基于径向基函数网络的宣城市空气质量预测

提出了一种应用人工神经网络进行空气质量预测的方法,即采用径向基函数神经网络进行短期的空气质量预测;并采用了主成分分析方法降低神经网络学习矩阵维数,浓缩预测信息,降维去噪.选取宣城市气象局2003年到2005年地面气象观测资料作为预测因子,宣城市环境保护监测中心提供的PM10、SO2浓度值作为预测对象,进行训练学习和预测验证.研究结果表明:将该方法应用于空气质量预测,效果良好,具有较强的实用性和推广能力.

Based on RBF Neural Network Prediction of Air Quality in the Xuancheng

A method of prediction of air quality is proposed based on RBF artificial neural network(ANN).making a low-dimension ANN learing matrix through principal component analysis,it could distill the main information of many factorials and remarkably decrease the dimension.According to the characteristics of the prediction of air quality,this paper selects the data in Xuancheng from 2003 to 2005 that greatly influence the air quality.The experimental result shows that the performance of prediction of air quality is favorable,the learning speed is fast and the rate of accurate is high,so it have a practical value.It provides a precise and generalization and efficient way for the prediction of air quality.