Convolutional Autoencoder-Based Flaw Detection for Steel Wire Ropes

Zhang, Guoyong; Tang, Zhaohui; Zhang, Jin; Gui, Weihua

doi:10.3390/s20226612

Cited by 13 publications

(8 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, recently developed recursive canonical correlation analysis algorithm do not require a reference for accurate assessment [17]. Another approach to ND also relies on a distance metric, but this time the distance is not measured from reference points stored in memory but from a sample's reconstructed version [5,[18][19][20][21]. In the last quoted approach, the boundary is defined by a trained model.…”

Section: Novelty Detection In Condition Monitoringmentioning

confidence: 99%

“…In articles [5,13,18,[22][23][24][25] novelty detection algorithms were presented for gearbox monitoring. Other applications include bearings fault detection [14,19,26,27], maintenance support system for gas turbines [10,11,28] and rope inspection [21]. The solutions presented in these articles are, however, usually at the experimental stage.…”

Section: Novelty Detection In Condition Monitoringmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of Novelty Detection Methods for Detection of Various Rotary Machinery Faults

Górski

Jabłoński

Heesch

et al. 2021

Sensors

View full text Add to dashboard Cite

Condition monitoring is an indispensable element related to the operation of rotating machinery. In this article, the monitoring system for the parallel gearbox was proposed. The novelty detection approach is used to develop the condition assessment support system, which requires data collection for a healthy structure. The measured signals were processed to extract quantitative indicators sensitive to the type of damage occurring in this type of structure. The indicator’s values were used for the development of four different novelty detection algorithms. Presented novelty detection models operate on three principles: feature space distance, probability distribution, and input reconstruction. One of the distance-based models is adaptive, adjusting to new data flowing in the form of a stream. The authors test the developed algorithms on experimental and simulation data with a similar distribution, using the training set consisting mainly of samples generated by the simulator. Presented in the article results demonstrate the effectiveness of the trained models on both data sets.

show abstract

Section: Novelty Detection In Condition Monitoringmentioning

confidence: 99%

Section: Novelty Detection In Condition Monitoringmentioning

confidence: 99%

Comparison of Novelty Detection Methods for Detection of Various Rotary Machinery Faults

Górski

Jabłoński

Heesch

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Nonetheless, the limited sample size and the reliance on hand-crafted features hampered robustness in noisy environments. This was evidenced in [24] which showed that the model accuracy drops to 80.5% when training/testing with a different dataset. Moreover, the utility of CNNs combined with image processing techniques was shown to increase the accuracy of the model in [22] from 93.3% to 95.5% [18].…”

mentioning

confidence: 93%

Real-Time Damage Detection in Fiber Lifting Ropes Using Convolutional Neural Networks

Jalonen¹,

Al-Sa'D²,

Mellanen³

et al. 2023

Preprint

View full text Add to dashboard Cite

The health and safety hazards posed by worn crane lifting ropes mandate periodic inspection for damage. This task is time-consuming, prone to human error, halts operation, and may result in the premature disposal of ropes. Therefore, we propose using deep learning and computer vision methods to automate the process of detecting damaged ropes. Specifically, we present a novel vision-based system for detecting damage in synthetic fiber rope images using convolutional neural networks (CNN). We use a camera-based apparatus to photograph the lifting rope's surface, while in operation, and capture the progressive wear-and-tear as well as the more significant degradation in the rope's health state. Experts from Konecranes annotate the collected images in accordance with the rope's condition; normal or damaged. Then, we pre-process the images, design a CNN model in a systematic manner, evaluate its detection and prediction performance, analyze its computational complexity, and compare it with various other models. Experimental results show the proposed model outperforms other techniques with 96.4% accuracy, 95.8% precision, 97.2% recall, 96.5% F1-score, and 99.2% AUC. Besides, they demonstrate the model's realtime operation, low memory footprint, robustness to various environmental and operational conditions, and adequacy for deployment in industrial systems.

show abstract

“…When the concentrated broken strand exceeds the scrap standard, the whole rope needs to be replaced, leading to a short life cycle. At the same time, in this short life cycle, the quantity of fracture defect data is small, which makes it difficult to use supervised learning methods to detect defects [7]. There is little literature on machine-vision-based defect detection for sealed wire ropes, but researchers have carried out a lot of work on surface defect detection for nonsealed ordinary wire ropes.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Steel Wire Rope Surface Defect Detection Based on Segmentation Template and Spatiotemporal Gray Sample Set

Zhang

Tang

Fan

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Machine-vision-based defect detection, instead of manual visual inspection, is becoming increasingly popular. In practice, images of the upper surface of cableway load sealing steel wire ropes are seriously affected by complex environments, including factors such as lubricants, adhering dust, natural light, reflections from metal or oil stains, and lack of defect samples. This makes it difficult to directly use traditional threshold-segmentation-based or supervised machine-learning-based defect detection methods for wire rope strand segmentation and fracture defect detection. In this study, we proposed a segmentation-template-based rope strand segmentation method with high detection accuracy, insensitivity to light, and insensitivity to oil stain interference. The method used the structural characteristics of steel wire rope to create a steel wire rope segmentation template, the best coincidence position of the steel wire rope segmentation template on the real-time edge image was obtained through multiple translations, and the steel wire rope strands were segmented. Aiming at the problem of steel wire rope fracture defect detection, inspired by the idea of dynamic background modeling, a steel wire rope surface defect detection method based on a steel wire rope segmentation template and a timely spatial gray sample set was proposed. The spatiotemporal gray sample set of each pixel in the image was designed by using the gray similarity of the same position in the time domain and the gray similarity of pixel neighborhood in the space domain, the dynamic gray background of wire rope surface image was constructed to realize the detection of wire rope surface defects. The method proposed in this paper was tested on the image set of Z-type double-layer load sealing steel wire rope of mine ropeway, and compared with the classic dynamic background modeling methods such as VIBE, KNN, and MOG2. The results show that the purposed method is more accurate, more effective, and has strong adaptability to complex environments.

show abstract

Convolutional Autoencoder-Based Flaw Detection for Steel Wire Ropes

Cited by 13 publications

References 24 publications

Comparison of Novelty Detection Methods for Detection of Various Rotary Machinery Faults

Comparison of Novelty Detection Methods for Detection of Various Rotary Machinery Faults

Real-Time Damage Detection in Fiber Lifting Ropes Using Convolutional Neural Networks

Steel Wire Rope Surface Defect Detection Based on Segmentation Template and Spatiotemporal Gray Sample Set

Contact Info

Product

Resources

About