Breakage Detection Research Articles

Lipoblastoma in one adult and 35 pediatric patients: Retrospective analysis of 36 cases.

Lipoblastoma is a rare benign mesenchymal neoplasm that typically occurs at various sites in infants and children but may also occur in adults. Thus, differential diagnoses are often performed. To understand this tumor type, the present study described clinicopathological features, diagnosis and differential diagnosis of different morphological lipoblastomas. A single-institution retrospective review of 36 lipoblastoma cases diagnosed between 2015 and 2021 was performed. Formalin-fixed paraffin-embedded tissue was used for S-100, CD34, P16 and desmin immunohistochemistry analysis, along with rapid fluorescence in situ hybridization (FISH) detection with pleiomorphic adenoma gene 1 (PLAG1). The 36 cases included 14 females and 22 males [age range, 7 days to 33 years (median, 16.5 years); 28 patients were aged ≤3 years] and the tumors were located in the trunk (n=16), limbs (n=12), head and neck (n=6), and perineum (n=2). Histologically, lipoblastomas were divided into classic (n=15), lipoma-like (n=13) and myxoid (n=8) subtypes. They comprised lobules of mature adipose tissue of varying size and a fine capillary network surrounded by mucinous stroma. Single- or multivesicular lipoblasts positive for S-100 (29/36, 81%) were observed, with occasional mature adipocytes. Peripheral vessels and cytoplasm of primitive mesenchymal cells were diffusely positive for CD34 (36/36, 100%), whereas primitive mesenchymal cells and striated muscle tissue were positive for desmin (26/36, 72%). Most tumor cells were negative while only few were positive for P16 (8/36, 22%). FISH revealed PLAG1 breakage and rearrangement in 24/32 (75%) patients. In total, 28 patients were followed up post-operatively (range, 2-84 months; median, 41 months; 3 patients relapsed and 8 were lost to follow-up). In conclusion, diagnosis of a typical lipoblastoma is not difficult and PLAG1 breakage detection is key for the diagnosis.

A comparison of four machine learning techniques and continuous wavelet transform approach for detection and classification of tool breakage during milling process

In machining, the tool condition has to be monitored by condition monitoring techniques to prevent damage by the use of tools and the workpiece. Cutting forces acting on the tool between zero and maximum values cause the cutting edge to crack and break. Predetection of this situation in the cutting tool is very important to prevent any negative situation that may occur. This study introduces a vibration-based intelligent tool condition monitoring technique to detect involute form cutter faults such as tool breakage at different levels during gear production on a milling machine. Machine learning algorithms such as artificial neural network, random forest, support vector machine, and K-nearest neighbor were used to detect the broken teeth and its level of breakage. According to the results obtained, it was observed that all the algorithms are successful in detecting faults in different teeth; also they have identification advantages according to different fault levels. In addition, the time and frequency domain analysis and continuous wavelet transform were used to determine the local faults. The developed machine learning-based detection performances compared the classical time and frequency domain analyses and continuous wavelet transform to prove the effectiveness and precision of the proposed methods. The results showed that all of the machine learning techniques have satisfactory performance to be used as fast and precise detection tools without complex calculations for detecting tool breakage.

An Enhanced Pathfinder Algorithm based MCSA for rotor breakage detection of induction motor

This paper proposes a new broken rotor bar (BRB) finding technique based on a novel hybrid algorithm, namely the Enhanced Pathfinder Algorithm (EPFA). Compared to existing optimization algorithms, this proposed approach provides more balanced exploration and exploitation. The main task regarding rotor fault detection is to identify sideband frequency components (SFCs) corresponding to the fault and in this context proposed method overcomes the disadvantages of traditional frequency domain analysis, such as short data length, effects of spectral leakage, and insufficient sampling frequency. To achieve the proposed diagnosis approach, a signal model for faulty current and a function of residual error are used for the formulation of a nonlinear least-square problem. The method is first tested on a synthetic fault signal, and finally, to validate the proposed technique, a Finite Element Model (FEM) of the induction motor is simulated in the ANSYS 2D workbench.

