Detection Of Wear Particles Research Articles

The Target Detection of Wear Particles in Ferrographic Images Based on the Improved YOLOv8

An enhanced YOLOv8 algorithm is proposed in the following paper to address challenging issues encountered in ferrographic image target detection, such as the identification of complex-shaped wear particles, overlapping and intersecting wear particles, and small and edge-wear particles. This aim is achieved by integrating the main body network with the improved Deformable Convolutional Network v3 to enhance feature extraction capabilities. Additionally, the Dysample method is employed to optimize the upsampling technique in the neck network, resulting in a clearer fused feature image and improved precision for detecting small and edge-wear particles. In the head network, parameter sharing simplifies the detection head while enhancing convergence speed and precision through improvements made to the loss function. The experimental results of the present study demonstrate that compared to the original algorithm, this enhanced approach achieves an average precision improvement of 5.6% without compromising the detection speed (111.6FPS), therefore providing valuable support for online monitoring device software foundations.

A New Target Detection Method of Ferrography Wear Particle Images Based on ECAM-YOLOv5-BiFPN Network.

For mechanical equipment, the wear particle in the lubrication system during equipment operation can reflect the lubrication condition, wear mechanism, and severity of wear between equipment friction pairs. To solve the problems of false detection and missed detection of small, dense, and overlapping wear particles in the current ferrography wear particle detection model in a complex oil background environment, a new ferrography wear particle detection network, EYBNet, is proposed. Firstly, the MSRCR algorithm is used to enhance the contrast of wear particle images and reduce the interference of complex lubricant backgrounds. Secondly, under the framework of YOLOv5s, the accuracy of network detection is improved by introducing DWConv and the accuracy of the entire network is improved by optimizing the loss function of the detection network. Then, by adding an ECAM to the backbone network of YOLOv5s, the saliency of wear particles in the images is enhanced, and the feature expression ability of wear particles in the detection network is enhanced. Finally, the path aggregation network structure in YOLOv5s is replaced with a weighted BiFPN structure to achieve efficient bidirectional cross-scale connections and weighted feature fusion. The experimental results show that the average accuracy is increased by 4.46%, up to 91.3%, compared with YOLOv5s, and the detection speed is 50.5FPS.

Online Gear Wear Particle Detection and Categorization Using a Convolutional Neural Network Algorithm Integrated with Cascade Classifier

Online Gear Wear Particle Detection and Categorization Using a Convolutional Neural Network Algorithm Integrated with Cascade Classifier

Numerical simulation of wear particle deposition in ferrograph based on coupled analysis of magnetic, fluid and solid field

Ferrography is a significant means of wear particle detection for mechanical equipment. In view of the issue that the wear particle deposition process by a ferrograph is difficult to accurately adjust to avoid accumulation or escape, the deposition behavior of wear particles coupled with the magnetic, fluid and solid field in a ferrograph is studied by the finite element method. The wear particle force and its deposition principle are first analyzed, and the theoretical deposition law is obtained. By building 3D finite element model, setting the magnetic, flow field and wear particle parameters, the coupled simulation based on the particle tracking is carried out. According to the process of wear particles flowing out of the oil tube and then depositing on the ferrogram slide, the effect of oil pressure, viscosity, density, ferrogram slide inclination angle, distance between oil tube and ferrogram slide and the particle release mode on the particle deposition behavior is studied. Based on the velocity and displacement of particles, the reasonable parameter ranges are obtained, and the sensitive variables are determined. In order to verify the numerical law through experiments, the area and perimeter of wear particles are adopted, meanwhile the oil viscosity is selected as the adjustable parameter, and the tested results prove the numerical simulation correct.

Research on the characteristics of micro planar capacitance sensor for multi wear particle detection

Wear particles are rich in information about the running state of mechanical equipment. Identifying wear particles is of great importance for monitoring and maintaining the safety and reliability of mechanical equipment. This paper investigated the characteristics of the micro planar capacitance sensor used for wear particle detection. The structural parameters of the micro planar capacitance sensor were optimized through finite element numerical simulation. The influences of excitation frequency and particle velocity on the sensor characteristics were studied through different experiments. The results show that when the excitation frequency is 1.8 MHz, and the particle velocity is 4 mm/s, the sensor has higher sensitivity for particle recognition. The relationship between the sensor signal pulse width and particle velocity is a power function. The sensor can distinguish and detect various metal and non-metallic particulate pollutants and recognize 78 μm iron particles and 165 μm copper particles. This study provides a new method for wear condition monitoring.

