Fault Diagnosis Using Deep Learning for Wastewater Treatment Processes

During the long operation of wastewater treatment systems, there is a risk of various failures due to aging equipment and environmental climate change. Hence, to ensure the stability of the wastewater treatment procedure, it is necessary to perform prompt and efficient fault diagnosis. This paper presents an optimized fusion deep learning network designed for feature extraction and fault diagnosis within wastewater treatment processes. The method combines knowledge-based fault diagnosis advantages and creates a reliable model using a multi-scale convolutional neural network to capture spatio-temporal features. The proxy-nearest neighborhood component analysis optimizes model parameters and improves fault classification performance compared to traditional cross entropy and the fault classification performance of the model was validated using faults based on the wastewater treatment process system BSM1. It achieves a 81.52% average accuracy for 11 fault types, outperforming conventional models and single sub-models with different loss functions. This study demonstrates the greater potential of the proposed model approach for fault diagnosis in wastewater treatment.