基于机器学习的短期PM2.5预测

Short-term PM2.5 Predictions Based on Machine Learning

由于环境和快速发展之间的不平衡,城市空气质量问题变得越来越突出。PM2.5作为空气污染的主要成分,会对人体造成很大伤害。因此,准确地预测PM2.5浓度对于保护人们健康具有重要意义。首先选取了其他空气质量数据(PM_(10)、NO_(2)、CO_(2)、O_(3))作为影响因素,构建了基于机器学习(多元线性回归、岭回归、套索回归、决策树、随机森林和人工神经网络)的PM2.5预测模型;其次利用这些模型预测山西省太原市未来1小时PM2.5浓度;最后通过MAE、RMSE、R^(2)来等指标评价各模型的预测性能,实验结果表明,基于随机森林的预测模型具有最高的预测精度。

Due to the imbalance between the environment and the rapid development,the urban air quality problem has become more and more prominent.PM2.5,as the main component of air pollution,will cause great harm to the human body.Therefore,accurately predicting PM2.5 concentration is important to protect people's health.In this paper,other air quality data(PM_(10),NO_(2),CO_(2),O_(3))are selected as influencing factors,and PM2.5 prediction model based on machine learning(multiple linear regression,ridge regression,lasso regression,decision tree,random forest and artificial neural network)is constructed;secondly,these models are used to predict the PM2.5 concentration in the next hour in Taiyuan;finally,the prediction performance of each model is evaluated by MAE,RMSE and R^(2).The experimental results show that the prediction model based on random forest has the highest prediction accuracy.