Ensemble model of wastewater treatment plant based on rich diversity of principal component determining by genetic algorithm for status monitoring |
| |
Affiliation: | 1. Department of Electrical and Computer Engineering, University of Calgary, 2500 University Dr NW, Calgary, AB T2N 1N4, Canada;2. Department of Chemical and Biological Engineering, University of British Columbia, 2360 East Mall, Vancouver, BC V6T 1Z3, Canada |
| |
Abstract: | Wastewater treatment plants (WWTPs) is a complex process, effective process monitoring can make it stable and prevent the destruction of the ecological environment. Principal component analysis (PCA) has been widely used in process monitoring. However, most PCA-based methods construct a single PCA model using several principal components (PCs), causing loss of information on some faults and less generalization ability of the PCA model. Thus, this study proposed a novel ensemble process monitoring method based on genetic algorithm (GA) for selective diversity of PCs. GA is used to determine a set of principal component subspaces with the greatest diversity as the base models. Bayesian inference is adopted to combine the results of base models into a probability index. Cases study on TE benchmark process and an actual WWTP show the excellent performance of the proposed method compared with several PCA-based methods and the strong generalization ability of the ensemble model. |
| |
Keywords: | Principal component analysis Ensemble learning Bayesian inference Process monitoring Genetic algorithm |
本文献已被 ScienceDirect 等数据库收录! |
|