常规方法难以处理实际流域水安全复杂系统的综合评价问题。本文在提出流域水安全评价方法论及其数学表达形式的基础上,探讨了用遗传算法、人工神经网络方法、模糊数学方法和蒙特卡洛方法等智能方法进行流域水安全评价的理论基础和这几种方法的适用范围。分析结果表明,采用智能方法进行流域水安全评价,有助于揭示流域水安全系统的复杂特征。推进这方面的研究将有助于发展和完善水资源综合评价理论,是水资源系统工程新的发展方向,具有广泛的应用前景。
The intelligem methods for evaluating water security in watershed, such as the genetic algorithm, artificial neural networks, fuzzy mathematical method and Monte Carlo method, are discussed from the viewpoints of methodology and their mathematical expressions, and the suitable application scopes for these methods are analyzed. It is found that the intelligent evaluation methods are very useful in revealing the complicated characteristics of watershed security. To improve the theory for intelligent evaluation of watershed water security is a new subject in developing the technology for reasonable utilization of watershed water resources.