基于投影寻踪回归方法的微压过滤冲洗池水头损失与过滤效率预测模型  

Prediction model for head loss and filtration efficiency of micro-pressure filtration flushing tanks based on projection pursuit regression

作　　者：陶洪飞[1,2] 杨玉敏 吴梓境 马合木江·艾合买提 李巧[1,2] 姜有为 杨文新[1,2] TAO Hongfei;YANG Yumin;WU Zijing;Mahemujiang Aihemat;LI Qiao;JIANG Youwei;YANG Wenxin(College of Hydraulic and Civil Engineering,Xinjiang Agricultural University,Urumqi 830052,China;Xinjiang Key Laboratory of Hydraulic Engineering Safety and Water Disaster Prevention,Urumqi 830052,China)

机构地区：[1]新疆农业大学水利与土木工程学院,新疆乌鲁木齐830052 [2]新疆水利工程安全与水灾害防治重点实验室,新疆乌鲁木齐830052

出　　处：《水资源与水工程学报》2024年第3期207-216,共10页Journal of Water Resources and Water Engineering

基　　金：国家自然科学基金项目(52369013);新疆维吾尔自治区重大科技专项(2022A02003-4);昌吉州“两区”科技发展计划——科技成果转化示范项目(2023LQG08);新疆维吾尔自治区创新环境(人才、基地)建设专项——青年科学基金项目(2021D01B58);2019年度新疆维吾尔自治区人民政府公派出国留学成组配套项目。

摘　　要：以微压过滤冲洗池的水头损失与过滤效率为考核指标,考虑进水流量、矿化度、含沙量与滤网孔径4个因素进行正交试验设计,采用投影寻踪回归方法PPR分析各个因素对考核指标的影响权重,选取20组试验数据建立含沙微咸水条件下微压过滤冲洗池水头损失和过滤效率的预测模型,探究了含沙微咸水对微压过滤冲洗池的水头损失和过滤效率的影响。研究结果表明,影响微压过滤冲洗池水头损失的因素由大到小依次为进水流量、含沙量、矿化度、滤网孔径;影响过滤效率的因素由大到小依次为含沙量、进水流量、矿化度、滤网孔径;构建的PPR预测模型的预测精度整体合格率达100%;当进水流量为6~7 m^(3)/h、含沙量为0.5~1.0 g/L、矿化度为0~2.0 g/L及滤网孔径为0.125~0.180 mm时,水头损失存在最小值;当进水流量为9~10 m^(3)/h、含沙量为1.75~2.00 g/L、矿化度为0~3.0 g/L及滤网孔径为0.125~0.150 mm时,过滤效率存在最大值。物理模型的试验结果可为微压过滤冲洗池的实际应用提供研究基础。Using the head loss and filtration efficiency of the micro-pressure filtration flushing tank as the assessment indicators,an orthogonal experimental design of four factors,including inlet flow rate,mineralization degree,sediment concentration and filter diameter,was carried out.The projection pursuit regression(PPR)method was used to analyze the weight of the four factors,and then 20 sets of experimental data were selected to establish a prediction model for the head loss and filtration efficiency of the micro-pressure filtration flushing tank under the condition of sandy and brackish water,by which the influence of sandy and brackish water on the head loss and filtration efficiency of the micro-pressure filtration flushing tank was explored.The research results indicate that the factors affecting the head loss of the micro-pressure filtration flushing tank ranked in descending order are as follows:inlet flow rate,sediment concentration,mineralization degree,and filter diameter;whereas those affecting filtration efficiency ranked in descending order are:sediment concentration,inflow rate,salinity,and filter diameter.In addition,the PPR model can produce an overall qualification rate of 100%for prediction accuracy.According to the prediction results,when the inlet flow rate is 6-7 m^(3)/h,the sediment concentration is 0.5-1.0 g/L,the mineralization degree is 0-2.0 g/L,and the filter diameter is 0.125-0.180 mm,there is a minimum value of head loss;when the inlet flow rate is 9-10 m^(3)/h,the sediment concentration is 1.75-2.00 g/L,the mineralization degree is 0-3.0 g/L,and the filter diameter is 0.125-0.150 mm,there is a maximum value of filtration efficiency.The experimental results of the physical model can provide a technical basis for the practical application of micro-pressure filtration flushing tanks.

关 键 词：投影寻踪回归 矿化度 水头损失 过滤效率 预测模型 进水流量 含沙量 

分 类 号：S275.6[农业科学—农业水土工程] TV93[农业科学—农业工程]