Prediction of Status Particulate Matter 2.5 Using State Markov Chain Stochastic Process and HYBRID VAR-NN-PSO

Rezzy Eko Caraka,Rung Ching Chen,Shih Hung Wu,Toni Toharudin,Bens Pardamean,Hasbi Yasin

doi:10.1109/access.2019.2950439

Abstract

Air pollution is the entry or inclusion of living things, energy substances, and other components into the air. Moreover, Air pollution is the presence of one or several contaminants in the outside atmospheric air such as dust, foam, gas, fog, smoke or steam in large quantities with various properties and time intervals of the contaminants in the air resulting in disturbances to the lives of humans, plants or animals. One of the parameters measured in determining air quality is PM 2.5 . However, PM 2.5 has a higher probability of being able to enter the lower respiratory tract because small particle diameters can potentially pass through the lower respiratory tract. In this paper, we will get two different insight. First, the probability of status change using Markov chain and second, forecasting by using VAR-NN-PSO. More details we classify by three classifications no risk (1-30), medium risk (30-48), and moderate (>49) in Pingtung and Chaozhou. This data is starting from January 2014 to May 2019 and it can be modeled with the Markov chain. At the same time, we perform Hybrid VAR-NN-PSO to forecast PM 2.5 in Pingtung and Chaozhou. In this optimization, the search for best solutions is carried out by a population consisting of several particles. Based on the results of the discussion, opportunities for the transition from monthly status change are obtained continuous stochastic time with a stationary probability distribution. Regarding the VAR-NN-PSO, we obtained the mean absolute percentage error (MAPE) 3.57% for PM 2.5 data in Pingtung and 4.87% for PM 2.5 data in Chaozhou, respectively. This model can be predicted to forecasting 180 days ahead. Besides, the population in PSO has generated randomly with the smallest value and the high value the accuracy.

Highlights

Air pollution is a change in the composition of air substances so that the air quality of these substances becomes reduced
The three-state Markov chain model can be used to determine the change of PM2.5 status
A combination will be made between lags to determine the existence of nonlinear patterns in the PM2.5 data

Summary

INTRODUCTION

Air pollution is a change in the composition of air substances so that the air quality of these substances becomes reduced. Particulates emitted into the atmosphere will undergo a process of changing their shape, size, and chemical composition by several mechanisms that will continue to occur until the particles undergo a deposition process. Not all phenomena are entirely deterministic because unknown factors can occur and affect these physical phenomena [11] In this case, the time-dependent phenomenon is needed in stochastic models [12]. Neural Network [13] is a nonparametric model that can be used for modelling time series data that does not require various residual assumptions. The application of Neural Network in time series prediction models [18] is expected to provide more accurate and robust results against data fluctuations [19]. The problem that appears in neural network modelling for multivariate time series data is how to determine the lag. A breakthrough is needed to find alternative methods that can overcome this problem

MARKOV CHAIN

VECTOR AUTOREGRESSIVE

CONCLUSION