• Proceedings of the Korean Society of Marine Engineers Conference
• /
• 2001.11a
• /
• pp.112-119
• /
• 2001

Parameter tuning methods by Ziegler-Nickels for control systems are generally classified into Z-N(1) and Z-N(2). The purpose of this paper is to describe what relations exist between methods of Z-N(1) and Z-N(2), or how Z-N(1) method can be originated from Z-N(2) method by analyzing one loop control system of P or PI controller and time delay process. The formulas of Z-N(1) consist of process parameters, L(time delay), $K_m$(gain) and $T_m$(time constant), but Z-N(2) method is based only on the ultimate gain $K_u$ and the ultimate period $T_u$ acquired normally by practical trial without any parameters of Z-N(1). In this paper, for the first step to seek mutual relations, the simple formulas of Z-N(2) are transformed into the formulas composed of the same parameters as Z-N(1) which is derived from the analysis of frequency characteristics. Then, the approximation of the actual ultimate frequency is proposed as important premise in the translation between Z-N(1) and (2). Such equalization and approximation brings a simple approximated formula which can explain how Z-N(1) is originated from the Z-N(2) in the form of formula. And a model system is adopted to compare the approximated formula to Z-N(1) and Z-N(2) methods, the results of which show the effectiveness of the proposals.

• Proceedings of the KIEE Conference
• /
• 2007.04a
• /
• pp.201-203
• /
• 2007

The paper concerns the successive optimization for structure and parameters of fuzzy inference systems that is based on parallel Genetic Algorithms (PGA) and information data granulation (IG). PGA is multi, population based genetic algorithms, and it is used tu optimize structure and parameters of fuzzy model simultaneously, The granulation is realized with the aid of the C-means clustering. The concept of information granulation was applied to the fuzzy model in order to enhance the abilities of structural optimization. By doing that, we divide the input space to form the premise part of the fuzzy rules and the consequence part of each fuzzy rule is newly' organized based on center points of data group extracted by the C-Means clustering, It concerns the fuzzy model related parameters such as the number of input variables to be used in fuzzy model. a collection of specific subset of input variables, the number of membership functions according to used variables, and the polynomial type of the consequence part of fuzzy rules, The simultaneous optimization mechanism is explored. It can find optimal values related to structure and parameter of fuzzy model via PGA, the C-means clustering and standard least square method at once. A comparative analysis demonstrates that the Dnmosed algorithm is superior to the conventional methods.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.8
• /
• pp.423-431
• /
• 2000

This paper presents a systematic approach to the input-output data-based fuzzy modeling for the complex and uncertain nonlinear systems, in which the conventional mathematical models may fail to give the satisfying results. To do this, we propose a new method that can yield a successful fuzzy model using a mGA hybrid schemes with a fine-tuning method. We also propose a new coding method fo chromosome for applying the mGA to the structure and parameter identifications of fuzzy model simultaneously. During mGA search, multi-purpose fitness function with a penalty process is proposed and adapted to guarantee the accurate and valid fuzzy modes. This coding scheme can effectively represent the zero-order Takagi-Sugeno fuzzy model. The proposed mGA hybrid schemes can coarsely optimize the structure and the parameters of the fuzzy inference system, and then fine tune the identified fuzzy model by using the gradient descent method. In order to demonstrate the superiority and efficiency of the proposed scheme, we finally show its applications to two nonlinear systems.

• Structural Engineering and Mechanics
• /
• v.26 no.3
• /
• pp.343-361
• /
• 2007

This article presents a numerical tool for dimensioning two-threaded fasteners connecting prismatic parts subjected to fatigue tension loads that are coplanar with the screw axis. A simplified numerical model is developed from unidirectional finite elements, modeling the connected parts and screws with bent elements and the elastic contact layer between the parts with springs. An algorithm updating the contact stiffness matrix, calculating forces and displacements at each node of the structure and thus normal stresses in the screws in both static and fatigue is further developed using C language. An experimental study is also conducted in parallel with the numerical approach to validate the developed model assumptions, the numerical model and the 3D finite element results. Since stiffness values for the compressive zones in the parts are analytically difficult to determine, a statistical software method is used, from which a tuning factor is derived for identifying these stiffness values. The method is also applied to set out the influence of each parameter on the fatigue behaviour of each screw. Finally, the developed model will be used to establish a new, sophisticated, fast and accurate tool for dimensioning bolted mechanical structures.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.44 no.1
• /
• pp.25-32
• /
• 2007

In this paper, a new method for first order plus dead time(FOPDT) model identification is proposed, which can identity multiple points on a process step response in terms of classification of time response. The process input and output to the test are decomposed into the transient part and the steady-state part. The steady-state part express one FOPDT model and the transient part express variously FOPDT model using least square estimation method. The optimum parameter tuning algorithm for PID controller of the Smith Predictor is proposed through ITAE as performance index. The Simulation results show the validity and improvement of performance for various processes.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1377-1379
• /
• 1996

