大气污染物浓度的神经网络预报

首次将神经网络应用于实测大气污染物浓度的预报.结果表明,预报值与观测值符合得较好.由于人工神经网络具有逼近精度高、学习速度快、对资料长度要求不高等优点,因而在非线性时间序列的预报中显示出了独特的优越性,并将会在大气环境和气候预报中得到越来越多的应用.

Prediction of atmospheric pollutant concentration using artificial neural network

In this paper, artificial neural network is first used to predict the observational data of atmospheric pollutant concentration. The results show that predicted and observed values are in relatively good agreements. Owing to such virtues as high approaching precision, fast learning rate, no stringent requirement on length of data and so on, artificial neural network is advantageous to predict nonlinear time series and will be applied to forecast the atmospheric environment and the climatic change more extensively.