal of Hydrology, Vol. 561, June 2018, pp- 918-929.
[12] Ryan G.Hefron, Brett J.Borghetti, James C.Christensen, Christine M. Schubert Kabban, “Deep long short-term memory structures model temporal dependencies improving cognitive workload estimation”, Pattern Recognition Letters 94 (2017) pp-96–104.