【摘要】 针对PM2.5浓度预测问题，提出一种基于小波变换的模型。在北京市六个大气污染监测站测得的PM2.5浓度数据上，运用小波分解算法对原始数据序列进行特征提取，使用BiLSTM对高频序列进行预测，同时使用ARMA对低频序列进行预测，最后将各个子序列的预测值进行小波重构得到最终预测结果。实验结果表明，相较于传统单一模型和组合模型，该模型的性能和预测精度均有提高。更多还原

【Abstract】 A model based on wavelet transform(WT) is proposed for PM2.5 concentration prediction. Based on the PM2.5concentration data measured by six national air pollutant control stations in Beijing, the wavelet decomposition algorithm is used to extract features from the original data series. The BiLSTM neural network is used to predict the high-frequency series, while the ARMA model is used to predict the low-frequency series. Finally, the predicted values of each subsequence are reconstructed to obtain the final prediction results. The experimental results show that compared with the traditional single model and combination model, the performance and prediction accuracy of the model have been improved, which has certain application significance.更多还原