【摘要】 研究水质评价问题,针对水环境保护和治理,由于水质类型和分级标准存在着模糊性,水质因子与水质类型间存在非线性关系,无法建立传统精确的数学模型。为了提高水质评价的准确性,提出模糊BP神经神经网络水质评价模型。利用模糊理论相对隶属度能反映各因子的质量相对状态,对模糊性具有很强的识别精度,可将水质因子模糊隶属度矩阵输入到BP神经网络中,通过非线性分析处理的BP神经网络进行学习和测试,得到输出水质评价等级。实验结果表明,模糊神经网络较传统的水质评价方法具有较高的识别精度,提高水质评价等级的准确性,使评价的结果更具有科学性。更多还原

【Abstract】 Ambiguity exist inwater type and classification standards,and the nonlinear relation exists between influence factors and water quality.The relative membership degree of fuzzy theory,can well reflect the quality of each factor’s fuzzy relative state,and the BP neural network is of strong nonlinear analysis ability.The paper put forward the water quality evaluation model based on BP neural network and fuzzy theory ans applied this model into practical quality evaluation.Using indicators relative membership degree matrix interpolation,a neural network training samples were produced.Then,the model was applied to assess the quality of water.The results show that the fuzzy neural network makes the evaluation result more scientific than BP neural network.更多还原