LDA-based deep transfer learning for fault diagnosis in industrial chemical processes

Wang, Yalin; Wu, Dongzhe; Yuan, Xiaofeng

doi:10.1016/j.compchemeng.2020.106964

Cited by 67 publications

(30 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the testing set of each fault type, the fault is introduced at the 160th sample. Thus, the first 160 samples are in the normal state and the last 800 samples are in the corresponding fault state [53]. Because many FDD algorithms have been tested using a standard dataset by other researchers, it is convenient to compare the results to the general methods by experimenting on the same standard dataset.…”

Section: Resultsmentioning

confidence: 99%

Improved Random Forest Algorithm Based on Decision Paths for Fault Diagnosis of Chemical Process with Incomplete Data

Zhang

Luo

et al. 2021

Sensors

View full text Add to dashboard Cite

Fault detection and diagnosis (FDD) has received considerable attention with the advent of big data. Many data-driven FDD procedures have been proposed, but most of them may not be accurate when data missing occurs. Therefore, this paper proposes an improved random forest (RF) based on decision paths, named DPRF, utilizing correction coefficients to compensate for the influence of incomplete data. In this DPRF model, intact training samples are firstly used to grow all the decision trees in the RF. Then, for each test sample that possibly contains missing values, the decision paths and the corresponding nodes importance scores are obtained, so that for each tree in the RF, the reliability score for the sample can be inferred. Thus, the prediction results of each decision tree for the sample will be assigned to certain reliability scores. The final prediction result is obtained according to the majority voting law, combining both the predicting results and the corresponding reliability scores. To prove the feasibility and effectiveness of the proposed method, the Tennessee Eastman (TE) process is tested. Compared with other FDD methods, the proposed DPRF model shows better performance on incomplete data.

show abstract

Section: Resultsmentioning

confidence: 99%

Improved Random Forest Algorithm Based on Decision Paths for Fault Diagnosis of Chemical Process with Incomplete Data

Zhang

Luo

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Linear discriminant analysis (LDA) finds a linear combination of features of two classes of objects, in order to effectively characterize or distinguish between them [26]. The resulting combination can be used as a linear classifier.…”

Section: Methodsmentioning

confidence: 99%

Characterization of Intracellular Structure Changes of Microcystis under Sonication Treatment by Polarized Light Scattering

Zou

Peng

et al. 2021

Biosensors

View full text Add to dashboard Cite

Cyanobacterial bloom is one of the most urgent global environmental issues, which eventually could threaten human health and safety. Sonication treatment (ST) is a potential effective method to control cyanobacteria blooms in the field. Currently, the bottleneck of extensive application of ST is the difficulty to estimate the ST effect on the cyanobacterial cells and then determine suitable ST times in the field. In this study, cyanobacterial Microcystis samples sonicated at different times were first measured by a spectrophotometer to calculate the removal efficiency of Microcystis cells. Additionally, they were observed by TEM to reveal the intracellular structure changes of the cells. Then the samples were measured by an experimental setup based on polarized light scattering to measure the polarization parameters. Experimental results indicated that the polarization parameters can effectively characterize the intracellular structural changes of Microcystis cells with different ST times, which is quite consistent with the results for removal efficiency and TEM images. Further, the optimal ST time can be inferred by the polarization parameters. These results demonstrate that polarized light scattering can be a potentially powerful tool to explore suitable times for sonication treatment of cyanobacteria blooms.

show abstract

“…The methodology was applied to the vibration signal of wind turbines. In Reference [ 59 ], a linear discriminant analysis (LDA)–based on Deep transfer network is proposed for fault classification of chemical processes as the Tennessee Eastman benchmark and real hydrocracking processes. A maximum mean discrepancy based loss function is used to extract similar latent features and reduce the discrepancy of distributions between the source and target data.…”

Section: Related Workmentioning

confidence: 99%

RNN- and LSTM-Based Soft Sensors Transferability for an Industrial Process

Curreri

Patané

Xibilia

2021

Sensors

View full text Add to dashboard Cite

The design and application of Soft Sensors (SSs) in the process industry is a growing research field, which needs to mediate problems of model accuracy with data availability and computational complexity. Black-box machine learning (ML) methods are often used as an efficient tool to implement SSs. Many efforts are, however, required to properly select input variables, model class, model order and the needed hyperparameters. The aim of this work was to investigate the possibility to transfer the knowledge acquired in the design of a SS for a given process to a similar one. This has been approached as a transfer learning problem from a source to a target domain. The implementation of a transfer learning procedure allows to considerably reduce the computational time dedicated to the SS design procedure, leaving out many of the required phases. Two transfer learning methods have been proposed, evaluating their suitability to design SSs based on nonlinear dynamical models. Recurrent neural structures have been used to implement the SSs. In detail, recurrent neural networks and long short-term memory architectures have been compared in regard to their transferability. An industrial case of study has been considered, to evaluate the performance of the proposed procedures and the best compromise between SS performance and computational effort in transferring the model. The problem of labeled data scarcity in the target domain has been also discussed. The obtained results demonstrate the suitability of the proposed transfer learning methods in the design of nonlinear dynamical models for industrial systems.

show abstract

LDA-based deep transfer learning for fault diagnosis in industrial chemical processes

Cited by 67 publications

References 42 publications

Improved Random Forest Algorithm Based on Decision Paths for Fault Diagnosis of Chemical Process with Incomplete Data

Improved Random Forest Algorithm Based on Decision Paths for Fault Diagnosis of Chemical Process with Incomplete Data

Characterization of Intracellular Structure Changes of Microcystis under Sonication Treatment by Polarized Light Scattering

RNN- and LSTM-Based Soft Sensors Transferability for an Industrial Process

Contact Info

Product

Resources

About