Predicting urban PM2.5 concentration based on principal component analysis and BP neural network

Yiping Ying

doi:10.1117/12.2627895

22 April 2022 Predicting urban PM2.5 concentration based on principal component analysis and BP neural network

Yiping Ying

Proceedings Volume 12163, International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2021); 121631J (2022) https://doi.org/10.1117/12.2627895
Event: International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2021), 2021, Nanjing, China

Abstract

Ambient air pollutants have a direct influence on the human body and the environment, and accurate prediction of PM2.5 concentration is very crucial for air quality control. This paper aims to provide a reasonable model to predict and analyze PM2.5 concentration data of Hangzhou. Forty-five sample data from ten meteorological observation points in Hangzhou in June and July 2021 were selected; factors including PM2.5, PM10 and daily temperatures as input. The BP neural network model is often used to predict the concentration of PM2.5. However, due to large dimension of the input elements, it often reduces the efficiency. To handle the problem, principal component analysis (PCA) is utilized in the input layer to achieve the goal of dimensionality reduction. By training the training-set, the network model structure is determined, and prediction accuracy is tested. As a conclusion, PCA-BP neural network performs equally well compared with BP neural network, while it can significantly save the calculation time and simplify the network structure.

Citation Download Citation

Yiping Ying "Predicting urban PM2.5 concentration based on principal component analysis and BP neural network", Proc. SPIE 12163, International Conference on Statistics, Applied Mathematics, and Computing Science (CSAMCS 2021), 121631J (22 April 2022); https://doi.org/10.1117/12.2627895

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available