Water Quality Monitoring Missing Data Filling Method Based on Improved OCS-FCM

In order to improve the accuracy of filling missing data in water quality monitoring, an improved OCS-FCM (Fuzzy C-means clustering algorithm and Optimal Completion Strategy) method for filling missing data is proposed. The missing values of COD Mn , DO, pH and TP measured by water quality monitoring stations in Hangzhou were filled in, and the comparative experiments were carried out in the case of single attribute and multiple attribute missing data set, as well as in the case of different missing rate. The results show that the improved OCS-FCM water quality monitoring missing data filling method with real-time updating membership matrix has better filling accuracy than the similar algorithm in the comparative experiment, especially in the case of high missing rate. In addition, the accuracy of filling missing values in multi-attribute datasets is significantly higher than that in single-attribute datasets. The improved OCS-FCM water quality monitoring missing data filling method has better filling effect to avoid large missing data rate and multi-attribute data sets.