基于PSO-SVM算法的环境监测数据异常检测和缺失补全

Anomaly Detection and Missing Completion of Environment Monitoring Data based on PSO-SVM

针对环境监测数据异常和数据缺失问题,提出了基于支持向量机的粒子群优化数据异常检测和缺失补全算法。利用粒子群优化算法选取较优的支持向量机训练参数组合,以此建立非线性的支持向量机模型,并利用结果模型对测得的真实数据拟合预测。以宁夏回族自治区某污水处理厂的污染物测量数据作为实验数据,结果表明,利用该算法预测数据的准确率可达97.977%,检测异常数据准确度高,缺失数据补全正确。

For problems of abnormal data and missing data in environmental monitoring, an anomaly detectionand data missing completion algorithm was presented based on particle swarm optimization with support vectormachine （PSO-SVM ）. Non-linear SVM model was established by applying the PSO algorithm in selecting theappropriate training parameter set and fitting prediction of real data. Taking the experimental data from a sewageplant in Ningxia Hui Autonomous Region, the predictions by this algorithm had the accuracy rate of 97.977% ,showing high accuracy in abnormal data detection and missing data completion.