• Journal of the Korea Society for Simulation
• /
• v.28 no.2
• /
• pp.35-48
• /
• 2019

Domestic hydraulic drifters till now have been developed by benchmarking products from overseas leading companies. However, they do not have excellent impact performance as they are not suitable for characteristics (large flow rate and low pressure) of Korean hydraulic drill power pack, and therefore, research on the optimum design has not made much headway. This study performs multi-objective function optimization for hydraulic drifters whose capacity has been redesigned to deal with the large flow rate, and also with the help of this function, it aims to improve impact power and reduce supply and surge pressure. A summary of the research study is as follows: First, we set goals for improving impact power, supply pressure, and surge pressure, and then perform multi-objective function optimization on them. After that, we secure the reliability of the optimized analytical model by comparing the test results of the prototype built by the optimized design with the analysis results of the analytical model. This study used SimulationX, that is the hydraulic system analysis software, and EasyDesign, which is a multi-objective function optimization program. Through this research, we have achieved the results that satisfy the goal of developing high power drifters suitable for Korean type hydraulic drills.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.15 no.4
• /
• pp.165-174
• /
• 2011

This research aims at the optimal constitution of sensors required to identify the structural shortcoming of roller-coaster. In this research we analyzed the dynamic characteristics of roller-coaster by three dimensional FE modelling, decided on the appropriate location and number of sensors through optimal transducer theory, abstracted the mathematical value of modal features before and after damage on the basis of optimally placed and numbered sensors. and then presented it as a primary information about the basic structure which would be applied to damage estimation. As a target structure, the roller-coater at Seoul Children's Grand Park was chosen and built as a model reduced by one twentieth in size. In order to consider the Kinetics features particular to the roller-coaster structure, we made an exact three-dimensional FE modelling for the model structure by means of Spline function. As for the proper location and number of sensors, it was done by applying EIM and EOT. We also estimated the damage from the combination of strength, flexibility, and model corelation after abstracting the value of modal features. Finally the optimal transducer theory presented here in this research was proved to be valid, and the structural damage was well identified through changes in strength and flexibility. As a result, we were able to present the optimal constitution of sensors needed for the analysis of dynamic characteristics and the development of techniques in dynamic characteristics, which would ultimately contribute to the development of health monitoring for roller-coaster.

• Smart Structures and Systems
• /
• v.25 no.4
• /
• pp.487-499
• /
• 2020

This paper proposes a novel approach to model updating for a large-scale cable-stayed bridge based on ambient vibration tests coupled with a hybrid metaheuristic search algorithm. Vibration measurements are carried out under excitation sources of passing vehicles and wind. Based on the measured structural dynamic characteristics, a finite element (FE) model is updated. For long-span bridges, ambient vibration test (AVT) is the most effective vibration testing technique because ambient excitation is freely available, whereas a forced vibration test (FVT) requires considerable efforts to install actuators such as shakers to produce measurable responses. Particle swarm optimization (PSO) is a famous metaheuristic algorithm applied successfully in numerous fields over the last decades. However, PSO has big drawbacks that may decrease its efficiency in tackling the optimization problems. A possible drawback of PSO is premature convergence leading to low convergence level, particularly in complicated multi-peak search issues. On the other hand, PSO not only depends crucially on the quality of initial populations, but also it is impossible to improve the quality of new generations. If the positions of initial particles are far from the global best, it may be difficult to seek the best solution. To overcome the drawbacks of PSO, we propose a hybrid algorithm combining GA with an improved PSO (HGAIPSO). Two striking characteristics of HGAIPSO are briefly described as follows: (1) because of possessing crossover and mutation operators, GA is applied to generate the initial elite populations and (2) those populations are then employed to seek the best solution based on the global search capacity of IPSO that can tackle the problem of premature convergence of PSO. The results show that HGAIPSO not only identifies uncertain parameters of the considered bridge accurately, but also outperforms than PSO, improved PSO (IPSO), and a combination of GA and PSO (HGAPSO) in terms of convergence level and accuracy.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.1
• /
• pp.109-115
• /
• 2019

In arc-welding applying to the present automobile part manufacturing process, a wrist-hollow type arc welding robot can shorten the welding cycle time, because feedability of a welding wire is not affected by a robot posture and thus facilitates high-quality arc welding, based on stable feeding with no entanglement. In this paper, we will propose the optimization of wrist design for a wrist-hollow type 6-Axis articulated robot. Specifically, we will perform the investigation on the optimized design of inner diameter of hollow arms (Axis 4 and Axis 6) and width of the upper arm by using the simulation of robot motion characteristics, using a Genetic Algorithm (i.e., GA). Our simulations are based on $SolidWorks^{(R)}$ for robot modeling, $MATLAB^{(R)}$ for GA optimization, and $RecurDyn^{(R)}$ for analyzing dynamic characteristics of a robot. Especially $RecurDyn^{(R)}$ is incorporated in the GA module of $MATLAB^{(R)}$ for the optimization process. The results of the simulations will be verified by using $RecurDyn^{(R)}$ to show that the driving torque of each axis of the writs-hollow 6-axis robot with the optimized wrist design should be smaller than the rated output torque of each joint servomotor. Our paper will be a guide for improving the wrist-hollow design by optimizing the wrist shape at a detail design stage when the driving torque of each joint for the wrist-hollow 6-axis robot (to being developed) is not matched with the servomotor specifications.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.19 no.1
• /
• pp.51-56
• /
• 2013

