MATLAB神经网络在湖库富营养化程度评价中的应用

Application of MATLAB Neural Network in Lake and Reservoir Eutrophication Evaluation

基于MATLAB神经网络和我国湖库富营养化评价标准,运用人工神经网络模式识别理论和方法,分别建立了BP,PNN,GRNN和Elman神经网络湖泊富营养化等级评价模型,对全国24个主要湖泊富营养化程度进行评价,并与文献[5]和文献[9]的评价结果进行比较.结果表明:基于MATLAB神经网络模型评价湖库富营养化程度是可行的,且评价模型简单易行,评价精度高,为湖库富营养化程度评价提供了一种新方法.

Based on MATLAB neural network and lake eutrophication evaluation crieria,using artificial neural network pattern recoginition theory and methods,the neural network evaluation models of lake eutrophication level were established by BP,PNN,GRNN and Elman to analyse the eutrophication level of 24 major lakes in our country.The analytical results were compared with the results from literature [5] and literature [9].The results show that the lake eutrophication evaluation level model based on MATLAB is feasible,and the evaluation model is simple and high precision,and is a new method for evaluating eutrophication level of lakes.