Theoretical and experimental evaluation of the distance dependence of fiber-based fluorescence and reflection measurements for laser lithotripsy

Objectives. In laser lithotripsy, a green aiming beam overlying the infrared (IR) treatment radiation gives rise to reflection and fluorescence signals that can be measured via the treatment fiber. While stone autofluorescence is used for target detection, the condition of the fiber can be assessed based on its Fresnel reflection. For good applicability, fluorescence detection of stones should work even when the stone and fiber are not in direct contact. Fiber breakage detection, on the other hand, can be falsified if surfaces located in front of the fiber reflect light from the aiming laser back into it. For both applications, therefore, a fundamental investigation of the dependence of the signal amplitude on the distance between fiber and surface is important. Methods. Calculations of the signal drop of fluorescence or diffuse and specular reflection with increasing fiber distance were performed using ray tracing based on a simple geometric model for different fiber core diameters. Reflection signals from a mirror, diffuse reflector, human calculi, and porcine renal tissue placed in water were measured at varying distances (0–5 mm). For human calculi, fluorescence signals were recorded simultaneously. Results. The calculations showed a linear signal decrease down to ∼60% of the maximum signal (fiber in contact). The distance z at which the signal drops to for example 50% depends linearly on the diameter of the fiber core. For fibers used in lithotripsy and positioned in water, z 50% ranges from 0.55 mm (200 μm core diameter) to 2.73 mm, (1 mm core diameter). The calculations were in good agreement with the experimental results. Conclusions. The autofluorescence signals of stones can be measured in non-contact mode. Evaluating the Fresnel signal of the end face of the fiber to detect breakage is possible unless the fiber is situated less than some millimeters to reflecting surfaces.

Automatic detection of snow breakage at single tree level using YOLOv5 applied to UAV imagery

The assessment of forest abiotic damages such as snow breakage is important to ensure compensation to forest owners. Currently, information on the extent of snow breakage is gathered through time-consuming and potentially biased field surveys. In such situations where field surveys are still common practice, unmanned aerial vehicles (UAVs) are increasingly being used to provide a more cost-efficient and objective methods to answer forest information needs. Further, the advent of sophisticated computer vision techniques such as convolutional neural networks (CNNs) offers new ways to analyze image data more efficiently and accurately. We proposed an object detection method to automatically identify trees and classify them according to the damage by snow based on a YOLO CNN architecture. UAV imagery collected across 89 study areas and over the course of the entire year were manually annotated into a total of >55 K single trees classified as healthy, damaged, or dead. The annotated trees, along with the corresponding UAV imagery were used to train a YOLOv5 object detection model. Furthermore, we tested the effect of seasonality, and varying atmospheric and lighting conditions on the model’s performance. Based on an independent test set of data we found that the general model including all of the data (i.e. any seasons, atmospheric conditions, and time of the day) outperformed all other tested scenarios (i.e. precision = 62 %; recall = 61 %). Furthermore, we found that despite the fact that the snow damaged trees represented a minority class (i.e. 16 % of the annotated trees), they were detected with the largest precision (76 %) and recall (78 %). Finally, the general model transferred well across the variation in seasons, atmospheric and illumination conditions, making it suitable for usage for any new UAV image acquisition.

Insulator Breakage Detection Utilizing a Convolutional Neural Network Ensemble Implemented With Small Sample Data Augmentation and Transfer Learning

Online fault detection of insulators is a necessary requirement for the development of a smart grid, which directly affects the safety and reliability of power system operations. Breakage is an important factor causing insulator abnormalities, and a large number of image samples are often required for training to complete defect identification with the help of AI deep learning. In this paper, we propose a convolutional neural network ensemble implemented using small samples, data augmentation and transfer learning for insulator breakage detection. We used data augmentation to expand the small sample of 200 insulator images by 15 times to 3000 images, built a VGG16 transfer network, pre-trained the VGG16 network convolutional basis weights, and improved the training performance and efficiency of the model with freeze training and fine-tuning training. The training on 2700 images and testing results on 300 images show that the proposed method can achieve an accuracy of 0.9871 for insulator breakage recognition.

