This paper presents the development of a design optimization module for achieving the best performance of hydrostatic bearings. The design optimization module consists of two components: a bearing performance analysis module and an optimization module that utilizes optimization algorithms. Widely recognized global search methods, genetic algorithm (GA), and particle swarm optimization (PSO) algorithm, were employed as the optimization algorithms. The design optimization problem was defined for hydrostatic bearings. Optimization design processes were carried out to improve load capacity, stiffness, and flow rate. Subsequent experimental validation was conducted through the fabrication of a practical experimental setup. The design optimization model demonstrated superior performance compared to the initial model while satisfying design conditions and constraints. This confirms the practical applicability of the design optimization module developed in this study.
Magnetic bearings are being actively adopted by the turbo-chiller industry because of their higher efficiency during partial load, quieter operation, and smaller footprint than that which machines with ball bearings provide. Since magnetic bearings are open-loop unstable, feedback control is necessary. In the industry, traditional PID-based control is preferred to model-based control, because of its simplicity. When traditional control algorithms are used, significant resources are required to obtain and tune control parameters, which is an impediment to the widespread use of magnetic bearing technology in the industry. In this paper, we propose a mixed optimization method by combining genetic algorithm and sequential quadratic programming. To obtain the initial guess to be used for the mixed optimization, a phase-margin maximization algorithm is also proposed, based on the rigid-body model of the system. Mixed optimization results in suitable control parameters in less than 2.8% of the time it takes a genetic algorithm only to find similar solutions. The proposed optimization also ensures the robustness of the control parameters. The output sensitivity measured from a prototype compressor with magnetic bearings confirms the validity of the control parameters.
This paper presents a novel method of designing an efficient locomotion pattern generating algorithm for snake robots by a genetic algorithm (GA). In search and rescue operations in disaster areas, a snake robot requires multiple locomotion patterns. To overcome the complexity of snake robot control, we used a central pattern generator (CPG)-based control method which mimics the motion of a biological snake. GA was used to optimize CPG parameters to maximize locomotion performance. The locomotion performance according to the CPG parameters change was analyzed using the snake robot simulator. The proposed locomotion pattern generation algorithm evolved quickly for the target performance and obtained CPG parameters for the desired locomotion.
Citations
Citations to this article as recorded by
A Study on I-PID-Based 2-DOF Snake Robot Head Control Scheme Using RBF Neural Network and Robust Term Sung-Jae Kim, Jin-Ho Suh Journal of Korea Robotics Society.2024; 19(2): 139. CrossRef
A Study on the Design of Error-Based Adaptive Robust RBF Neural Network Back-Stepping Controller for 2-DOF Snake Robot’s Head Sung-Jae Kim, Maolin Jin, Jin-Ho Suh IEEE Access.2023; 11: 23146. CrossRef
When the penetrator collides with the target, the penetrator has different penetrating characteristics and residual velocity after penetration, according to the geometry of the penetrator. In this study, we optimized the geometry of the penetrator using the artificial neural network and the genetic algorithm to derive the best penetration performance. The Latin hypercube sampling method was used to collect the sample data, Simulation for predicting the behavior of the penetrator was conducted with the finite cavity pressure method to generate the training data for the artificial neural network. Also, the optimal hyper parameter was derived by using the Latin hypercube sampling method and the artificial neural network was used as the fitness function of the genetic algorithm to optimize the geometry of the penetrator. The optimized geometry presented the deepest penetration depth.
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
This paper presents a search methodology for the optimal operational path of robots using a genetic algorithm. The work scheduled to be performed using a robot was characterized. Collision avoidance between the robot including the working tool and the target object was considered. In this study, we followed the general steps of data mining. We compared the time taken by the robot moving along the path created by our proposed methodology with the time taken for the robot along the path created by real humans. The results show that the path generated by this study was more efficient than that of humans.
Fault diagnosis and condition monitoring of rotating machines are important for the maintenance of the gas turbine system. In this paper, the Lab-scale rotor test device is simulated by a gas turbine, and faults are simulated such as Rubbing, Misalignment and Unbalance, which occurred from a gas turbine critical fault mode. In addition, blade rubbing is one of the gas turbine main faults, as well as a hard to detect fault early using FFT analysis and orbit plot. However, through a feature based analysis, the fault classification is evaluated according to several critical faults. Therefore, the possibility of a feature analysis of the vibration signal is confirmed for rotating machinery. The fault simulator for an acquired vibration signal is a rotor-kit based test rig with a simulated blade rubbing fault mode test device. Feature selection based on GA (Genetic Algorithms) one of the feature selection algorithm is selected. Then, through the Support Vector Machine, one of machine learning, feature classification is evaluated. The results of the performance of the GA compared with the PCA (Principle Component Analysis) for reducing dimension are presented. Therefore, through data learning, several main faults of the gas turbine are evaluated by fault classification using the SVM (Support Vector Machine).
Citations
Citations to this article as recorded by
A Study on Machine Learning-Based Feature Classification for the Early Diagnosis of Blade Rubbing Dong-hee Park, Byeong-keun Choi Sensors.2024; 24(18): 6013. CrossRef
Feature selection and feature learning in machine learning applications for gas turbines: A review Jiarui Xie, Manuel Sage, Yaoyao Fiona Zhao Engineering Applications of Artificial Intelligence.2023; 117: 105591. CrossRef
Simultaneous Fault Diagnostics for Three-Shaft Industrial Gas Turbine Waleligne Molla Salilew, Syed Ihtsham Gilani, Tamiru Alemu Lemma, Amare Desalegn Fentaye, Konstantinos G. Kyprianidis Machines.2023; 11(8): 832. CrossRef
A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps Kang Whi Kim, Jihoon Kang, Seung Hwan Park Journal of the Korean Society for Precision Engineering.2021; 38(4): 269. CrossRef
Performance Improvement of Feature-Based Fault Classification for Rotor System Won-Kyu Lee, Deok-Yeong Cheong, Dong-Hee Park, Byeong-Keun Choi International Journal of Precision Engineering and Manufacturing.2020; 21(6): 1065. CrossRef
First
Prev