A sensor containing high permeability material for mechanical wear particle detection

A sensor containing high permeability material for mechanical wear particle detection

Novelty Sensor for Detection of Wear Particles in Oil Using Integrated Microwave Metamaterial Resonators With Neodymium Magnets

Sensors for detecting wear particles in fluid lubricants have been widely studied owing to their ability to diagnose mechanical machine conditions and to help in preventing mechanical failures caused by wear particles or contamination particles. The proposed novelty sensor was fabricated using photolithography to make the complementary split ring resonator (CSSR) on PCB and it utilized neodymium magnets at the center of the CSRR structure to attract ferrous wear particles. The proposed sensor was tested, and we found that it is highly sensitive and could detect 50 ppm of ferrous wear particles larger than <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$10 \mu \text{m}$ </tex-math></inline-formula> dissolved in mineral white oil ISO VG 15. The sensor of this article has a low-cost and can be used in different situations to measure the ferrous wear particles while monitoring the real-time mechanical equipment wear, as it can be easily integrated into them. The sensor offers good accuracy, and is insensitive to fluid transparency and bubble formation, unlike other sensors for detection of wear particles. Therefore, the proposed novelty sensor of this work is in line with Industry 4.0, which requires many low-cost sensors for continuous monitoring to apply predictive maintenance techniques for extending the lifespan of infrastructures.

Online Wear Particle Detection Sensors for Wear Monitoring of Mechanical Equipment—A Review

Online wear particle detectors have been widely studied due to their ability to monitor the wear state of machines, and to help to prevent serious mechanical failures caused by excessive wear of components. This paper presents a review of the state-of-the-art wear debris detectors based on different principles. That mainly includes optical/imaging particle detectors, electrical particle detectors, ultrasonic particle detectors and magnetic debris detection sensors. Meanwhile, the characteristics and the performance (sensitivity, maximum flow rate, and the detectable information) of each type of sensor are detailed discussed. In conclusion, the optical/imaging and magnetic debris detectors have become promising technologies in the field of wear monitoring. Meanwhile, that the rapid extraction algorithm and three-dimensional reconstruction method of wear debris for optical/imaging debris detectors, and improving the sensitivity and detectability of magnetic debris detectors are the two important research contents in the field. Besides that, it can be predicted that the integrated wear monitoring system that can collect multiple wear-information in-situ, and the intelligent wear particle detector that have the ability of wear monitoring, wear evaluation and fault warning are the main research directions in the future.

Semantic segmentation of ferrography images for automatic wear particle analysis

Automatic wear particle detection and classification has remained a high priority research area for wear condition monitoring and failure analysis. In this study, a deep convolutional neural network (DCNN) with three modules, namely, an encoder, atrous spatial pyramid pooling (ASPP), and a decoder, is constructed. Instead of using handcrafted features, the DCNN can automatically learn features through a layer-wise representation and realize semantic segmentation, i.e., segmentation and identification concurrently, of five types of wear particles in ferrograph images using end-to-end processing. Experimental results show that the DCNN achieves 82.5% accuracy. This proposed method unifies the segmentation, classification, and edge location of the wear particles into a single model, avoids the accumulation and transmission of errors caused by numerous steps applied in a traditional linear process, and improves the efficiency and accuracy of the wear particle analysis.

A Wear Particle Sensor Using Multiple Inductive Coils Under a Toroidal Magnetic Field

Inductive wear particle sensors have been widely studied due to their ability to monitor the wear status of equipment in real time. To detect wear particles more precisely, a wear particle sensor based on multiple inductive coils under a toroidal magnetic field is proposed in this paper. With inductive coils located near a stronger magnetic field, the sensor can not only ensure the flow rate, but also enhance wear particle detection accuracy. The proportional relationship between the signal and the wear particle velocity and the excitation current is discussed in this paper. The crosstalk phenomenon among different inductive coils was also investigated by single particle experiments and lubricating oil dynamic experiments. In the dynamic oil experiment, the sensor detected $13~\mu \text{m}$ particles under a flow rate of 570 mL/min, which satisfied most industrial online monitoring demands.

WP-DRnet: A novel wear particle detection and recognition network for automatic ferrograph image analysis

Ferrography plays an important role in wear analysis for machine condition monitoring, in which effective and efficient wear particle analysis is regarded as a crucial pre-requisite. An automatic wear particle detection and classification process is developed here using a cascade of two convolutional neural networks and a support vector machine (SVM) classifier. The neural networks are used for particle detection and recognition while particle classification is conducted in the SVM. This structure ensures that the computation expense is reduced and the accuracy is improved. The proposed network is verified using a large number of ferrograph images. Results show that high classification accuracies are obtained. Furthermore, the proposed approach can be further developed and applied in online machine condition monitoring applications.

