Deep learning-based fault diagnosis systems for prognostics and health management of mechanical systems is an active research topic. Notably, the absence and class imbalance of fault data (insufficient fault data compared to normal data) have been shown to cause many challenges in developing fault diagnosis systems for the manufacturing fields. Therefore, this paper presents case studies using deep learning algorithms in the absence or class imbalance of fault data. Auto-encoder-based anomaly detection method, which can be used when fault data is absent, was applied to diagnose faults in a robotic spot welding process. The anomaly detection threshold was set based on the reconstruction error of trained normal data and the confidence level of the distribution of normal data. The anomaly detection performance of the auto-encoder was verified using non-trained normal data and three sets of fault data through the threshold. As a case study for insufficient fault data, synthetic data was generated based on cGAN and applied to diagnose fault of bearing. Using the imbalanced dataset to generate synthetic fault data and to reduce the imbalance ratio, it was confirmed that the accuracy of the synthetic data generation-based 2DCNN fault diagnosis model was improved.
Citations
Citations to this article as recorded by
Distribution of Force Applied to a Lateral Damper during EMU Operation Hyun Moo Hur, Kyung Ho Moon, Seong Kwang Hong Journal of the Korean Society for Precision Engineering.2024; 41(9): 673. CrossRef
This paper discusses flow characteristics of nanofluid minimum quantity lubrication (MQL) in the milling process of a titanium alloy by usingnumerical analysis. A mist of nanofluids including nanodiamond and hexagonal boron nitride (hBN) particles is sprayed into a tool-workpiece interface with conditions varying by spray angle and flow rate. The milling. Are experimentally measured and minimized by the determined optimal spray angle and flow rate. The subsequent numerical analysis based on a computational fluid dynamics (CFD) approach is conducted to calculate the penetration ratios of the nanofluid droplets into a tool. At the experimentally obtained optimal spray angle and flow rate of the nanofluids’ mist, the calculated ratio of penetration is highest and, therefore, the optimal spray conditions of the nanofluids are numerically validated.
This paper presents a numerical study on the thermal characteristics of a milling process of titanium alloy with nanofluid minimum-quantity lubrication (MQL). The computational fluid dynamics (CFD) approach is introduced for establishing the numerical model for the nanofluid MQL milling process, and estimated temperatures for pure MQL and for nanofluid MQL using both hexagonal boron nitride (hBN) and nanodiamond particles are compared with the temperatures measured by thermocouples in the titanium alloy workpiece. The estimated workpiece temperatures are similar to experimental ones, and the model is validated.
Recently, titanium alloys have been widely used in aerospace, biomedical engineering, and military industries due to their high strength to weight ratio and corrosion resistance. However, it is well known that titanium alloys are difficult-to-cut materials because of a poor machinability characteristic caused by low thermal conductivity, chemical reactivity with all tool materials at high temperature, and high hardness. To improve the machinability of titanium alloys, cryogenic cooling with LN2 (Liquid Nitrogen) and nanofluid MQL (Minimum Quantity Lubrication) technologies have been studied while turning a Ti-6Al-4V alloy. For the analysis of turning process characteristics, the cutting force, the coefficient of friction, and the surface roughness are measured and analyzed according to varying lubrication and cooling conditions. The experimental results show that combined cryogenic cooling and nanofluid MQL significantly reduces the cutting forces, coefficients of friction and surface roughness when compared to wet condition during the turning process of Ti-6Al-4V.
Citations
Citations to this article as recorded by
Current research trends in coolant application for machining Ti-6Al-4V alloy: a state-of-the-art review Prianka B. Zaman, N. R. Dhar Advances in Materials and Processing Technologies.2024; : 1. CrossRef
Comprehensive analysis of cutting temperature, tool wear, surface integrity and tribological properties in sustainable milling of Ti6Al4V alloy: LN2, nanofluid and hybrid machining Emine Şirin, Çağrı Vakkas Yıldırım, Şenol Şirin, Turgay Kıvak, Murat Sarıkaya Journal of Manufacturing Processes.2024; 131: 1360. CrossRef
Recent advancements in nano-lubrication strategies for machining processes considering their health and environmental impacts Kishan Zadafiya, Prassan Shah, Alborz Shokrani, Navneet Khanna Journal of Manufacturing Processes.2021; 68: 481. CrossRef
Determination of Flow Stress and Cutting Force Prediction of Ti-6Al-4V Material for 3D Printer using S-K Constitutive Equation Dae-Gyoun Park, Tae-Ho Kim, Eon-Chan Jeon Journal of the Korean Society of Manufacturing Process Engineers.2018; 17(6): 68. CrossRef
Friction and Wear Characteristics of Surface-Modified Titanium Alloy for Metal-on-Metal Hip Joint Bearing Hyeon-hwa Lee, Sungcheul Lee, Jong-Kweon Park, Minyang Yang International Journal of Precision Engineering and Manufacturing.2018; 19(6): 917. CrossRef
CP (Commercially Pure) titanium has been widely used in various industries such as in energy plants and bio-materials because of an excellent corrosion resistance and its non-toxicity to the human body. But there are limitations for usage as structural materials due to low strength. The tensile properties of CP titanium could be improved by microstructure refinement such as in a SPD (Severe Plastic Deformation) process. However, high strengthening of CP titanium wire is impossible by SPD processes like ECAP (Equal Channel Angular Pressing), HPT (High-Pressure Torsion), and the ARB (Accumulative Roll Bonding) process. The study purposes are to increase the strength of CP titanium wire by optimization of the cold drawing process and the harmonization with mechanical properties by heat treatments for the next forming process. The optimization process was investigated with regard to the design of drawing dies and the reduction ratio of cross sections. The elongations of high strength CP titanium were controlled by heat treatment.
