Search results

The purpose of this paper is to propose a method of qualitative ferrographic analysis by quantitative parameters of wear debris characteristics.

The amount of the wear debris needed for analysis on the ferrogram made by rotary ferrograph is discussed based on the theory of debris group. Quantitative parameters are constituted to express the characteristics of wear debris group, and correlation coefficients are employed to establish the relationship between wear debris and wear condition. The reliability of the method was verified by wear test experiments and ferrographic analysis.

The wear condition of machines should be determined by studying all the debris together as a group rather than by focusing on individual debris. In the proposed method, the qualitative analysis result is obtained by synthetic analysis of quantitative parameters of wear debris characteristics using a computer program, which makes the judgment of the wear system condition more objective and precise.

In the procedure of wear condition monitoring by the proposed method, because the weight factors and correlation coefficients introduced in this paper are determined according to the experiences deriving from practice among mining machinery, further rectifications may be needed if they are applied to other industrial field.

The paper illustrates a more objective and precise ferrographic analysis method for wear condition monitoring.

The dependency on human expertise for analysis and interpretation is the main reason for wear debris analysis not being used in industry to its full potential and becoming one of the most powerful machine condition monitoring strategies. The dependency on human expertise makes the interpretation and result subjective in nature, costly and time consuming. The purpose of this paper is to review work being done to develop an automatic, reliable and objective wear particle classification system as a solution to the above problem. At the same time it also aims to discuss some common off line test methods being practiced for wear debris analysis.

Computer image analysis is a solution for some of the problems associated with the conventional techniques. First it is tried to efficiently describe the characteristics of computer images of different types of wear debris using a few numerical parameters. Then using some Artificial Intelligence tools, the wear particle classification system can be developed.

Many shape, size and surface parameters are discussed in the paper. Out of these, nine numerical parameters are selected to describe and distinguish six common type of wear debris. Once the type of debris is identified, the mode of wear and hence the machine condition can be assessed.

The present process of fault and condition monitoring of an equipment by wear debris analysis involves human judgment of debris formations. A set‐up standard for comparison of debris will enable the maintenance team to diagnose faults in a comparatively better way.

The aim of this paper is to discuss the difficulties in identifying wear particles and finding out the exact health of equipment, which, due to its subjective nature, is influenced by human errors. An objective method with certain standards for classification of wear particles compatible with an artificial intelligence system will yield some flawless results of wear debris analysis, which has not been attempted in the past as per available literature.

The purpose of this paper is to identify smaller wear particles and improve the calculation speed, identify more abrasive particles and promote industrial applications.

This paper studies a new intelligent recognition method for equipment wear debris based on the YOLO V5S model released in June 2020. Nearly 800 ferrography pictures, 23 types of wear debris, about 5,000 wear debris were used to train and test the model. The new lightweight approach of wear debris recognition can be implemented in rapidly and automatically and also provide for the recognition of wear debris in the field of online wear monitoring.

An intelligent recognition method of wear debris in ferrography image based on the YOLO V5S model was designed. After the training, the GIoU values of the model converged steadily at about 0.02. The overall precision rate and recall rate reached 0.4 and 0.5, respectively. The overall MAP value of each type of wear debris was 40.5, which was close to the official recognition level of YOLO V5S in the MS COCO competition. The practicality of the model was approved. The intelligent recognition method of wear debris based on the YOLO V5S model can effectively reduce the sensitivity of wear debris size. It also has a good recognition effect on wear debris in different sizes and different scales. Compared with YOLOV. YOLOV, Mask R-CNN and other algorithms%2C, the intelligent recognition method based on the YOLO V5S model, have shown their own advantages in terms of the recognition effect of wear debris%2C the operation speed and the size of weight files. It also provides a new function for implementing accurate recognition of wear debris images collected by online and independent ferrography analysis devices.

To the best of the authors’ knowledge, the intelligent identification of wear debris based on the YOLO V5S network is proposed for the first time, and a large number of wear debris images are verified and applied.

The purpose of this paper is to show the development of an image processing-based portable equipment for an automatic wear debris analysis. It can analyze both the qualitative and quantitative features of machine wear debris: size, quantity, size distribution, shape, surface texture and material composition via color.

It comprises hardware and software components which can take debris in near real-time from a machine oil sump and process it for features diagnosis. This processing provides the information of the basic features on the user screen which can further be used for machine component health diagnosis.

The developed system has the capacity to replace the existing off-line methods due to its cost effectiveness and simplicity in operation. The system is able to analyze debris basic quantitative and qualitative features greater than 50 micron and less than 300 micron.

Wear debris basic features analysis tool is developed and discussed. The portable and near real-time analysis offered by the discussed work can be more technically effective as compared to the existing off-line and online techniques.

In real working condition, signal is highly disturbed and even drowned by noise, which extremely interferes in detecting results. Therefore, this paper aims to provide an effective de-noising method for the debris particle in lubricant so that the ultrasonic technique can be applied to the online debris particle detection.

For completing the online ultrasonic monitoring of oil wear debris, the research is made on some selected wear debris signals. It applies morphology component analysis (MCA) theory to de-noise signals. To overcome the potential weakness of MCA threshold process, it proposes fuzzy morphology component analysis (FMCA) by fuzzy threshold function.

