【摘要】 为了准确、高效地评定地下水水质,提出了一种遗传算法与神经网络相结合的混合评价算法,针对水质评价的多变量和非线性,采用BP神经网络对其进行综合评价计算,BP算法易陷入局部极小的缺点则通过引入遗传算法来克服,将两者有机的结合起来实现神经网络的训练和知识库的建立.通过算法比较和实例结果分析,证明了该算法的有效性.更多还原

【Abstract】 A hybrid evaluation algorithm is proposed in this paper by combining BP neural network with genetic algorithms,in order to evaluate groundwater quality accurately and efficiently.In view of the multi-variable and nonlinear characteristics of water assessment,BP neural network is introduced here to make comprehensive evaluation and calculation.As for the shortcoming that BP algorithm is easily trapped to a local optimum,it can be overcomed through the introduction of genetic algorithms,and the two will work organically together to achieve the training and knowledge base establishment of the neural network.Through the comparison of the algorithms and analysis of the results of the examples,the result shows this algorithm is valid.更多还原