Deep learning approach for sustainable WWTP operation: A case study on data-driven influent conditions monitoring

A. Dairi, T. Cheng, F. Harrou, Y. Sun, T.O. Leiknes
Sustainable Cities and Society, volume 50, pp. 101670, (2019)

Deep learning approach for sustainable WWTP operation: A case study on data-driven influent conditions monitoring

Keywords

Wastewater treatment plant, Influent conditions monitoring, Machine learning, Unsupervised deep learning

Abstract

Wastewater treatment plants (WWTPs) are sustainable solutions to water scarcity. As initial conditions offered to WWTPs, influent conditions (ICs) affect treatment units states, ongoing processes mechanisms, and product qualities. Anomalies in ICs, often raised by abnormal events, need to be monitored and detected promptly to improve system resilience and provide smart environments. This paper proposed and verified data-driven anomaly detection approaches based on deep learning methods and clustering algorithms. Combining both the ability to capture temporal auto-correlation features among multivariate time series from recurrent neural networks (RNNs), and the function to delineate complex distributions from restricted Boltzmann machines (RBM), RNN-RBM models were employed and connected with various classifiers for anomaly detection. The effectiveness of RNN based, RBM based, RNN-RBM based, or standalone individual detectors, including expectation maximization clustering, K-means clustering, mean-shift clustering, one-class support vector machine (OCSVM), spectral clustering, and agglomerative clustering algorithms were evaluated by importing seven years ICs data from a coastal municipal WWTP where more than 150 abnormal events occurred. Results demonstrated that RNN-RBM-based OCSVM approach outperformed all other scenarios with an area under the curve value up to 0.98, which validated the superiority in feature extraction by RNN-RBM, and the robustness in multivariate nonlinear kernels by OCSVM. The model was flexible for not requiring assumptions on data distribution, and could be shared and transferred among environmental data scientists.

Code

DOI: 10.1016/j.scs.2019.101670

Sources

Website PDF

See all publications 2019