为验证混合蛙跳算法(Shuffled Frog Leaping Algorithm,SFLA)-投影寻踪(Projection Pursuit,PP)模型应用于区域水资源利用效率综合评价中的有效性和可行性。从综合、工业、农业、生活和生态环境5个方面遴选15个指标构建区域水资源利用效率评价指标体系,利用SFLA算法优化PP模型最佳投影方向,提出SFLA-PP水资源利用效率评价模型,与构建的入侵杂草优化(Invasive Weed Optimization,IWO)算法、粒子群优化(Particle Swarm Optimization,PSO)算法和布谷鸟搜索(Cuckoo Search,CS)算法优化的PP模型作对比,并基于各模型投影系列均值及标准差σ构造4个等级的水资源利用效率评价标准,以文山州8县市水资源利用效率评价为例进行实例研究。结果表明:SFLA算法优化PP模型获得的适应度值为715.800 2,均优于IWO,PSO和CS算法PP模型,具有较好的求解精度和极值寻优能力;SFLA-PP模型对文山市、砚山县水资源利用效率评价为“高水平”,对麻栗坡县评价为“低水平”,其余5县评价为“中等水平”,全州水资源利用效率综合评价为“较高水平”;SFLA-PP模型对实例评价及排序结果与IWO-PP模型相同;与PSO-PP模型、CS-PP模型在评价结果及排序上均存在差异。实例验证了4种智能算法PP模型的求解精度对区域水资源利用效率的评价结果起到关键作用。

In the aim of testifying the efficiency and feasibility of shuffled frog leaping algorithm(SFLA)- projection pursuit (PP) model in assessing utilization efficiency of regional water resources, an evaluation system consisting 15 indexes in terms of comprehensive aspect, industry, agriculture, living and ecological environment was established. Furthermore, SFLA algorithm was used to optimize the best projection direction of PP model. In consequence, a SFLA-PP model for evaluating utilization efficiency was proposed and compared with PP model optimized by invasive weed optimization (IWO) algorithm, particle swarm optimization (PSO) algorithm and cuckoo search (CS) PP algorithm, respectively. Then an evaluation standard of four levels (high, medium-high, medium and low) was constructed on the basis of the mean value and standard deviation σ of the model projection series. Eight counties of Wenshan autonomous prefecture were chosen for case analysis. Results showed that the fitness value of PP model optimized with SFLA algorithm was 715.800 2, superior to that with IWO, PSO and CS algorithm, with better precision and ability of seeking optimum extreme value. According to SFLA-PP model, the utilization efficiency of Wenshan City and Yanshan County was evaluated as high level, and that of Malipo County was evaluated as low level, and the remaining five counties as medium level. The utilization efficiency of water resources in the study area was at high level in general. Case evaluation and sorted results of SFLA-PP model were the same with those of IWO-PP model, but different from PSO-PP and CS-PP model. Finally, examples demonstrated that the solving accuracy of the four intelligent algorithms was crucial for the evaluation results.