城市空气质量指数的时间序列分析和预测--以扬州市为例

Time Series Analysis and Forecast of Urban Air Quality Index:A Case Study of Yangzhou

城市空气质量指数是反映城市环境宜居水平的重要指标,预测城市空气质量指数可以为政策制定提供重要的参考依据。选取扬州市2014-2022年的空气质量指数(AQI)进行时间序列分析,采用的时间序列模型有自回归积分滑动平均模型(ARIMA)、先知模型(Prophet)以及人工智能领域的长短期记忆模型(LSTM)。通过这些模型分析扬州市空气质量指数的周期性、季节性和趋势特征,并对空气质量指数进行预测。结果显示,基于人工智能的长短期记忆模型具有很强的预测能力。

Urban air quality index is an important indicator that reflects the level of livability of urban environment,and the forecast of the urban air quality index can be an important reference for policy formulation.This paper selects the air quality index(AQI)of Yangzhou City from 2014 to 2022 for time series analysis,and uses the time series models including the autoregressive integral moving average model(ARIMA),the prophet model(Prophet),and the long short-term memory model(LSTM)in the field of artificial intelligence to analyze the periodicity,seasonality and trend characteristics of the air quality index in Yangzhou City,and to forecast the air quality index.The results show that the AI-based LSTM has a strong predictive power.