Design of electromagnetic sensor for metal wear particle detection in oil

Wear is one of the common faults in the mechanical transmission system. Detection of wear particles in lubricating oil can detect and evaluate the wear degree of mechanical equipment. The electromagnetic sensor for metal wear particle detection can be connected in series with the oil circuit, which has the advantages of full liquid flow, simple structure, distinguishing metal properties, and not being influenced by bubbles and vibration. In view of the shortcomings of the existing electromagnetic sensor for wear particle detection, the detection principle and the induction electromotive force model were studied, and the magnetic field change of the metal particles through three coil differential solenoids was analysed. The optimisation design method was proposed here, which was about the related parameters, such as the wire diameter, the coil turns, the coil widths and so on. The experimental analysis of the sensor was carried out based on detection hardware system that could provide the phase-locked amplification function. The 10-mm inner diameter sensor could detect the minimum 100-μm ferromagnetic metal abrasive particles after optimisation. The developed sensor could provide data information for the abrasive particle detection, fault diagnosis and early warning of wear degree of mechanical equipment.

Analysis of Spatial Sensitivity Based on Electrostatic Monitoring Technique in Oil-Lubricated System

In this paper, for the detection of wear particles in lubricating oil roads, by an-alyzing the particle belt motor, a mathematical model of the oil circuit electro-static sensor considering the dielectric constant is proposed. At the same time, the concept of spatial sensitivity and visual field is introduced. The influencing factors of the position of the charged abrasive particles, the axial length of the sensor and the radial radius were studied respectively. The influence of each variable on the sensitivity was obtained, and the accuracy of the mathematical model was also verified. The simulation results show that reducing the radial radius of the sensor can effectively increase the spatial sensitivity of the electrostatic sensor; the larger the axial length L of the electrode, the higher the sensitivity, and the more uniform the cross-sectional sensitive field distribution; in the axial and radial directions, the larger the shaft diameter ratio, the more sensitive the sensor is and the more sensitive it is.

A non-reference evaluation method for edge detection of wear particles in ferrograph images

Abstract Edges are one of the most important features of wear particles in a ferrograph image and are widely used to extract parameters, recognize types of wear particles, and assist in the identification of the wear mode and severity. Edge detection is a critical step in ferrograph image processing and analysis. Till date, there has been no single algorithm that guarantees the production of good quality edges in ferrograph images for a variety of applications. Therefore, it is desirable to have a reliable evaluation method for measuring the performance of various edge detection algorithms and for aiding in the selection of the optimal parameter and algorithm for ferrographic applications. In this paper, a new non-reference method for the objective evaluation of wear particle edge detection is proposed. In this method, a comprehensive index of edge evaluation is composed of three components, i.e., the reconstruction based similarity sub-index between the original image and the reconstructed image, the confidence degree sub-index used to show the true or false degree of the edge pixels, and the edge form sub-index that is used to determine the direction consistency and width uniformity of the edges. Two experiments are performed to illustrate the validity of the proposed method. First, this method is used to select the best parameters for an edge detection algorithm, and it is then used to compare the results obtained using various edge detection algorithms and determine the best algorithm. Experimental results of various real ferrograph images verify the effectiveness of the proposed method.

Photonic Low Cost Micro-Sensor for in-Line Wear Particle Detection in Flowing Lube Oils.

The presence of microscopic particles in suspension in industrial fluids is often an early warning of latent or imminent failures in the equipment or processes where they are being used. This manuscript describes work undertaken to integrate different photonic principles with a micro- mechanical fluidic structure and an embedded processor to develop a fully autonomous wear debris sensor for in-line monitoring of industrial fluids. Lens-less microscopy, stroboscopic illumination, a CMOS imager and embedded machine vision technologies have been merged to develop a sensor solution that is able to detect and quantify the number and size of micrometric particles suspended in a continuous flow of a fluid. A laboratory test-bench has been arranged for setting up the configuration of the optical components targeting a static oil sample and then a sensor prototype has been developed for migrating the measurement principles to real conditions in terms of operating pressure and flow rate of the oil. Imaging performance is quantified using micro calibrated samples, as well as by measuring real used lubricated oils. Sampling a large fluid volume with a decent 2D spatial resolution, this photonic micro sensor offers a powerful tool at very low cost and compacted size for in-line wear debris monitoring.

