Application of principal component analysis (PCA) to the assessment of parameter correlations in the partial-nitrification process using aerobic granular sludge. (15th June 2021)

• Record Type: Journal Article
• Authors: Cui, Fenghao; Kim, Minkyung; Park, Chul; Kim, Dokyun; Mo, Kyung; Kim, Moonil
• Is Part Of: Journal of environmental management. Volume 288(2021)
• Publication Date: 2021-06-15
• Subjects: Aerobic granular sludge -- Microbial communities -- Nitrogen removal -- Partial-nitrification -- Principal component analysis; Environmental policy -- Periodicals; Environmental management -- Periodicals; Environment -- Periodicals; Ecology -- Periodicals; 363.705
• DOI: 10.1016/j.jenvman.2021.112408
• Languages: English
• ISSNs: 0301-4797

Abstract

Abstract: For the first time, principal component analysis (PCA) was used to extract relevant information hidden in the partial-nitrification process using aerobic granular sludge. The objectives of this research are (a) to determine total ammonia nitrogen (TAN), total nitrite nitrogen (NO2 –N), nitrate nitrogen (NO3 –N), and other water quality parameters; (b) to identify the diversity of nitrification and denitrification bacterial community of wastewater samples during the partial-nitrification process using aerobic granular sludge and; (c) to analyze the correlation of available parameters using PCA. The nitrite accumulation ratio was determined from TAN, NO2 –N, and NO3 –N. Other water quality parameters were mixed liquor volatile suspended solids (MLVSS), alkalinity, total nitrogen (TN) and sludge volume index (SVI), pH, and dissolved oxygen (DO). The identification of bacterial community was conducted using 16S rRNA gene-based pyrosequencing by GS Junior Sequencing system. The water quality parameters were computed for PCA using software MATLAB. A nitrite accumulation ratio (NAR) between 0.55 and 0.85 was determined while maintaining the aerobic granular sludge's compact and dense structure. The PCA was used to reduce the data dimensionality from the original 8 variables to 2 principal components explaining 75% of the total data variance. Applying PCA to the data analysis in biological wastewater treatment can support detecting data anomalies and separating useful information from unwanted interferences. Graphical abstract: Image 1 Highlights: A partial-nitrification process was performed in an aerobic granulation reactor. PCA was applied to find relevant information hidden within experimental data. Hotelling' T 2 statistic could be used to the detection of anomalies in the dataset. Biplots show highly correlated nitrification, denitrification, and granulation. PCA finding was validated using bacterial community identification.