A Continual Learning-Based Framework for Developing a Single Wind Turbine Cybertwin Adaptively Serving Multiple Modeling Tasks

This article proposes a generalized neural continual learning-based cybertwin (GNC) modeling framework to realize developing one wind turbine (WT) cybertwin serving multiple modeling tasks in the wind farm operations and maintenance (O&M). A generalized WT cybertwin modeling problem, which considers modeling one cybertwin for multiple tasks without additional computational burden, is studied for the first time. Fully connected neural networks are adopted as the backbone for developing the GNC model. The online elastic weight consolidation method is incorporated to mitigate the catastrophic forgetting phenomenon among different modeling tasks. Computational experiments are conducted to validate the effectiveness of the proposed GNC framework based on the supervisory control and data acquisition data. Modeling tasks in three important problems of the wind farm O&M, the WT gearbox failure detection, WT blade breakage detection, and wind power prediction, are considered in the experiment. Compared with other benchmarking models, such as the multiple neural cybertwins, neural cybertwin, and regularized neural cybertwin, the proposed GNC can achieve high accuracies on both new tasks and existing tasks, which further verifies the WT cybertwin generalization via the proposed GNC.

High-impedance fault detection on downed conductor in overhead distribution networks

High impedance faults (HIF) may occur in the medium voltage distribution systems and constitute a danger to human beings. Many cases of HIF are caused due to downed power cables in the distribution network. These cables in contact with the ground generate a great risk to the population, animals and fire risk. In this paper, a methodology for detecting HIF, generated by cable breakage, is presented. The algorithm is based on the analysis of the harmonic and symmetrical components of the current. The method aims to gain reliability in HIF detection, where the detection of conductor breakage incorporated into the method allows a better distinction of HIF from other events. A distribution network based on a Brazilian real feeder was implemented, as well as two well know HIF models. The method is validated using real-time digital simulations and hardware implementation, or HIL (hardware-in-the-loop) concept with OPAL-RT real-time simulator. A comparison with other methods is also performed in order to evaluate the performance of the method. The results show that the proposed methodology has presented a better performance under different types of HIF, models of HIF, and achieved a general accuracy of 90% and security of 100% for the tests performed.

Integrated YOLO and CNN Algorithms for Evaluating Degree of Walkway Breakage

The focus of policymaking in Korea has changed from vehicle-centric road environments to people-centric environments. As the importance of walking has increased, the construction of pedestrian paths and interest in pedestrian environments have also increased. However, problem recognition and resolution require considerable time in the event of a problem in a pedestrian path. People with reduced mobility tend to resist changes in roads that they use. Thus, damaged pedestrian paths and obstacles pose a considerable risk and economic loss to transportation. In this study, we aimed to minimize the time and cost required for the evaluation of pedestrian paths by developing an automatic system for determining damage using integrated You Only Look Once (YOLO) and convolutional neural network (CNN) image deep learning algorithms. We constructed a model using image deep learning by dividing the steps into walkway breakage detection and score evaluation according to the degree of breakage. The accuracy of the model was determined to be 92%. In the future, the evaluation of pedestrian path damage is expected to be automated using images and videos, thereby reducing the time required for the detection and restoration of damage.

Insulator Breakage Detection Based on Improved YOLOv5

Aerial images have complex backgrounds, small targets, and overlapping targets, resulting in low accuracy of intelligent detection of overhead line insulators. This paper proposes an improved algorithm for insulator breakage detection based on YOLOv5: The ECA-Net (Efficient Channel Attention Network) attention mechanism is integrated into its backbone feature extraction layer, and the effective distinction between background and target is achieved by increasing the weight of important channels. A bidirectional feature pyramid network is added to the feature fusion layer, and large-scale images with more original information are combined to effectively retain small target features. Incorporating a flexible detection frame selection algorithm Soft-NMS (Soft Non-Maximum Suppression) into the prediction layer to re-screen the target frame, thereby reducing the probability of mistaken deletion of overlapping targets. The effectiveness of the improved YOLOv5 algorithm is verified in the actual aerial image dataset, and the results show that the mean Average Precision (mAP) of the improved algorithm is 95.02% and the detection speed FPS (Frames Per Second) can reach 49.4 frames/s, which meets the real-time and accuracy requirements of engineering applications.

