Research

14 pages, 3903 KiB

Open AccessArticle

Fault Detection of Planetary Gears Based on Signal Space Constellations

by Maite Martincorena-Arraiza, Carlos A. De La Cruz Blas, Antonio Lopez-Martin, Cristián Molina Vicuña and Ignacio R. Matías

Sensors 2022, 22(1), 366; https://0-doi-org.brum.beds.ac.uk/10.3390/s22010366 - 04 Jan 2022

Cited by 2 | Viewed by 1509

Abstract

A new method to process the vibration signal acquired by an accelerometer placed in a planetary gearbox housing is proposed, which is useful to detect potential faults. The method is based on the phenomenological model and consists of the projection of the healthy [...] Read more.

A new method to process the vibration signal acquired by an accelerometer placed in a planetary gearbox housing is proposed, which is useful to detect potential faults. The method is based on the phenomenological model and consists of the projection of the healthy vibration signals onto an orthonormal basis. Low pass components representation and Gram–Schmidt’s method are conveniently used to obtain such a basis. Thus, the measured signals can be represented by a set of scalars that provide information on the gear state. If these scalars are within a predefined range, then the gear can be diagnosed as correct; in the opposite case, it will require further evaluation. The method is validated using measured vibration signals obtained from a laboratory test bench. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

14 pages, 3301 KiB

Open AccessArticle

Early Fault Diagnosis Method for Batch Process Based on Local Time Window Standardization and Trend Analysis

by Yuman Yao, Yiyang Dai and Wenjia Luo

Sensors 2021, 21(23), 8075; https://0-doi-org.brum.beds.ac.uk/10.3390/s21238075 - 02 Dec 2021

Cited by 1 | Viewed by 1686

Abstract

The products of a batch process have high economic value. Meanwhile, a batch process involves complex chemicals and equipment. The variability of its operation leads to a high failure rate. Therefore, early fault diagnosis of batch processes is of great significance. Usually, the [...] Read more.

The products of a batch process have high economic value. Meanwhile, a batch process involves complex chemicals and equipment. The variability of its operation leads to a high failure rate. Therefore, early fault diagnosis of batch processes is of great significance. Usually, the available information of the sensor data in batch processing is obscured by its noise. The multistage variation of data results in poor diagnostic performance. This paper constructed a standardized method to enlarge fault information as well as a batch fault diagnosis method based on trend analysis. First, an adaptive standardization based on the time window was created; second, utilizing quadratic fitting, we extracted a data trend under the window; third, a new trend recognition method based on the Euclidean distance calculation principle was composed. The method was verified in penicillin fermentation. We constructed two test datasets: one based on an existing batch, and one based on an unknown batch. The average diagnostic rate of each group was 100% and 87.5%; the mean diagnosis time was the same; 0.2083 h. Compared with traditional fault diagnosis methods, this algorithm has better fault diagnosis ability and feature extraction ability. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

20 pages, 2621 KiB

Open AccessArticle

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

by Yuequn Zhang, Lei Luo, Xu Ji and Yiyang Dai

Sensors 2021, 21(20), 6715; https://0-doi-org.brum.beds.ac.uk/10.3390/s21206715 - 09 Oct 2021

Cited by 11 | Viewed by 2044

Abstract

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) [...] Read more.

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. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

25 pages, 3880 KiB

Open AccessArticle

Deep-Compact-Clustering Based Anomaly Detection Applied to Electromechanical Industrial Systems

by Francisco Arellano-Espitia, Miguel Delgado-Prieto, Artvin-Darien Gonzalez-Abreu, Juan Jose Saucedo-Dorantes and Roque Alfredo Osornio-Rios

Sensors 2021, 21(17), 5830; https://0-doi-org.brum.beds.ac.uk/10.3390/s21175830 - 30 Aug 2021

Cited by 9 | Viewed by 2841

Abstract

The rapid growth in the industrial sector has required the development of more productive and reliable machinery, and therefore, leads to complex systems. In this regard, the automatic detection of unknown events in machinery represents a greater challenge, since uncharacterized catastrophic faults can [...] Read more.

