Prediction of anaerobic digestion performance and identification of critical operational parameters using machine learning algorithms

• Published in: Bioresource technology Vol. 298; p. 122495
• Main Authors: Wang, Luguang, Long, Fei, Liao, Wei, Liu, Hong
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.02.2020
• Subjects: Algorithms; anaerobic digesters; Anaerobic digestion; Anaerobiosis; artificial intelligence; carbon; Cluster Analysis; data collection; Logistic Models; Machine Learning; Methane production; Operational parameters; Prediction; regression analysis; Operational parameters; Methane production; Anaerobic digestion; Machine learning; Prediction
• ISSN: 0960-8524; 1873-2976; 1873-2976
• DOI: 10.1016/j.biortech.2019.122495

[Display omitted] •Accurate prediction of methane production via machine learning algorithms was achieved.•Regression and classification models were both examined using various algorithms.•The removal of outliers in the validation set could improve prediction accuracy.•Feature importance revealed total carbon as the determinant operational parameter. Machine learning has emerges as a novel method for model development and has potential to be used to predict and control the performance of anaerobic digesters. In this study, several machine learning algorithms were applied in regression and classification models on digestion performance to identify determinant operational parameters and predict methane production. In the regression models, k-nearest neighbors (KNN) algorithm demonstrates optimal prediction accuracy (root mean square error = 26.6, with the dataset range of 259.0–573.8), after narrowing prediction coverage by excluding extreme outliers from the validation set. In the classification models, logistic regression multiclass algorithm yields the best prediction accuracy of 0.73. Feature importance reveals that total carbon was the determinant operational parameter. These results demonstrate the great potential of using machine learning algorithms to predict anaerobic digestion performance.