Abstract
Fault detection and diagnosis in the chemical industry is a challenging task due to the large number of measured variables and complex interactions among them. To solve this problem, a new fault diagnosis method named Fisher discriminative sparse representation (FDSR), based on deep belief network (DBN), is proposed in this paper. We used DBN to extract the features of all faulty and normal modes. Features extracted by the DBN were used to learn subdictionaries, then the overcomplete dictionary was constructed by cascading all subdictionaries in order, and each dictionary atom corresponded to class labels. The Fisher discrimination criterion (FDC) was applied to the dictionary learning to ensure smaller within-class scatter but greater between-class scatter. The quadratic programming method was applied to estimate the sparse coefficients simultaneously class by class. Therefore, both the reconstruction error and sparse coefficients were discriminative, so that the reconstruction error after sparse coding can be used for pattern classification. An experiment performed on the Tennessee Eastman (TE) process indicated that compared with the traditional monitoring methods, the FDSR based on DBN produced more interpretable results and achieved superior performance for feature extraction and classification in the field of complex system fault diagnosis. T-distributed stochastic neighbor embedding (t-SNE) appropriately visualized the performance of the proposed method and produced reliable fault diagnosis results.
Highlights
With the increase in the complexity of operation and multiloop control processes, a robust fault detection and diagnosis (FDD) method has become increasingly necessary for industrial processes.Fault detection and diagnosis involves developing a method that detects faults quickly, diagnosing the fault type, and determining the root causes of the fault with the least number of false alarms [1,2].A good fault detection and diagnosis system guarantees the quality of the product and improves the safety of the system.Fault diagnosis methods are categorized into data-based, model-based, knowledge-based, and hybrid methods [3]
The deep belief network (DBN)-Fisher discriminative sparse representation (FDSR) method was applied to the Tennessee Eastman (TE) process to validate its performance
The connections can be recognized as a filter since they eliminate useless information from inputs while maintaining information that reflects certain fault information
Summary
With the increase in the complexity of operation and multiloop control processes, a robust fault detection and diagnosis (FDD) method has become increasingly necessary for industrial processes. To reveal the deeper connection between the variables and create a proper representation of physical meaning, this paper proposes a novel feature extraction scheme for complex system fault diagnosis using deep learning and sparse representation. Proposed the linear superposition of multiple DBNs with quantum intervals in the last hidden layer to construct a deep quantum inspired neural network and applied it to the fault diagnosis of aircraft fuel systems [24]. To reduce the dimension of variables and determine the relationship with interpretability, we introduced deep belief networks to extract features of raw data from industrial processes. Wu et al proposed a novel fault diagnosis and detection method based on sparse representation classification (SRC), with the main contribution being the model training and multiclassification strategy [31].
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call