This study analyzed acoustic emission (AE) signals generated during ultrasonic machining of SiC cathodes and evaluated classification performances of various machine learning models. AE data were collected in both waveform and hit formats, enabling signal characterization through statistical analysis and frequency domain examination. Various machine learning models, including XGBoost, KNN, Logistic Regression, SVM, and MLP, were applied to classify machining states. Results showed that XGBoost achieved the highest classification accuracy across all sensor positions, particularly at the upper part of the worktable with an accuracy of 98.35%. Additional experiments confirmed the consistency of these findings, highlighting the influence of sensor placement on classification performance. This study demonstrates the feasibility of monitoring AE-based machining state using machine learning and emphasizes the importance of sensor placement and signal analysis in improving classification accuracy. Future research should incorporate defect data and deep learning approaches to further enhance classification performance and process monitoring capabilities.
Estimating energy expenditure is essential in monitoring the intensity of physical activity and health status. Energy expenditure can be estimated based on wearable sensors such as inertial measurement unit (IMU). While a variety of methods have been developed to estimate energy expenditure during day-to-day activities, their performances have not been thoroughly evaluated under walking conditions according to various speeds and inclines. This study investigated IMU-based neural network models for energy expenditure estimation under various walking conditions and comparatively analyzed their performances in terms of sensor attachment locations and training/testing datasets. In this study, two neural network models were selected based on a previous study (Slade et al., 2019): (M1) a multilayer perceptron using sensor signals during each gait cycle, and (M2) a recurrent neural network using sensor signal sequences of a fixed window size. The results revealed the following: (i) the performance of the foot attachment model was the best among the five sensor attachment locations (0.89 W/kg for M1 and 1.14 W/kg for M2); and (ii) although the performance of M1 was superior to that of M2, M1 requires accurate gait detection for data segmentation by each stride, which hinders the usefulness of M2.
Citations
Citations to this article as recorded by
Development of a Novel Ventilation Estimation Model Based on Convolutional Neural Network (CNN) Jeongyeon Chu, Jaehyon Baik, Kangsu Jeong, Seungwon Jung, Youngjin Park, Hosu Lee Journal of Korea Robotics Society.2025; 20(1): 138. CrossRef
Tire-related crashes account for a large proportion of all types of car accidents. The causes of tire-related accidents are inappropriate tire temperature, pressure, and wear. Although temperature and pressure can be monitored easily with TPMS, there exists no system to monitor tire wear regularly. This paper proposes a system that can estimate tire wear using a 3-axis accelerometer attached to the tread inside the tire. This system utilizes axial acceleration, extracts feature from data acquired with the accelerometer and estimates tire wear by feature classification using machine learning. In particular, the proposed tire wear estimation method is designed to estimate tread depth in four types (7, 5.6, 4.2, and 1.4 mm) at speeds of 40, 50, and 60 kmph. Based on the data obtained during several runs on a test track, it has been found that this system can estimate the tread depth with reasonable accuracy.
Citations
Citations to this article as recorded by
A Study on Wheel Member Condition Recognition Using 1D–CNN Jin-Han Lee, Jun-Hee Lee, Chang-Jae Lee, Seung-Lok Lee, Jin-Pyung Kim, Jae-Hoon Jeong Sensors.2023; 23(23): 9501. CrossRef
As the digitization of the manufacturing process is accelerating, various data-driven approaches using machine learning are being developed in chemical mechanical polishing (CMP). For a more accurate prediction in contact-based CMP, it is necessary to consider the real-time changing pad surface roughness during polishing. Changes in pad surface roughness result in non-uniformity of the real contact pressure and friction applied to the wafer, which are the main causes of material removal rate variation. In this paper, we predicted the material removal rate based on pressure and surface roughness using a deep neural network (DNN). Reduced peak height (Rpk) and real contact area (RCA) were chosen as the key parameters indicative of the surface roughness of the pad, and 220 data were collected along with the process pressure. The collected data were normalized and separated in a 3 : 1 : 1 ratio to improve the predictive performance of the DNN model. The hyperparameters of the DNN model were optimized through random search techniques and 5 cross-validations. The optimized DNN model predicted the material removal rate with high accuracy in ex-situ CMP. This study is expected to be utilized in data-driven machine learning decision making for cyber-physical CMP systems in the future.
Citations
Citations to this article as recorded by
Precision Engineering and Intelligent Technologies for Predictable CMP Somin Shin, Hyun Jun Ryu, Sanha Kim, Haedo Jeong, Hyunseop Lee International Journal of Precision Engineering and Manufacturing.2025; 26(9): 2121. CrossRef
Prediction of Normalized Material Removal Rate Profile Based on Deep Neural Network in Five-Zone Carrier Head CMP System Yonsang Cho, Myeongjun Kim, Munyoung Hong, Joocheol Han, Hong Jin Kim, Hyunki Kim, Hyunseop Lee International Journal of Precision Engineering and Manufacturing-Green Technology.2025; 12(3): 869. CrossRef
The collaboration of robots and humans sharing workspace, can increase productivity and reduce production costs. However, occupational accidents resulting in injuries can increase, by removing the physical safety around the robot, and allowing the human to enter the workspace of the robot. In preventing occupational accidents, studies on recognizing humans, by installing various sensors around the robot and responding to humans, have been proposed. Using the LiDAR (Light Detection and Ranging) sensor, a wider range can be measured simultaneously, which has advantages in that the LiDAR sensor is less impacted by the brightness of light, and so on. This paper proposes a simple and fast method to recognize humans, and estimate the path of humans using a single stationary 360° LiDAR sensor. The moving object is extracted from background using the occupied grid map method, from the data measured by the sensor. From the extracted data, a human recognition model is created using CNN machine learning method, and the hyper-parameters of the model are set, using a grid search method to increase accuracy. The path of recognized human is estimated and tracked by the extended Kalman filter.