In this paper, a real-time condition diagnosis system for the lens injection molding process is developed through the use of LabVIEW. The built-in-sensor (BIS) mold, which has pressure and temperature sensors in their cavities, is used to capture real-time signals. The measured pressure and temperature signals are processed to obtain features such as maximum cavity pressure, holding pressure and maximum temperature by the feature extraction algorithm. Using those features, an injection molding condition diagnosis model is established based on a response surface methodology (RSM). In the real-time system using LabVIEW, the front panels of the data loading and setting, feature extraction and condition diagnosis are realized. The developed system is applied in a real industrial site, and a series of injection molding experiments are conducted. Experimental results show that the average real-time condition diagnosis rate is 96%, and applicability and validity of the developed real-time system are verified.
This study investigates the effects of the size of copper sheets on the plastic deformation behavior in a microscale deep drawing process. Tensile tests are conducted on the copper sheets to study the flow stress of the materials with different grain sizes before carrying out the microscale deep drawing experiments. After the tensile tests, a novel desktop-sized microscale deep drawing system is used to perform the microscale deep drawing process. A series of microscale deep drawing experiments are subsequently performed, and the experimental results indicate that an increase in the grain size results in the reduction of the deformation load of the copper sheets due to the effects of the surface grain. The results also show that the blank holder gap improves both the formability of copper sheets and the material flow.
In this paper, a new condition diagnosis algorithm for the lens injection molding process using various features extracted from cavity pressure, nozzle pressure and screw position signals is developed with the aid of probability neural network (PNN) method. A new feature extraction method is developed for identifying five (5), seven (7) and two (2) critical features from cavity pressure, nozzle pressure and screw position signals, respectively. The node energies extracted from cavity and nozzle pressure signals are also considered based on wavelet packet decomposition (WPD). The PNN method is introduced to build the condition diagnosis model by considering the extracted features and node energies. A series of the lens injection molding experiments are conducted to validate the model, and it is demonstrated that the proposed condition diagnosis model is useful with high diagnosis accuracy.
As demands on micro-products increase significantly with raising functional integration and increasing complexity, microfoming attracts a lot of attention in the manufacture of microproducts. Since the conventional big forming systems are not adequate to achieve sufficient tolerances of micro-scale parts, it is necessary to reduce the scale of the forming equipment and devices. In addition, understandings on the size effects, which exist in the material behavior and process characterization of microforming processes, need to be expanded. In this study, a miniaturized forming system based on the ball screw and servo motor actuator was developed for the efficient micro-parts production. In addition, tensile tests and cylindrical upsetting experiments were performed to evaluate the performance of the microforming system and to investigate the flow stress and friction size effects in microforming processes.
This study investigates the non-traditional manufacturing process of dry wire electrical discharge machining (EDM) in which liquid dielectric is replaced by a gaseous medium. Wire EDM experiments of thin workpieces were conducted both in wet and dry EDM conditions to examine the effects of spark cycle (T), spark on-time (Ton), thickness of workpieces, and work material on machining performance. The material removal rate (MRR) in the dry wire EDM case was much lower than that in the wet wire EDM case. In addition, the thickness of workpiece and workmaterial were found to be critical factors influencing the MRR for dry EDM process. The relative ratios of spark, arc and short circuit were also calculated and compared to examine the effectiveness of processes of dry and wet wire EDM.
This paper addresses the design and fabrication of desktop die-sinking dry electrical discharge machining (EDM) system and its experimental performance analysis. The developed desktop dry EDM machine has the horizontal configuration with the size of 300×200×260㎜. The experimental performance analysis is conducted to investigate the effects of EDM conditions and dielectric gas temperature on the surface roughness of EDMed slots and number of EDM sparks. The experimental results demonstrate that low feed rate and large electrode displacement are good for better surface roughness and more number of EDM sparks. In addition, low temperature of dielectric gas results in better surface roughness.
This paper presents the development of a new inchworm actuation system using the shearing deformation of the piezoelectric actuators. In this new actuation system, piezoelectric shearing/expanding actuators, an inertial mass and an advanced preload system are configured innovatively to generate the motion of an inertial mass. There are two modes in the new actuation system: (1) stick mode, and (2) clamp mode. In stick mode, the deformation of the piezoelectric shearing actuators drives an inertial mass by means of the friction force at their contact interface. On the other hand, in clamp mode, the piezoelectric expanding actuators provide the gripping force to an inertial mass and, as a result, eliminate its backward motion following the rapid backward deformation of the piezoelectric shearing actuators. To investigate the feasibility of the proposed new actuation system, the experimental system is built up, and the static performance evaluation and dynamic analysis are conducted. The open-loop performance of the linear motion of the proposed new actuation system is evaluated. In dynamic analysis, the mathematical model for the contact interface is established based on the LuGre friction model and the equivalent parameters are identified.
First
Prev