A Study on Tool Breakage Detection During Milling Process Using LSTM-Autoencoder and Gaussian Mixture Model

A Study on Tool Breakage Detection During Milling Process Using LSTM-Autoencoder and Gaussian Mixture Model

CFRP drilling process control based on spindle power consumption from real production data in the aircraft industry

Ongoing challenges in advanced manufacturing highlight the need of improved control of the production system, higher production speed, lower tool wear, top quality standards together with reduced material waste to minimize associated costs, time and environmental footprint. The organization of production resources through the integration of data along the value chain using Information Technologies is required to achieve these challenges. Thus, the emergence of the fourth industry revolution brings with it the organization of productive resources through computational intelligence and connectivity. This research seeks to analyze the spindle power consumption in Carbon-fiber-reinforced polymer composites (CFRPs) drilling operations as a process control indicator in terms of tool wear. This signal stands out among others available because it can be obtained in real time, with high quality and through a non-intrusive methodology. In particular, the study is framed in the real production system in factories of Airbus. The industrial process data were directly collected from the manufacturing plant in Getafe (in the Madrid-Spain region) and correspond to more than 3000 holes drilled with diamond-coated tungsten carbide tools. The variability of machining conditions in the aeronautical component drilling process and inherent noise level of signals obtained in industrial environments required the development of a data wrangling methodology to structure and clean the information. As a result, different magnitudes were obtained from spindle power consumption signal related to tool wear with low levels of dependence on drilling conditions. The conclusions of this work are directly applicable to the control of industrial production systems within the framework of Industry 4.0, searching new improvement opportunities through Data analytics and Artificial Intelligence such as tool breakage detection or the optimization of cycle times.

Sensor signal selection for tool wear curve estimation and subsequent tool breakage prediction in a drilling operation

ABSTRACT Tool condition monitoring have an important role in machining processes to reduce defective component and ensure quality requirements. Stopping the process before the tool breaks or an excessive tool wear is reached can avoid costs resulting from that undesirable situation. This research work presents the results obtained in drilling process monitoring carried out on Inconel 718. Monitoring systems should be light and scalable. Following this idea, multiple sensors for external signal acquisition are used in this work (cutting forces, vibrations, and acoustic emissions) and several machine internal signals are collected. The main objective is to evaluate the capacity of each acquisition source for the reconstruction of the tool wear curve and subsequently detection of tool breakage. Given the difficulty of using all of these signals in a real system, the methodology used to analyse the data makes it possible to have a comparative analysis of the potential of each of these sources for tool wear monitoring during the drilling process. The results indicate cutting forces whether they come from internal signals or external signals can carry out this task accurately. At the same time of data acquisition, detailed tool wear measurements were added.

Wear and Breakage Detection of Integral Spiral End Milling Cutters Based on Machine Vision.

Tool wear and breakage detection technologies are of vital importance for the development of automatic machining systems and improvement in machining quality and efficiency. The monitoring of integral spiral end milling cutters, however, has rarely been investigated due to their complex structures. In this paper, an image acquisition system and image processing methods are developed for the wear and breakage detection of milling cutters based on machine vision. The image acquisition system is composed of three light sources and two cameras mounted on a moving frame, which renders the system applicable in cutters of different dimensions and shapes. The images captured by the acquisition system are then preprocessed with denoising and contrast enhancing operations. The failure regions on the rake face, flank face and tool tip of the cutter are extracted with the Otsu thresholding method and the Markov Random Field image segmentation method afterwards. Eventually, the feasibility of the proposed image acquisition system and image processing methods is demonstrated through an experiment of titanium alloy machining. The proposed image acquisition system and image processing methods not only provide high quality detection of the integral spiral end milling cutter but can also be easily converted to detect other cutting systems with complex structures.

An imbalanced data learning method for tool breakage detection based on generative adversarial networks

An imbalanced data learning method for tool breakage detection based on generative adversarial networks

Quick assessment of DNA damage in cervical epithelial cells using a chromatin dispersion test.

This study was aimed to quantify genomic DNA breakages in the cervical epithelium cells of patients diagnosed with different grades of cervical lesions using a quick test based on chromatin dispersion after controlled protein depletion. The association between the progressive stages of cervical dysplasia and the levels of DNA damage, taking into account the presence of papillomavirus human (HPV) infection, was investigated. A hospital-based unmatched case-control study was conducted during 2018 with a sample of 78 women grouped according to histological diagnosis as follows: 23 women with low grade-squamous intraepithelial lesion (LG-SIL), 34 women with high grade- squamous intraepithelial lesion (HG-SIL), and three women with cervical carcinoma (CC). In parallel, 15 women without cervical lesions were included as a Control cohort. DNA damage levels in cervical epithelial cells were assessed using the chromatin dispersion test (CDT) and controlled in parallel with DNA breakage detection coupled with florescent in situ hybridization (DBD‒FISH) using whole genomic DNA probes. CDT produces different morphotypes in the cervical epithelium that can be associated with the level of DNA breakage revealed with DBD‒FISH. A significant increase of DNA damage was correlated with the histological progression of the patients and human papillomavirus (HPV) infection. The CDT is a simple, accurate and inexpensive morphological bioassay to identify different levels DNA damage that can be associated with the level of abnormal cells present in the cervical epithelium in patients who commonly present HPV infection.

