论文部分内容阅读
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.
In order to evaluate the quality of groundwater accurately and efficiently, a hybrid evaluation algorithm based on genetic algorithm and neural network is proposed. According to the multivariable and nonlinear water quality evaluation, BP neural network is used to evaluate and evaluate the groundwater quality. The shortcoming of local minima is overcome by the introduction of genetic algorithm, and the two are combined organically to train the neural network and build the knowledge base.Through the comparison of algorithms and the analysis of example results, the effectiveness of the algorithm is proved.