前馈神经网络在环境评价和预测中的应用进展

Application Progress of Feedforward Neural Networks for Environment Impact Assessment and Prediction

本文针对水环境中复杂的不确定性及非线性关系,在水环境不确定性分析的基础上,详细阐述了以BP网络和RBF网络为代表的前馈神经网络法的基本原理,分析了两种方法的优点。同时,本文对两种方法在水环境影响评价工作中的应用现状进行总结,分析了两种方法的研究发展趋势。

In this paper,based on the analysis of uncertainty of water environment,the essential principle of BP and RBF networks were introduced in detail as representatives of the feedforward neural networks.The advantages of the two methods were analyzed.Meanwhile the status quo of application of the two methods for water environment impact assessment was summarized.The developing trend of the two methods was analyzed as well.