基于活性污泥絮体微观参数的污泥沉降性能判别  

Discriminant of activated sludge settling ability based on floc microscopic parameters

作　　者：张新喜[1,2] 完颜健飞 胡小兵[1,2] 叶星[3] 欧阳英[1] 

机构地区：[1]安徽工业大学建筑工程学院,马鞍山243032 [2]生物膜法水质净化及利用技术教育部工程研究中心,马鞍山243032 [3]安徽工业大学能源与环境学院,马鞍山243032

出　　处：《环境科学学报》2015年第12期3815-3823,共9页Acta Scientiae Circumstantiae

基　　金：国家自然科学基金(No.51208001);安徽省高校省级科学研究项目(No.KJ2013A059);国家级创新创业计划项目(No.201210360078)~~

摘　　要：针对活性污泥沉降性检测仍采用人工检测污泥容积指数(SVI)的现状,采用活性污泥絮体微观图像分析技术,以实现对污泥沉降性简单、快速的判别.将20个微观参数通过主成分分析归纳为3个污泥絮体微观特征综合指标,即大小因子、形态因子和浓度因子,并以3个综合指标建立污泥沉降性能Fisher判别模型,再分别以原始训练样本数据和2座污水厂3个月的检测数据作为测试样本对模型可靠程度进行检验.结果发现,判别正确率分别为79.8%和80.6%.研究表明,模型判别结果较为直观,能直接表达为"正常"或"膨胀",可为实现污泥沉降性的在线自动判别提供技术基础.To determine the activated sludge settling ability simply and quickly, the microscopic image analysis on activated sludge floc is adopted to improve the traditional method based on the measurement of Sludge Volume Index （SVI）. Three comprehensive indexes, namely size factor, morphology factor and concentration factor, can be achieved from 20 microscopic parameters through the principal component analysis （PCA）. Based on the indexes, a discriminant model is established to make discriminant analysis （DA） on the sludge setting ability. To study the reliability of the model, data from the original training samples and monitoring samples on two wastewater treatment plants during three months are used to test independently and the corresponding accuracy rates are 79.8% and 80.6%, respectively. Results indicate that the discriminant model can provide a simple and practical method for on-line determining the ＂normal＂ or ＂bulking＂ state of activated sludge.

关 键 词：沉降性 污泥膨胀 图像分析 主成分分析 判别分析 

分 类 号：X703.1[环境科学与工程—环境工程]