基于自由搜索的投影寻踪水质综合评价方法

为解决水质评价指标的不相容性以及现有评价方法对等级的分辨率较粗略等问题,提出了水质状况评价的投影寻踪方法(Projection Pursuit,PP).根据地表水环境等级划分标准,构建了能够反映水质状况变异信息的投影指标函数;引入自由搜索(Free Search,FS)算法,优化投影指标函数寻求最佳投影方向;利用最佳投影方向信息研究了各种水质指标对水质状况评价的影响程度.以山西省汾河流域为例,进行水质评价分析,并与基于加速遗传算法优化投影寻踪模型(RAGAPP)和利用粒子群优化投影寻踪模型(PSOPP)得到的投影值结果进行对比.实例应用表明,该模型评估结果误差精度优于RAGAPP和PSOPP,评价结果符合当地水质状况实际情况,且对水质等级划分清晰.

Projection pursuit model for comprehensive evaluation of water quality based on free search

The method of projection pursuit(PP)was presented to evalute the water quality for solving the problems such as the incompatibility of water quality indexes and the rough grade levels evaluteed by other methods. Based on the grade division standard of water quality, the projection function which could reflect the variability of the water quality was constructed,and free search (FS) was introduced to optimize the function and seek the optimum projection vector, while the data of the vector were used to research the influential level of each water quality index. Take Fenhe River Basin for instance to evaluate the water quality, compared with the resules which evaluteed by RAGAPP and PSOPP. The achieved results suggested that the error precision of FSPP is superior to RAGAPP and PSOPP, the conclusion assessed by the model was accordant with the actual water quality condition, and the projection pursuit model could identify the grade of water quality, FSPP model is an effectivemethod for evaluating water quality. It is effective and general, which can be applied to comprehensive evaluation of other environmental quality.