A method for filling missing values ​​of water quality monitoring data

Abstract

The invention discloses a method for filling missing values ​​of water quality monitoring data. The present invention regards the water quality monitoring data of the water quality monitoring station as a data set, which contains normal water quality data and missing water quality data, and then uses the improved OCS-FCM algorithm to solve the missing data in the data set, mainly as Continuously update the membership matrix iteratively until the preset number of iterations is reached, and then stop the iteration. Finally, the solved missing data is filled into the original data set to obtain a complete data set without missing data. The method of the invention overcomes the disadvantage of difficult selection of membership degree matrix parameters of the traditional FCM algorithm, adopts a method of updating the membership degree matrix in real time, and realizes the improvement of the correct rate of missing data filling, especially in the case of a data set with a large missing rate.

Description

technical field [0001] The invention belongs to the field of water quality monitoring, and in particular relates to a method for filling missing values ​​of water quality monitoring data. Background technique [0002] In water quality monitoring engineering applications, the data collected by front-end sensors is often not directly available, and data preprocessing has become an indispensable step in engineering applications. Due to sensor aging or system instability, there are often data loss phenomena in the process of water quality monitoring data collection, resulting in the absence of water quality monitoring data sets. An important part of the processing link. The missing value is the NULL value in the database. At present, the common NULL value processing methods in the field of water quality monitoring are as follows: [0003] (1) Direct discard method. That is, delete the NULL value in the water quality monitoring database. This method is simple and crude, and is...

AI Technical Summary

Problems solved by technology

In order to solve the problem of low filling accuracy of missing water quality data, it is necessary to propose a new filling method for missing water quality data. This method is based on an improved OCS-FCM water quality monitoring data filling method for missing values. The parameters of the OCS-FCM algorithm are optimized by using the degree matrix method to improve the clustering performance of the algorithm, and finally a complete water quality monitoring data set with an ideal filling effect is obtained.

Images