The rapid growth in the industrial sector has required the development of more productive and reliable machinery, and therefore, leads to complex systems. In this regard, the automatic detection of unknown events in machinery represents a greater challenge, since uncharacterized catastrophic faults can occur. However, the existing methods for anomaly detection present limitations when dealing with highly complex industrial systems. For that purpose, a novel fault diagnosis methodology is developed to face the anomaly detection. An unsupervised anomaly detection framework named deep-autoencoder-compact-clustering one-class support-vector machine (DAECC-OC-SVM) is presented, which aims to incorporate the advantages of automatically learnt representation by deep neural network to improved anomaly detection performance. The method combines the training of a deep-autoencoder with clustering compact model and a one-class support-vector-machine function-based outlier detection method. The addressed methodology is applied on a public rolling bearing faults experimental test bench and on multi-fault experimental test bench. The results show that the proposed methodology it is able to accurately to detect unknown defects, outperforming other state-of-the-art methods. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

14 pages, 2243 KiB

Open AccessEditor’s ChoiceArticle

A Deep Learning Model for Predictive Maintenance in Cyber-Physical Production Systems Using LSTM Autoencoders

by Xanthi Bampoula, Georgios Siaterlis, Nikolaos Nikolakis and Kosmas Alexopoulos

Sensors 2021, 21(3), 972; https://0-doi-org.brum.beds.ac.uk/10.3390/s21030972 - 01 Feb 2021

Cited by 59 | Viewed by 7784

Abstract

Condition monitoring of industrial equipment, combined with machine learning algorithms, may significantly improve maintenance activities on modern cyber-physical production systems. However, data of proper quality and of adequate quantity, modeling both good operational conditions as well as abnormal situations throughout the operational lifecycle, [...] Read more.

Condition monitoring of industrial equipment, combined with machine learning algorithms, may significantly improve maintenance activities on modern cyber-physical production systems. However, data of proper quality and of adequate quantity, modeling both good operational conditions as well as abnormal situations throughout the operational lifecycle, are required. Nevertheless, this is difficult to acquire in a non-destructive approach. In this context, this study investigates an approach to enable a transition from preventive maintenance activities, that are scheduled at predetermined time intervals, into predictive ones. In order to enable such approaches in a cyber-physical production system, a deep learning algorithm is used, allowing for maintenance activities to be planned according to the actual operational status of the machine and not in advance. An autoencoder-based methodology is employed for classifying real-world machine and sensor data, into a set of condition-related labels. Real-world data collected from manufacturing operations are used for training and testing a prototype implementation of Long Short-Term Memory autoencoders for estimating the remaining useful life of the monitored equipment. Finally, the proposed approach is evaluated in a use case related to a steel industry production process. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

22 pages, 4392 KiB

Open AccessArticle

Integrated Diagnostic Framework for Process and Sensor Faults in Chemical Industry

by Jiaxin Zhang, Wenjia Luo and Yiyang Dai

Sensors 2021, 21(3), 822; https://0-doi-org.brum.beds.ac.uk/10.3390/s21030822 - 26 Jan 2021

Cited by 6 | Viewed by 2120

Abstract

This study considers the problem of distinguishing between process and sensor faults in nonlinear chemical processes. An integrated fault diagnosis framework is proposed to distinguish chemical process sensor faults from process faults. The key idea of the framework is to embed the cycle [...] Read more.

This study considers the problem of distinguishing between process and sensor faults in nonlinear chemical processes. An integrated fault diagnosis framework is proposed to distinguish chemical process sensor faults from process faults. The key idea of the framework is to embed the cycle temporal algorithm into the dynamic kernel principal component analysis to improve the fault detection speed and accuracy. It is combined with the fault diagnosis method based on the reconstruction-based contribution graph to diagnose the fault variables and then distinguish the two fault types according to their characteristics. Finally, the integrated fault diagnosis framework is applied to the Tennessee Eastman process and acid gas absorption process, and its effectiveness is proved. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

22 pages, 4722 KiB

Open AccessArticle

Application of Composite Spectrum in Agricultural Machines

by Fernando Feijoo, Francisco Javier Gomez-Gil and Jaime Gomez-Gil

Sensors 2020, 20(19), 5519; https://0-doi-org.brum.beds.ac.uk/10.3390/s20195519 - 26 Sep 2020

Cited by 4 | Viewed by 2012

Abstract

Composite spectrum (CS) is a data-fusion technique that reduces the number of spectra to be analyzed, simplifying the analysis process for machine monitoring and fault detection. In this work, vibration signals from five components of a combine harvester (thresher, chopper, straw walkers, sieve [...] Read more.

Composite spectrum (CS) is a data-fusion technique that reduces the number of spectra to be analyzed, simplifying the analysis process for machine monitoring and fault detection. In this work, vibration signals from five components of a combine harvester (thresher, chopper, straw walkers, sieve box, and engine) are obtained by placing four accelerometers along the combine-harvester chassis in non-optimal locations. Four individual spectra (one from each accelerometer) and three CS (non-coherent, coherent and poly-coherent spectra) from 18 cases are analyzed. The different cases result from the combination of three working conditions of the components—deactivated (off), balanced (healthy), and unbalanced (faulty)—and two speeds—idle and maximum revolutions per minute (RPM). The results showed that (i) the peaks can be identified in the four individual spectra that correspond to the rotational speeds of the five components in the analysis; (ii) the three formulations of the CS retain the relevant information from the individual spectra, thereby reducing the number of spectra required for monitoring and detecting rotating unbalances within a combine harvester; and, (iii) data noise reduction is observed in coherent and poly-coherent CS with respect to the non-coherent CS and the individual spectra. This study demonstrates that the rotating unbalances of various components within agricultural machines, can be detected with a reduced number of accelerometers located in non-optimal positions, and that it is feasible to simplify the monitoring with CS. Overall, the coherent CS may be the best composite spectra formulation in order to monitor and detect rotating unbalances in agricultural machines. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

