Hierarchical cognize framework for the multi-fault diagnosis of the interconnected system based on domain knowledge and data fusion

Zhang, Tong; Tao, Laifa; Wang, Xiaoding; Zhang, Cong; Li, Shangyu; Jie, Hao; Lu, Chen; Suo, Mingliang

doi:10.1016/j.eswa.2022.116503

Cited by 13 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, it is found that data association knowledge becomes more interpretable, intuitive, and transparent than numerical information, which provides new ideas in dealing with the problem of insufficient data information faced by machine learning. Zhang et al [ 29 ] tried to integrate causal correlation knowledge into neural network learning, which is considered a new way for complex chemical process fault diagnoses. However, most of the existing methods can only provide qualitative knowledge‐based analyses, which still demonstrate difficulty in being merged with data characteristics.…”

Section: Introductionmentioning

confidence: 99%

ACNNapproach based on correlation metrics to chemical process fault classifications with limited labelled data

Yin

2023

Can J Chem Eng

View full text Add to dashboard Cite

This paper proposes a novel correlation metrics‐based convolutional neural network (CNN) classification model for chemical process fault diagnoses, creating a heuristic representation concerning process variable locations in grey correlation space (GCS) in terms of the copula entropy to guide the learning of classifiers. The proposed method based on correlation metrics can help solve the problem of insufficient information caused by a lack of labelled data. Specifically, variable correlations are fused into a heuristic matrix to provide prior knowledge for network learning in compensating data information before the CNN is employed to build the classifier for mining features in GCS. Driven by this mechanism, fault classifications in the case of small numbers of fault samples are successfully implemented. With successful simulation experiments carried out on the Tennessee Eastman (TE) process platform, we found that in GCS, different fault samples can represent hugely different features, while data resulting from the same fault rarely contribute to different ones. This observation lays a solid foundation for constructing superior fault classifiers. In addition, compared with conventional approaches, the proposed method has demonstrated better fault classification performances in the case of limited labelled fault samples.

show abstract

Section: Introductionmentioning

confidence: 99%

ACNNapproach based on correlation metrics to chemical process fault classifications with limited labelled data

Yin

2023

Can J Chem Eng

View full text Add to dashboard Cite

show abstract

“…In our case, we would add one more category for the relationship between components, components can be related if they affect the recorded data in a similar manner. In [16], the authors do multi-fault prediction focusing on the co-occurrence of faults. The assumptions of their approach is that there is appropriate data to represent all of the failures modes of the system, and that two fault modes must be discernible with the given data.…”

Section: Introducing Similar Faultsmentioning

confidence: 99%