SMOTE_Bagging integrated sewage treatment fault diagnosis method based on weighted extreme learning machine

Abstract

The invention discloses an SMOTE_Bagging integrated sewage treatment fault diagnosis method based on a weighted extreme learning machine, the method comprises the following steps that (1) the defect items of samples with incomplete attributes in sewage data are supplemented with an averaging method and normalized to be in an interval of [0,1]; (2) the number of base classifiers and the optimal parameters of hidden nodes of the base classifiers are set; (3) independent oversampling is performed to the training sample corresponding to each base classifier with an improved SMOTE algorithm aimingat each base classifier, and the base classifiers are trained; (4) the output weight of each classifier is determined on the basis of a G-mean method; (5) integration is performed to all base classifiers after training, and a final integration classifier is obtained. According to the SMOTE_Bagging integrated sewage treatment fault diagnosis method based on the weighted extreme learning machine, the diversity among the base classifiers is improved while the unbalancedness of sewage data is effectively reduced, the classification accuracy of sewage treatment fault classes is improved, and further the whole performance of fault diagnosis in the sewage treatment process is effectively improved.

Description

technical field [0001] The invention relates to the field of sewage treatment fault diagnosis, in particular to a SMOTE_Bagging integrated sewage treatment fault diagnosis method based on a weighted extreme learning machine. Background technique [0002] Wastewater treatment is a complex, multivariable biochemical process. The failure of sewage treatment plants can easily lead to a series of serious sewage pollution problems. The fault diagnosis of sewage treatment process is a classification problem of pattern recognition. The sewage dataset consists of data collected on a regular basis from the sewage plant and its current working status. Due to the low frequency of failure of qualified sewage treatment plants, the data in the failure state in the sewage data set is far less than the data in the normal state. As a result, the sewage dataset is a highly unbalanced distribution, that is, the fault diagnosis of sewage treatment process is a classification problem with unba...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the defect of poor identification effect of sewage faults in the existing learning algorithm, and propose a SMOTE_Bagging integrated sewage treatment fault diagnosis method based on weighted extreme learning machine, which effectively reduces the imbalance of sewage data At the same time, it improves the diversity of base classifiers, improves the classification accuracy of sewage treatment faults, and effectively improves the overall performance of fault diagnosis in sewage treatment.

Images