【摘要】 基于徐州市2013年12月—2018年11月的空气质量指数日均值,建立了时间序列自回归输入的GA-BP神经网络模型用于空气质量指数预测。结果表明,所建立的网络模型能够准确预测徐州市空气质量指数的变化趋势,其中夏季预测相对误差18. 23%,仿真均方根误差(RMSE)为14. 59;冬季预测相对误差9. 14%,仿真RMSE为11. 47。更多还原

【Abstract】 Based on the daily average of air quality index in Xuzhou from December 2013 to November 2018,this paper established a GA-BP neural network model with autoregressive input of time series for air quality index prediction. The experimental results showed that the established network model could accurately predict the change trend of Xuzhou air quality index. The relative error of summer forecast was 18. 23%,the simulation RMSE was 14. 59,the winter forecast relative error was 9. 14%,and the simulation RMSE was 11. 47.更多还原