Characterising the friction and wear behaviour of lubricated metal‐metal pairings with an optical online particle detection system

AbstractIn this study, a simple ball‐on‐prism test set‐up is combined with an optical online wear particle detection system. The set‐up is used to study the friction and wear characteristics of some exemplary oil lubricated metal‐metal pairings. In addition to the time‐resolved evolutions of the coefficient of friction and the volumetric wear, the number of wear particles, the particle sizes and the particle shapes are obtained during the test by using an online particle detection system. The number of small (&lt;5 μm) wear particles increases steadily during the tests, while the number of larger (&gt;10 μm) particles only increases during the running‐in period and remains almost constant afterwards. Simultaneous drops in the curves for the coefficient of friction and the trend curves for the number of wear particles indicate a possible interrelation. The individual curves for the different particle sizes suggest that mainly small particles (&lt;5 μm) contribute to that relation. © 2017 John Wiley &amp; Sons, Ltd.

Analiza tkiva okrog omajanih umetnih kolčnih debel iz poliacetala z uporabo nuklearnih metod

IntroductionThe influence of polyacetal wear particles on aseptic loosening of non-cemented isoelastic femoral stems with polyacetal coating remains unclear. The aim of our study was to use nuclear methods to prove the presence of polyacetal wear particles, to determine their morphology and to check their distribution in the tissues around loosend hip prosthetic components.MethodsTissue samples obtained during retrieval of 4 aseptic loosened primary hip prostheses with isoelastic stems made of polyacetal were subjected to nuclear analyses. Proton microbrobe method (mikro-PIXE) was used to prove the presence of polyacetal wear particles and to check for their morphology by detection of barium, which is molecularly in BaSO4 embedded in polyacetal. Thermal neutron activation was used to determine distribution pattern of polyacetal wear in the peri-prosthetic tissues.ResultsAgainst expectations, polyacetal wear particles were found rather rare, larger than 100 µm and present in pseudo-membrane samples around the loosened stem, but virtually absent in tissues away from their origin. Concentration of BaSO4 in polyacetal wear particles in pseudo-membrane samples was similar to the one in polyacetal coating (conc. Ba = 14217 µg/g and 14800 µg/g, respectively).ConclusionAccording to the results, the primary cause of the loosening of the isoelastic stems with polyacetal coating is most probably mechanical restlessness, which is responsible for local production of the large polyacetal wear particles responsible to accelerate the process of loosening. PIXE method and thermal neutron activation are sensitive quantitative nuclear methods suitable for direct or indirect detection of wear particles in the tissue around loosened prostheses and to determine morphology of wear particles and their distribution in the tissues.

Wear rate evaluation of a novel polycarbonate-urethane cushion form bearing for artificial hip joints

There is growing interest in the use of compliant materials as an alternative to hard bearing materials such as polyethylene, metal and ceramics in artificial joints. Cushion form bearings based on polycarbonate-urethane (PCU) mimic the natural synovial joint more closely by promoting fluid-film lubrication. In the current study, we used a physiological simulator to evaluate the wear characteristics of a compliant PCU acetabular buffer, coupled against a cobalt-chrome femoral head. The wear rate was evaluated over 8 million cycles gravimetrically, as well as by wear particle isolation using filtration and bio-ferrography (BF). The gravimetric and BF methods showed a wear rate of 9.9-12.5mg per million cycles, whereas filtration resulted in a lower wear rate of 5.8mg per million cycles. Bio-ferrography was proven to be an effective method for the determination of wear characteristics of the PCU acetabular buffer. Specifically, it was found to be more sensitive towards the detection of wear particles compared to the conventional filtration method, and less prone to environmental fluctuations than the gravimetric method. PCU demonstrated a low particle generation rate (1-5×10⁶ particles per million cycles), with the majority (96.6%) of wear particle mass lying above the biologically active range, 0.2-10μm. Thus, PCU offers a substantial advantage over traditional bearing materials, not only in its low wear rate, but also in its osteolytic potential.

Effective Condition Monitoring Methods for Diesel Main Bearing -High Frequency Vibration Measurement and Wear Particle Detection

Extension on overhaul inspection period of the main bearing of main diesel engine is highly desired by carrying out its condition monitoring. The effective condition monitoring method is required instead of the temperature method which rises out the warning alarm after bearings appeared in critical condition. It was found also for plane bearing that the contact between the shaft and bearing may occur locally even in the oil film lubrication state. Result from this study, the high frequency vibration measurement and wear particle detection in lubrication oil were suggested to be the effective condition monitoring methods.