According to simulated and experimental results, it eliminates most of the wear debris signal noises by using FMCA through the signal comparison. According to the comparison of simulation evaluation index, it has highest signal noise ratio, smallest root mean square error and largest similarity factor.

The rapid movement of the debris particles, as well as the lubricant temperature, may influence the measuring signals. Researchers are encouraged to solve these problems further.

This paper includes implications for the improvement in the online debris detection and the development of the ultrasonic technique applied in online debris detection.

This paper provides a promising way of applying the MCA theory to de-noise signals. To avoid the potential weakness of the MCA threshold process, it proposes FMCA through fuzzy threshold function. The FMCA method has great obvious advantage in de-noising wear debris signals. It lays the foundation for online ultrasonic monitoring of lubrication wear debris.

It is a challenging task to analysis oxide wear particles when they are stuck together with other types of wear particles in complex ferrography images. Hence, this paper aims to propose a method of ferrography image segmentation to analysis oxide wear debris in complex ferrography images.

First, ferrography images are segmented with watershed transform. Then, two region merging rules are proposed to improve the initial segmentation results. Finally, the features of each particle are extracted to detect and assess the oxide wear particles.

The results show that the proposed method outperforms other methods of ferrography image segmentation, and the overlapping wear particles in complex ferrography images can be well separated. Moreover, the features of each separated wear particles can be easily extracted to analysis the oxide wear particles.

The proposed method provides a useful approach for the automatic detection and assessment of oxide wear particles in complex ferrography images.

The colours, edges and position information of wear debris are considered in the proposed method to improve the segmentation result. Moreover, the proposed method can not only detect oxide wear particles in ferrography images but also evaluate oxide wear severity in ferrography images.

This study provides a unique integrated diagnosis system to investigate the causes of low productivity, profitability, machinery health conditions and wear severity of medium-size biscuit industry assets in Taiz, Yemen.

The evaluation is based on an integrating of the overall equipment effectiveness (OEE) and oil-based maintenance (OBM) approaches. The data are collected using the company's operational records, interviews and observations, while the used lubricating oil samples are also collected from production lines' machineries. Scanning electron microscope (SEM) is used to study the wear debris particle features and wear mechanism. Different other analysis tools such as fishbone, 5 whys and Pareto charts are also used to investigate the root causes and plausible recovery solutions of machinery failures.

This study demonstrated that a large proportion of machinery failures and production loss are of management concerns. Also, this study inferred that the analysis of wear debris is unique and informative for determining machinery wear severity and useful life. Finally, the current conditions of production lines are clarified and suggestions to use a mixed preventive/predictive maintenance management approach are also elucidated.

This work implemented an integrated OEE/OBM diagnostic maintenance system to investigate the root causes of low productivity and machine failures in real production lines and suggested robust decisions on the maintenance duties.

This paper aims to investigate the effect of misalignment on wear of spur gears and on oil degradation using online sensors.

The misalignment effect on gears is created through a self-alignment bearing, and is measured using laser alignment system. Several online sensors such as Fe-concentration sensor, moisture sensor, oil condition sensor, oil temperature sensor and metallic particle sensor are installed in the gear test rig to monitor lubricant quality and wear debris in real time to assess gearbox failure.

Offset and angular misalignments are detected in both vertical and horizontal planes. The failure of misaligned gear is observed at both the ends and on both the surfaces of the gear teeth. Larger-size ferrous and non-ferrous particles are traced by metallic particle sensor due to gear and seal wear caused by misalignment. Scanning electron microscope (SEM) images examine chuck, spherical and flat platelet particles, and confirm the presence of fatigue (pitting) and adhesion (scuffing) wear mechanism. Energy-dispersive X-ray spectroscopy analysis of SEM particles traces carbon (C) and iron (Fe) elements due to gear failure.

Gear misalignment is one of the major causes of gearbox failure and the lubricant analysis is as important as wear debris analysis. A reliable online gearbox condition monitoring system is developed by integrating wear and oil analyses for misaligned spur gear pair in contact.

Reports on the study of wear debris in the early 1980s organized by the Institution of Mechanical Engineers. Describes the use of Ferrography, spectrographic oil analysis, X‐ray fluorescence, inductive, magnetic and ultrasonic methods of debris detection with some examples of wear limits.

The probability distribution of major length and aspect ratio (major length/minor length) of wear debris collected from gear oil used in planetary gear drive were analysed and modelled. The paper aims to find an appropriate probability distribution model to forecast the kind of wear particles at different running hour of the machine.

Used gear oil of the planetary gear box of a slab caster was drained out and charged with a fresh oil of grade (EP-460). Six chronological oil samples were collected at different time interval between 480 and 1,992 h of machine running. The oil samples were filtered to separate wear particles, and microscopic study of wear debris was carried out at 100X magnification. Statistical modelling of wear debris distribution was done using Weibull and exponential probability distribution model. A comparison was studied among actual, Weibull and exponential probability distribution of major length and aspect ratio of wear particles.

Distribution of major length of wear particle was found to be closer to the exponential probability density function, whereas Weibull probability density function fitted better to distribution of aspect ratio of wear particle.

The potential of the developed model can be used to analyse the distribution of major length and aspect ratio of wear debris present in planetary gear box of slab caster machine.