The tower crane is widely used in construction and transportation engineering. To improve working efficiency and safety, input shaping methods have been applied. Input shaping is a method of reducing residual vibration of flexible systems by convolving a sequence of impulses with unit step command. However, input shaping is based on the linear system theory in which its control performances are degraded, in case of nonlinearity and unmatched dynamics of the control systems. In this paper, a new optimal reference input shape design method based on minimizing cost function is suggested and applied, to a simple cart-pendulum system which is a simplified model of tower cranes. Since pendulum dynamics is nonlinear, analytic solution does not exist. To overcome this problem, in this paper, a machine learning approach is suggested to find optimal reference input shape for the cart position control. The feasibility of the proposed design method is verified through some simulation examples by using MatLab.
Bone plates made of biodegradable polymers have been used to fix broken bones. 3D printers are used to produce the bone plates for fracture fixing in the industry. The dimensional accuracy of the product printed by a 3D printer is less than 80%. Fracture fixing plates with less than 80% dimensional accuracy cause problems during surgery. There is an urgent need to improve the dimensional accuracy of the product in the industry. In this paper, a methodology using machine learning was proposed to improve the dimensional accuracy. The proposed methodology was evaluated through case studies. The results predicted by the machine learning methodology proposed in this paper and the experimental results were compared through the experiment. After verification, results of the proposed prediction model and the experimental results were in good agreement with each other.
Recently, improvement of productivity of the paper cup forming machine has being conducted by increasing manufacturing speed. However, rapid manufacturing speed imposes high load on cams and cam followers. It accelerates wear and cracking, and increases paper cup failure. In this study, a failure diagnosis algorithm was suggested using vibration data measured from cam driving parts. Among various paper cup forming processes, a test bed imitating the bottom paper attaching process was manufactured. Accelerometers were installed on the test bed to collect data. To diagnose failure from measured data, the K-NN (K-Nearest Neighbor) classifier was used. To find a decision boundary between normal and abnormal state, learning data were collected from normal and abnormal state, and normal and abnormal cams. A few representative features such as mean and variance were selected and transformed to the relevant form for the classifier. Classification experiments were performed with the developed classifier and data gathered from the test bed. According to assigned K values, a successful classification result was obtained which means appropriate failure recognition.
Citations
Citations to this article as recorded by
A Study on 3D Printing Conditions Prediction Model of Bone Plates Using Machine Learning Song Yeon Lee, Yong Jeong Huh Journal of the Korean Society for Precision Engineering.2022; 39(4): 291. CrossRef
Cyber-Physical Production Systems (CPPS), which pursue the implementation of machine intelligence in manufacturing systems, receive much attention as an advanced technology in Smart Factories. CPPS significantly necessitates the selflearning capability because this capability enables manufacturing objects to foresee performance results during their process planning activities and thus to make data-driven autonomous and collaborative decisions. The present work designs and implements a self-learning factory mechanism, which performs predictive process planning for energy reduction in metal cutting industries based on a hybrid-learning approach. The hybrid-learning approach is designed to accommodate traditional machine-learning and transfer-learning, thereby providing the ability of predictive modeling in both data sufficient and insufficient environments. Those manufacturing objects are agentized under the paradigm of Holonic Manufacturing Systems to determine the best energy-efficient machine tool through their self-decisions and interactions without the intervention of humans’ decisions. For such purpose, this paper includes: the proposition of the hybrid-learning approach, the design of system architecture and operational procedure for the self-learning factory, and the implementation of a prototype system.
Citations
Citations to this article as recorded by
AI-Driven Innovations in 3D Printing: Optimization, Automation, and Intelligent Control Fatih Altun, Abdulcelil Bayar, Abdulhammed K. Hamzat, Ramazan Asmatulu, Zaara Ali, Eylem Asmatulu Journal of Manufacturing and Materials Processing.2025; 9(10): 329. CrossRef
Deaf people use their own national sign or finger languages for communication. They have a lot of inconvenience in both social and financial problems. In this study, a finger language recognition system using an ensemble machine learning algorithm with an armband sensor of 8 channel surface electromyography (sEMG) is introduced. The algorithm consisted of signal acquisition, digital filtering, feature vector extraction, and an ensemble classifier based on artificial neural network (EANN). It was evaluated with Korean finger language (14 consonants, 17 vowels and 7 numbers) in 20 normal subjects. EANN was categorized with the number of classifiers (1 to 10) and the size of training data (50 to 1500). Mean accuracies and standard deviations for each structure were then obtained. Results showed that, as the number of classifiers (1 to 8) and the size of training data (50 to 300) were increased, the average accuracy of the E-ANN classifier was increased while the standard deviation was decreased. Statistical analysis showed that the optimal E-ANN structure was composed with 8 classifiers and 300 training data. This study suggested that E-ANN was more accurate than the general ANN for sign/finger language recognition.
First
Prev