In this paper, each parameter was identified by the gradient descent method to overcome difficulty deciding fuzzy rules of FLC for the unknown process and the type of membership Junctions. Usually PID or optimal control theories have been mostly usee in control field so far. However, optimal control requires much time for calculation because of adaptation for disturbance and nonlinearity. And intricate technique such as MRAS which can be realized only by an expert are limited to be used in the systems requiring rapid and precise response because of comparatively longer calculating time and complicateness. Gradient descent method is a method to find Z minimizing a function about a certain vector Z. And required output of FLC is gained using gradient approaching method in order to adapt control rule parameters of FLC. Simulation proved validation of this algorithm.

• Steel and Composite Structures
• /
• v.39 no.1
• /
• pp.81-93
• /
• 2021

Splitting tensile strength (STS) is an important mechanical parameter of concrete. This study offers novel methodologies for the early prediction of this parameter. Artificial neural network (ANN), which is a leading predictive method, is synthesized with two metaheuristic algorithms, namely atom search optimization (ASO) and equilibrium optimizer (EO) to achieve an optimal tuning of the weights and biases. The models are applied to data collected from the published literature. The sensitivity of the ASO and EO to the population size is first investigated, and then, proper configurations of the ASO-NN and EO-NN are compared to the conventional ANN. Evaluating the prediction results revealed the excellent efficiency of EO in optimizing the ANN. Accuracy improvements attained by this algorithm were 13.26 and 11.41% in terms of root mean square error and mean absolute error, respectively. Moreover, it raised the correlation from 0.89958 to 0.92722. This is while the results of the conventional ANN were slightly better than ASO-NN. The EO was also a faster optimizer than ASO. Based on these findings, the combination of the ANN and EO can be an efficient non-destructive tool for predicting the STS.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.5
• /
• pp.23-29
• /
• 2020

The National Institute of Agricultural Sciences of the Rural Developement Administration (NAS, RDA) is conducting various studies on various crops, such as monitoring the cultivation environment and analyzing harvested seeds for high-throughput phenotyping. In this paper, we propose a deep learning-based rice seed segmentation method to analyze the seeds of various crops owned by the NAS. Using Mask-RCNN deep learning model, we perform the rice seed segmentation from manually taken images under specific environment (constant lighting, white background) for analyzing the seed characteristics. For this purpose, we perform the parameter tuning process of the Mask-RCNN model. By the proposed method, the results of the test on seed object detection showed that the accuracy was 82% for rice stem image and 97% for rice grain image, respectively. As a future study, we are planning to researches of more reliable seeds extraction from cluttered seed images by a deep learning-based approach and selection of high-throughput phenotype through precise data analysis such as length, width, and thickness from the detected seed objects.

• Korean Journal of Remote Sensing
• /
• v.40 no.1
• /
• pp.93-101
• /
• 2024

Accurate field crop classification is essential for various agricultural applications, yet existing methods face challenges due to diverse crop types and complex field conditions. This study aimed to address these issues by combining support vector machine (SVM) models with multi-seasonal unmanned aerial vehicle (UAV) images, texture information extracted from Gray Level Co-occurrence Matrix (GLCM), and RGB spectral data. Twelve high-resolution UAV image captures spanned March-October 2021, while field surveys on three dates provided ground truth data. We focused on data from August (-A), September (-S), and October (-O) images and trained four support vector classifier (SVC) models (SVC-A, SVC-S, SVC-O, SVC-AS) using visual bands and eight GLCM features. Farm maps provided by the Ministry of Agriculture, Food and Rural Affairs proved efficient for open-field crop identification and served as a reference for accuracy comparison. Our analysis showcased the significant impact of hyperparameter tuning (C and gamma) on SVM model performance, requiring careful optimization for each scenario. Importantly, we identified models exhibiting distinct high-accuracy zones, with SVC-O trained on October data achieving the highest overall and individual crop classification accuracy. This success likely stems from its ability to capture distinct texture information from mature crops.Incorporating GLCM features proved highly effective for all models,significantly boosting classification accuracy.Among these features, homogeneity, entropy, and correlation consistently demonstrated the most impactful contribution. However, balancing accuracy with computational efficiency and feature selection remains crucial for practical application. Performance analysis revealed that SVC-O achieved exceptional results in overall and individual crop classification, while soybeans and rice were consistently classified well by all models. Challenges were encountered with cabbage due to its early growth stage and low field cover density. The study demonstrates the potential of utilizing farm maps and GLCM features in conjunction with SVM models for accurate field crop classification. Careful parameter tuning and model selection based on specific scenarios are key for optimizing performance in real-world applications.