A novel dynamic ensemble air quality index forecasting system

Özet

The air quality index (AQI) can reflect the change of air quality in real time. It has linear characteristics, nonlinearand fuzzy features. However, a single model cannot fit the dynamic changes of AQI scientifically andreasonably. Therefore, this paper proposes a new dynamic ensemble forecasting system based on multi-objectiveintelligent optimization algorithm to forecast AQI, which has time-varying parameter weights and mainlycontains three module: data preprocessing module, dynamic integration forecasting module and system evaluationmodule. In the data preprocessing module, the off-line frequency domain filtering approach is applied toidentify and correct the outliers in the series. To better extract the series information and remove the randomnoise, the time series is decomposed into multi-level utilizing decomposition strategy and reconstructed. In thedynamic integration forecasting module, three hybrid models based on ARIMA, optimized extreme learningmachine and fuzzy time series model, named as HCA, HCME and HCFL respectively, are used to forecast thereconstructed series and time varying parameters are employed to dynamically combine the forecasting results.In the system evaluation module, the accuracy of the system was tested by parameter test method and nonparametrictest method respectively. The results demonstrate that the proposed dynamic integrated model is notonly superior to other comparison models in forecasting accuracy, but also provides strong technical support forair quality forecasting and treatment.