19 pages, 9034 KiB

Open AccessArticle

Series AC Arc Fault Detection Method Based on High-Frequency Coupling Sensor and Convolution Neural Network

by Ruobo Chu, Patrick Schweitzer and Rencheng Zhang

Sensors 2020, 20(17), 4910; https://0-doi-org.brum.beds.ac.uk/10.3390/s20174910 - 31 Aug 2020

Cited by 29 | Viewed by 4107

Abstract

Arc faults induced by residential low-voltage distribution network lines are still one of the main causes of residential fires. When a series arc fault occurs on the line, the value of the fault current in the circuit is limited by the load. Traditional [...] Read more.

Arc faults induced by residential low-voltage distribution network lines are still one of the main causes of residential fires. When a series arc fault occurs on the line, the value of the fault current in the circuit is limited by the load. Traditional circuit protection devices cannot detect series arcs and generate a trip signal to implement protection. This paper proposes a novel high-frequency coupling sensor for extracting the features of low-voltage series arc faults. This sensor is used to collect the high-frequency feature signals of various loads in series arc state and normal working state. The signal will be transformed into two-dimensional feature gray images according to the temporal-domain sequence. A neural network with a three-layer structure based on convolution neural network is designed, trained and tested using the various typical loads’ arc states and normal states data sets composed of these images. This detection method can simultaneously accurately identify series arc, as well as the load type. Seven different domestic appliances were selected for experimental verification, including a desktop computer, vacuum cleaner, induction cooker, fluorescent lamp, dimmer, heater and electric drill. Then, the stability and universality of the detection algorithm is also verified by using electronic load with adjustable power factor and peak factor. The experimental results show that the designed sensor has the advantages of simple structure and wide frequency response range. The detection algorithm comparison confirms that the classification accuracy of the seven domestic appliances’ work states in the fourteen categories could reach 98.36%. The adjustable load in the two categories could reach above 99%. The feasibility of hardware implementation based on FPGA of this method is also evaluated. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Graphical abstract

19 pages, 557 KiB

Open AccessArticle

Nonlinear Chemical Process Fault Diagnosis Using Ensemble Deep Support Vector Data Description

by Xiaogang Deng and Zheng Zhang

Sensors 2020, 20(16), 4599; https://0-doi-org.brum.beds.ac.uk/10.3390/s20164599 - 16 Aug 2020

Cited by 13 | Viewed by 2342

Abstract

As one classical anomaly detection technology, support vector data description (SVDD) has been successfully applied to nonlinear chemical process monitoring. However, the basic SVDD model cannot achieve a satisfactory fault detection performance in the complicated cases because of its intrinsic shallow learning structure. [...] Read more.

As one classical anomaly detection technology, support vector data description (SVDD) has been successfully applied to nonlinear chemical process monitoring. However, the basic SVDD model cannot achieve a satisfactory fault detection performance in the complicated cases because of its intrinsic shallow learning structure. Motivated by the deep learning theory, one improved SVDD method, called ensemble deep SVDD (EDeSVDD), is proposed in order to monitor the process faults more effectively. In the proposed method, a deep support vector data description (DeSVDD) framework is firstly constructed by introducing the deep feature extraction procedure. Different to the traditional SVDD with only one feature extraction layer, DeSVDD is designed with multi-layer feature extraction structure and optimized by minimizing the data-enclosing hypersphere with the regularization of the deep network weights. Further considering the problem that DeSVDD monitoring performance is easily affected by the model structure and the initial weight parameters, an ensemble DeSVDD (EDeSVDD) is presented by applying the ensemble learning strategy based on Bayesian inference. A series of DeSVDD sub-models are generated at the parameter level and the structure level, respectively. These two levels of sub-models are integrated for a holistic monitoring model. To identify the cause variables for the detected faults, a fault isolation scheme is designed by applying the distance correlation coefficients to measure the nonlinear dependency between the original variables and the holistic monitoring index. The applications to the Tennessee Eastman process demonstrate that the proposed EDeSVDD model outperforms the traditional SVDD model and the DeSVDD model in terms of fault detection performance and can identify the fault cause variables effectively. Full article

(This article belongs to the Special Issue Cyberphysical Sensing Systems for Fault Detection and Identification)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Cyberphysical Sensing Systems for Fault Detection and Identification

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI