BP神经网络在大坝安全综合评价中的应用

提出了基于改进的BP神经网络的大坝安全综合评价方法，将改进的BP神经网络应用于大坝综合评价，意在建立更加接近于人类思维模式的定性与定量相结合的综合评价模型。通过对给定学习样本模式的学习，获得取学习样本中所体现的评价专家的经验、知识、主观判断及对目标重要性的倾向。当需对评价对象作出新的综合评价时，该方法便可再现评价专家的经验、知识和直觉思维，从而实现定性分析与定量分析的有效结合，较好地保证评价结果的质量，示例表明，这一评价方法是合理、可行的。

Application of BP artificial neural network to comprehensive evaluation of dam safety

An improved BP neural network was applied to comprehensive evaluation of dam safety, and a comprehensive evaluation method close to the thinking mode of human beings, with qualitative and quantitative analysis combined, was developed. By study of some given samples, the experts' experiences, knowledge, subjective judgment, and tendency towards the importance of objectives embodied in the samples were obtained. When it is necessary to comprehensively evaluate a new object, the method can reproduce the experts' experiences, knowledge and instinctive thought, and effectively combine qualitative and quantitative analysis together, ensuring the quality of dam safety evaluation. Examples show that the method is reasonable and feasible.