From the time the product is manufactured until it is carried and ultimately used, the product is subjected to some form of handling and transportations. During this process, the product can be subjected to many potential hazards. One of them is the damage caused by shocks. In order to design a product-package system to protect the product, the peak acceleration or G force to the product that causes damage needs to be determined. When a corrugated fiberboard box loaded with products is dropped onto the ground, part of the energy acquired due to the action of the gravitational acceleration during the free fall is dissipated in the product and the package in various ways. The shock absorbing characteristics of the packaging cushion materials are presented as a family of cushion curves in which curves showing peak accelerations during impacts for a range of static loads are shown for several drop heights. The new method for determining the shock absorbing characteristics of cushioning materials for protective packaging has been described and demonstrated. It has been shown that cushion curves can be produced by combining the static compression and impact characteristics of the material. The dynamic factor was determined by the iterative least mean squares (ILMS) optimization technique in which the discrepancies between peak acceleration data predicted from the theoretical model and obtained from the impact tests are minimized. The approach enabled an efficient determination of cushion curves from a small number of experimental impact data.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.346-348
• /
• 2004

In this rarer, we introduce a new Fuzzy Polynomial Neural Networks (FPNNs)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNs based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNs-like structurenamed Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. In considering the structures of FPNN-like networks such as FPNN and FSPNN, they are almost similar. Therefore they have the same shortcomings as well as the same virtues on structural side. The proposed design procedure for networks' architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IG) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using gas furnace process dataset.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.25 no.12
• /
• pp.830-842
• /
• 2015

This paper aims to develop a SI(structural identification) technique using the kinetic energy optimization technique(KEOT) and the direct matrix updating method(DMUM) to decide on optimal location of sensors and to update FE model respectively, which ultimately contributes to a composition of more effective SHM. Owing to the characteristic structural flexing behavior of cable bridges, which makes them vulnerable to any vibration, systematic and continuous structural health monitoring (SHM) is pivotal for them. Since it is necessary to select optimal measurement locations with the fewest possible measurements and also to accurately assess the structural state of a bridge for the development of an effective SHM, a SI technique is as much important to accurately determine the modal parameters of the current structure based on the data optimally obtained. In this study, the KEOT was utilized to determine the optimal measurement locations, while the DMUM was utilized for FE model updating. As a result of experiment, the required number of measurement locations derived from KEOT based on the target mode was reduced by approximately 80 % compared to the initial number of measurement locations. Moreover, compared to the eigenvalue of the modal experiment, an improved FE model with a margin of error of less than 1 % was derived from DMUM. Finally, the SI technique for long-span bridges proposed in this study, which utilizes both KEOT and DMUM, is proven effective in minimizing the number of sensors while accurately determining the structural dynamic characteristics.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.433-436
• /
• 2004

In this paper, we introduce a new Fuzzy Polynomial Neural Networks (FPNNS)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNS based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNS-like structure named Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. The proposed design procedure for networks architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IC) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using the time series dataset of gas furnace process.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.10
• /
• pp.809-819
• /
• 2020

In this study, the twist angle and wing planform shapes were selected as design variables and optimized to secure the stability of the flying-wing type UAV. Flying-wing aircraft has no separated fuselage and tails, which has advantages in aerodynamic characteristics and stealth performance, but it is difficult to secure the flight stability. In this paper, the sweep back angle and twist angle were optimized to obtain the lateral stability, the static margin and wing planform shapes were optimized to improve the longitudinal stability of the flying-wing, then effect of the twist angle was confirmed by comparing the stability of the shape with the winglet and the shape with the twist angle. In the optimization formulation, focusing on improving stability, constraints were established, objective functions and design variables were set, then design variable sensitivity analysis was performed using the Sobol method. AVL was used for aerodynamic analysis and stability analysis, and SQP was used for optimization. The CFD analysis of the optimized shape and the simulation of the dynamic stability proved that the twist angle can be applied to the improvement of the lateral stability as well as the stealth performance in the flying-wing instead of the winglet.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.3
• /
• pp.639-647
• /
• 2011

In this paper, we introduce a design methodology of data-centroid Radial Basis Function neural networks with extended polynomial function. The two underlying design mechanisms of such networks involve K-means clustering method and Particle Swarm Optimization(PSO). The proposed algorithm is based on K-means clustering method for efficient processing of data and the optimization of model was carried out using PSO. In this paper, as the connection weight of RBF neural networks, we are able to use four types of polynomials such as simplified, linear, quadratic, and modified quadratic. Using K-means clustering, the center values of Gaussian function as activation function are selected. And the PSO-based RBF neural networks results in a structurally optimized structure and comes with a higher level of flexibility than the one encountered in the conventional RBF neural networks. The PSO-based design procedure being applied at each node of RBF neural networks leads to the selection of preferred parameters with specific local characteristics (such as the number of input variables, a specific set of input variables, and the distribution constant value in activation function) available within the RBF neural networks. To evaluate the performance of the proposed data-centroid RBF neural network with extended polynomial function, the model is experimented with using the nonlinear process data(2-Dimensional synthetic data and Mackey-Glass time series process data) and the Machine Learning dataset(NOx emission process data in gas turbine plant, Automobile Miles per Gallon(MPG) data, and Boston housing data). For the characteristic analysis of the given entire dataset with non-linearity as well as the efficient construction and evaluation of the dynamic network model, the partition of the given entire dataset distinguishes between two cases of Division I(training dataset and testing dataset) and Division II(training dataset, validation dataset, and testing dataset). A comparative analysis shows that the proposed RBF neural networks produces model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.