Title: Simulation of temporal and spatial distribution characteristics of air pollutant concentration in residential areas based on random forest

Authors: Jin-Yu Fan; Bo-Hong Zheng; Bo-Yang Zhang; Bo Chen

Addresses: School of Architecture and Art, Central South University, Hunan, 410011, China; College of Architecture and Urban Planning, Guizhou University, Guizhou, 550025, China ' School of Architecture and Art, Central South University, Hunan, 410011, China ' School of Architecture and Art, Central South University, Hunan, 410011, China; School of Urban and Rural Planning and Architectural Engineering, Guiyang University, Guizhou, 550025, China ' College of Architecture and Urban Planning, Guizhou University, Guizhou, 550025, China

Abstract: In order to overcome the problems of poor pollutant concentration correlation and poor simulation effect in the traditional simulation results of air pollution concentration temporal and spatial distribution characteristics, a new simulation method of air pollution concentration temporal and spatial distribution characteristics in residential areas based on random forest is proposed. First, divide the indicators of air pollutants in residential areas, and setup the data series of the temporal and spatial distribution characteristics of air pollutant concentration. Then, build the feature judgement matrix and complete feature extraction. The simulation model of temporal and spatial distribution characteristics of pollutant concentration is constructed by using random forest, and the model is solved to realise the simulation of temporal and spatial distribution characteristics of pollutant concentration. The experimental results show that the proposed method has good correlation and high simulation accuracy.

Keywords: random forest; residential areas; air pollution concentration; temporal and spatial distribution; feature simulation.

DOI: 10.1504/IJETM.2023.127345

International Journal of Environmental Technology and Management, 2023 Vol.26 No.1/2, pp.105 - 118

Received: 25 Oct 2021
Accepted: 14 Feb 2022
Published online: 30 Nov 2022 *