Sensor and Component Fault Detection and Diagnosis for Hydraulic Machinery Integrating LSTM Autoencoder Detector and Diagnostic Classifiers

Mallak, Ahlam; Fathi, Madjid

doi:10.20944/preprints202012.0058.v1

Cited by 3 publications

(3 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It was utilized to predict the conditions of a hydraulic system's four components as a case study. In the experiments, the proposed OMDC approach was applied to a wellknown and real-world hydraulic system's condition dataset used in many ML-based studies [23][24][25].…”

Section: Introductionmentioning

confidence: 99%

An Ordinal Multi-Dimensional Classification (OMDC) for Predictive Maintenance

Taşer¹

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Predictive Maintenance is a type of condition-based maintenance that assesses the equipment's states and estimates its failure probability and when maintenance should be performed. Although machine learning techniques have been frequently implemented in this area, the existing studies disregard to the natural order between the target attribute values of the historical sensor data. Thus, these methods cause losing the inherent order of the data that positively affects the prediction performances. To deal with this problem, a novel approach, named Ordinal Multi-dimensional Classification (OMDC), is proposed for estimating the conditions of a hydraulic system's four components by taking into the natural order of class values. To demonstrate the prediction ability of the proposed approach, eleven different multi-dimensional classification algorithms (traditional Binary Relevance (BR), Classifier Chain (CC), Bayesian Classifier Chain (BCC), Monte Carlo Classifier Chain (MCC), Probabilistic Classifier Chain (PCC), Classifier Dependency Network (CDN), Classifier Trellis (CT), Classifier Dependency Trellis (CDT), Label Powerset (LP), Pruned Sets (PS), and Random k-Labelsets (RAKEL)) were implemented using the Ordinal Class Classifier (OCC) algorithm. Besides, seven different classification algorithms (Multilayer Perceptron (MLP), Support Vector Machine (SVM), k-Nearest Neighbour (kNN), Decision Tree (C4.5), Bagging, Random Forest (RF), and Adaptive Boosting (AdaBoost)) were chosen as base learners for the OCC algorithm. The experimental results present that the proposed OMDC approach using binary relevance multi-dimensional classification methods predicts the conditions of a hydraulic system's multiple components with high accuracy. Also, it is clearly seen from the results that the OMDC models that utilize ensemble-based classification algorithms give more reliable prediction performances with an average Hamming score of 0.853 than the others that use traditional algorithms as base learners.

show abstract

Section: Introductionmentioning

confidence: 99%

An Ordinal Multi-Dimensional Classification (OMDC) for Predictive Maintenance

Taşer¹

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

show abstract

“…To enhance the temporal modeling capability, various methods used long short-term memory (LSTM) and gated recurrent units (GRU). 31–33 A fusion model based on sensor time series named 1D-CNN_GRU is developed to diagnose a chiller’s fault. 34 Due to the unique advantage of the “end-to-end” network structure, LSTM- and GRU-based encoder–decoder network (EDN) structures are used in the FDD of sensors.…”

Section: Introductionmentioning

confidence: 99%

A novel improved data-temporal attention network reconstruction model-based method for fault diagnosis of chiller sensors

Hong

Zhao

et al. 2022

Proceedings of the Institution of Mechanical Engineers, Part I:

View full text Add to dashboard Cite

Accurate measurement of sensors is the basis for the reliable operation of complex systems. To improve the fault diagnosis performance of the air-cooled chiller sensor, an improved data-temporal attention network is proposed, and an improved data-temporal attention network–based method for sensor fault diagnosis is established. First, the method combines the “memory” capability of the bidirectional improved long short-term memory network and the feature extraction capability of the multi-scale convolutional neural network to fully explore the time correlation of the chiller sensor, the data correlation between the physical quantities, and the dynamic response characteristics of the physical quantities. In addition, data and time attention mechanisms are introduced into the encoder and decoder, respectively, and valuable information is enhanced through weight distribution to capture relevant features further. Second, relying on the advantages of the “end-to-end” network structure of the improved data-temporal attention network model, the fault sensor is directly located by comparing the absolute reconstruction error vector with the fault threshold vector. Finally, it is verified by the datasets collected from a real air-cooled chiller platform that the proposed method has achieved an excellent fault diagnosis effect. The ablation experiment confirms that each improvement step can effectively improve the reconstruction performance of the network, which enhances the sensitivity of fault sensor identification. Compared with existing mainstream methods, the state-of-the-art framework presented in this article reveals its superiority.

show abstract

“…The SAE weights are computed by the unsupervised pre-training are applied for the fine-tuning supervised stage, which is a more effective strategy than random weight initialization [ 30 ]. For this reason, several industrial case-applied soft sensors have been proposed based on SAE [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 ]. The cited-above successful applications of SAE-based deep learning demonstrate a strong ability for feature extraction.…”

Section: Introductionmentioning

confidence: 99%

Industrial Semi-Supervised Dynamic Soft-Sensor Modeling Approach Based on Deep Relevant Representation Learning

Lima

Araújo

2021

Sensors

View full text Add to dashboard Cite

Soft sensors based on deep learning have been growing in industrial process applications, inferring hard-to-measure but crucial quality-related variables. However, applications may present strong non-linearity, dynamicity, and a lack of labeled data. To deal with the above-cited problems, the extraction of relevant features is becoming a field of interest in soft-sensing. A novel deep representative learning soft-sensor modeling approach is proposed based on stacked autoencoder (SAE), mutual information (MI), and long-short term memory (LSTM). SAE is trained layer by layer with MI evaluation performed between extracted features and targeted output to evaluate the relevance of learned representation in each layer. This approach highlights relevant information and eliminates irrelevant information from the current layer. Thus, deep output-related representative features are retrieved. In the supervised fine-tuning stage, an LSTM is coupled to the tail of the SAE to address system inherent dynamic behavior. Also, a k-fold cross-validation ensemble strategy is applied to enhance the soft-sensor reliability. Two real-world industrial non-linear processes are employed to evaluate the proposed method performance. The obtained results show improved prediction performance in comparison to other traditional and state-of-art methods. Compared to the other methods, the proposed model can generate more than 38.6% and 39.4% improvement of RMSE for the two analyzed industrial cases.

show abstract

Sensor and Component Fault Detection and Diagnosis for Hydraulic Machinery Integrating LSTM Autoencoder Detector and Diagnostic Classifiers

Cited by 3 publications

References 14 publications

An Ordinal Multi-Dimensional Classification (OMDC) for Predictive Maintenance

An Ordinal Multi-Dimensional Classification (OMDC) for Predictive Maintenance

A novel improved data-temporal attention network reconstruction model-based method for fault diagnosis of chiller sensors

Industrial Semi-Supervised Dynamic Soft-Sensor Modeling Approach Based on Deep Relevant Representation Learning

Contact Info

Product

Resources

About