摘要
将模糊系统与神经网络结合,提出了一种水质评价模型.根据水质评价过程,采用5层结构的FNN,且使用自适应学习步长以加速网络收敛速度.该模型具有推理过程清晰,泛化能力强的特点.为了验证该算法的性能,进行了仿真试验,结果表明:和常见的方法相比,该模型的评价结果更为准确.
The eutrophication of Taihu Lake affects local environment directly . It is important to control and assess the eutrophication precisely. In this paper neural network and fuzzy system are united together and we put forward a neural model for assessment of eutrophication. There are 5 layers in this model, which is based on the process of assessment for eutrophication. The model has many characteristics, such as clear reasoning, better generalization capabilities and so on . In order to prove the performance of the algorithm we carry out a simulation experiment. The results of experiment show that the model can get more precise results compared with common methods.
出处
《云南民族大学学报（自然科学版）》
CAS
2007年第3期255-258,共4页
Journal of Yunnan Minzu University:Natural Sciences Edition
基金
重庆市教委科学研究项目(KJ050809)
关键词
模糊
神经网络
富营养化
水质评价
fuzzy
neural network
eutrophication
water quality assessment
