从人水和谐的角度对水安全进行定义。在前人研究的基础上组建水安全评价的指标体系,引进支持向量机技术,利用其在统计学习理论中结构风险最小化的原理,以及其解决非线性和高维识别中的优势,特别是能克服主观因素的不利影响,对水安全进行综合评价,并以水资源总量比较丰富,但由于时空分布不均存在水安全问题的浙江省为例进行评价。 Water safety is defined from the perspective of human-water harmony. On the basis of previous researches, this paper establishes the index system of water safety evaluation, introduces the support vector machine technology, makes use of its principle of minimizing structural risk in statistical learning theory and its advantages in solving nonlinear and high-dimensional identification, especially Can overcome the adverse effects of subjective factors and make a comprehensive assessment of water security. With a relatively large amount of water resources, Zhejiang Province, an issue of water security with uneven spatial and temporal distribution, is taken as an example for evaluation.