High performance machining of continuous metal fibers with cascaded multi-stage profile tools

Metallic fibers are essential for applications in the chemical industry, the automotive industry for brake pads and, more recently, for the air conditioning modules. The most productive process to date for producing continuous fiber lines with fiber cross-sections between 50 and 150 µm is the application of cascaded multi-stage profile tools for cutting from semi-finished wire products. One challenge is the development of a technology for highly ductile materials made of aluminium, which are used as fiber materials in the area of refrigeration adsorbers. Preliminary studies have shown that high cutting speeds are necessary for machining.For this reason, a highly manually operated machine system was advanced to meet the requirements of high cutting speeds and to integrated a process monitoring system. For the first time, experimental investigations were able to show the interrelationships between set technology values and the effective machining forces for this process. Further investigations were done using the example of aluminium alloys 5019 for an analysis of a control system for wire breakage detection on the basis of force sensors.

Online Yarn Breakage Detection: A Reflection-Based Anomaly Detection Method

This article discusses an optical reflection-based method for equipment anomaly detection that enhances weak signals and has high sensitivity to abnormalities. Automatic warp knitting machine yarn breakage detection, which has become an acknowledged difficulty in the textile field, is achieved. To the best of our knowledge, this is the first time that visual inspection has been applied to yarn breakage detection in weaving. Furthermore, based on the periodicity of the yarn distribution and the periodic motion law of the machine, the combined wavelet and Seasonal and Trend decomposition using locally weighted regression (LOESS) (STL) decomposition method is proposed for yarn breakage detection. Finally, the efficiency and accuracy of the proposed method are verified experimentally. Our research is a successful application of reflection characteristics to the anomaly detection of non-Lambertian objects, which has implications for the high-precision anomaly detection of precision equipment.

Investigation of signal behaviors for sensor fusion with tool condition monitoring system in turning

Monitoring of the cutting area with different type of sensors requires confirmation for composing sensor fusion to obtain longer tool life and high-quality product. The complex structure of machining and interaction between variables affect the influence of parameters on quality indicators. Using multiple sensors provide comparison of information acquired from different resources and make easier to decide about tool and workpiece condition. In this experimental research for the first time, five different sensors were adopted to a lathe for collecting data to measure the capability of each sensor in reflecting the tool wear. Cutting forces, vibration, acoustic emission, temperature and current measurements were carried out during turning of AISI 5140 with coated carbide tools. Considering the graphical investigation, the successes of sensors on detection of progressive flank wear and tool breakage were investigated. Besides, the effects of cutting parameters on measured variables were interpreted considering graphs. According to results, temperature and acoustic emission signals seem to be effective about 74% for flank wear. In addition, fuzzy logic based prediction of flank wear was performed with the assistance of temperature and acoustic emission sensors with high accuracy which demonstrates their availability for sensor fusion. Tool breakage occurs instantly which can prevent with the assistance of sensor signals and tangential and feed cutting forces, acoustic emission and vibration signals seem as reliable indicators for approaching major breakage. Sensor fusion based turning provides confirmed information which enables more reliable, robust and consistent machining.

Research on the performance of heavy-haul rail breakage detection using track circuit

Abstract Rail breakage is one of the major safety risks in railway transportation. Because of the large axle weight, high density and large capacity of the heavy-haul railway, the damage to rail caused by heavy-haul train wheels will be more serious than that caused by ordinary passenger and cargo trains, resulting in a higher frequency of rail breakage. Taking the Daqin Railway Line as the research object, this paper analyses and discusses rail breakages occurring in interstation tracks and in-station tracks by establishing the ZPW-2000A track circuit calculation model considering the land leakage resistance between the rail line tracks; introduces the defining standards and measurement index of the broken rail coefficient to quantitatively analyse the influence of various influencing factors on the rail breakage inspection performance under the most unfavourable working conditions; and compares the model simulation data, the laboratory model data and the field test results to verify its effectiveness, so as to provide a reference and theoretical basis for the subsequent improvement and solution of the heavy-haul railway rail breakage problem.