基于随机训练样本的河流系统健康状况评价  

Evaluation of the Health Status of River Systems based on Random Training Samples

作　　者：张明[1,2] 周润娟[1] 和蕊[1] 

机构地区：[1]安徽工程大学建筑工程学院,安徽芜湖241000 [2]南京水利科学研究院水文水资源与水利工程科学国家重点实验室,南京210029

出　　处：《南水北调与水利科技》2012年第2期75-78,共4页South-to-North Water Transfers and Water Science & Technology

基　　金：安徽高校省级自然科学研究项目(KJ2012B016);水文水资源与水利工程科学国家重点实验室开放研究基金(2011492011)

摘　　要：河流系统健康状况的评价对于河流的可持续管理、区域生态环境建设具有重要的意义。在综述河流系统健康状况传统评价方法的基础上,建立了基于随机训练样本的BP神经网络评价模型。该模型采用随机方式分别在各标准评价等级上、下限之间随机产生若干评价样本,再经过建模、训练步骤得到用于河流系统健康状况评价的模型,一定程度上减少了由于评价样本较少带来的评价结果不确定性影响。海河三岔口河流实例研究结果显示:基于随机样本的河流系统健康状况评价模型中,实数码与二进制码输出形式都给出了2.95级良好状态的评价结果,二进制码形式还得到了该河流有向不健康状态演变的趋势。与模糊综合评价模型给出的结果相比,本文建立的模型评价结果更符合实际河流健康状况,为研究河流系统健康状况提供了一条新的解决途径。Evaluation of the health status of the river systems is important to the sustainable management of rivers and the regional ecological environment construction.Based on the summary of the traditional evaluation methods,this paper proposes the BP neural network assessment model in terms of the random samples.This model uses the stochastic method to generate the random evaluation samples between the higher and lower bounds for each standard evaluation grade,and the development of the model and training steps are conducted to obtain the final evaluation model for the health status of the river systems,which can reduce the effects of the uncertainty due to the lack of evaluation samples.The evaluation model is used practically in the Sanchakou river section in the Haihe River.The results show that the real coded and binary coded outputs result in a good health status of 2.95 grade from the evaluation model based on the random samples,and the binary coded output also suggests that this river has a tendency towards unhealthy conditions.Compared to the results obtained from the fuzzy comprehensive evaluation model,the evaluation model presented in this paper can yield the river health status closer to the actual conditions,which provides a new solution to study the river health status.

关 键 词：河流系统健康 随机样本 神经网络 BP算法 评价 

分 类 号：X82[环境科学与工程—环境工程] X171