【摘要】 为了克服传统模糊综合评价法中权重赋值的片面性及最大隶属度原则导致评价结果不合理的弊端,提出了基于层次分析法和熵权法的主客观组合权重,并结合相乘相加算子及加权平均原则,建立了改进的模糊综合评价法。选取TP、TN、DO、NH₃-N、COD_Mn和BOD₅为评价指标,用改进的模糊综合评价法对重庆某水库2012年7—12月的水质进行了综合评价。结果表明:TN为主要的污染物;水质最差时段为8月,评价结果为2.655,水质属于Ⅱ级,略偏于Ⅲ级。评价结果与人工神经网络模型所得结果基本一致。更多还原

【Abstract】 The paper is aimed at presenting a innovated fuzzy comprehensive evaluation method（FCEM） to replace the traditional one in hoping to improve the evaluation of the water quality of reservoir.The reason why we stress to improve the water quality of reservoir is that its water serves as one of the major sources of urban water supply,whose quality is immediately related to the health of urban residents.The single method of weight allocation in the traditional fuzzy comprehensive evaluation method（FCEM）,which is considered to be partial and the maximum membership principle may often be likely to cause the loss of data and therefore result in inadequate reliability in terms of evaluation.To overcome the said inadequacy,this paper would like to propose a renovated version of FCEM,which can join the analytical hierarchy process（AHP） and the entropy-weight method（EWM） into one to improve the part of the weight allocation in addition to the comprehensive use of the addition operator and average-weighing principle in the evaluation.In one word,the suggested innovation can not only include both the subjective and the objective weights,but also takes into account all the data including weight and membership.As a step of trial application we have successfully applied the improved method to the evaluation of the water quality of a reservoir in Chongqing.As the result of our actual evaluation of the reservoir water quahty,the following indexes have been chosen as the indicative pollution indexes for the water of the reservoirs in the city:TN,TP,DO,NH₃-N,COD_Mn,and BOD₅.The conclusion we have made from the result of evaluation can be stated as follows:the lowest point of the water quality comes out in August（Water quality grade Ⅲ） because of the high concentration of TN and TP（mostly TN,with an average weight coefficient of 0.317）.The reliability of the evaluation results prove to be consistent with the actual water quality.In addition,we have also done some other evaluations,whose results also turn to be well in accord with the artificial neural network method.Thus,it can be seen that the evaluation of the reservoir water quality is in a position to validate both the objectivity and practicability of the actual water situation.What is more,the improved version of the fuzzy comprehensive evaluation of the water quality can help to improve the water quahty maintenance and the environmental quality management.更多还原