• Journal of Intelligence and Information Systems
• /
• v.8 no.2
• /
• pp.189-203
• /
• 2002

In this paper we compare the performances of adaptive operators in genetic algorithm. For the adaptive operators, the crossover and mutation operators of genetic algorithm are considered. One fuzzy logic controller is developed in this paper and two heuristics is presented from conventional works for constructing the operators. The fuzzy logic controller and two conventional heuristics adaptively regulate the rates of the operators during genetic search process. All the algorithms are tested and analyzed in numerical examples. Finally, the best algorithm is recommended.

• Industrial Engineering and Management Systems
• /
• v.7 no.1
• /
• pp.9-22
• /
• 2008

As an extension of a previous work by the authors (Song and Esogbue, 2006), a new algorithm for automated modeling of nonstationary seasonal time series is presented in this paper. Issues relative to the methodology for building automatically seasonal time series models and periodic time series models are addressed. This is achieved by inspecting the trend, estimating the seasonality, determining the orders of the model, and estimating the parameters. As in our previous work, the major instruments used in the model identification process are correlograms of the modeling errors while the least square method is used for parameter estimation. We provide numerical illustrations of the performance of the new algorithms with respect to building both seasonal time series and periodic time series models. Additionally, we consider forecasting and exercise the models on some sample time series problems found in the literature as well as real life problems drawn from the retail industry. In each instance, the models are built automatically avoiding the necessity of any human intervention.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.721-723
• /
• 2004

A trust-region method is a quite attractive optimization technique. It is, in general, faster than the steepest descent method and is free of a learning rate unlike the gradient-based methods. In addition to its convergence property (between linear and quadratic convergence), ifs stability is always guaranteed, in contrast to the Newton's method. In this paper, we present an efficient implementation of the maximum likelihood independent component analysis (ICA) using the trust-region method, which leads to trust-region-based ICA (TR-ICA) algorithms. The useful behavior of our TR-ICA algorithms is confimed through numerical experimental results.

• Structural Engineering and Mechanics
• /
• v.7 no.6
• /
• pp.601-614
• /
• 1999

The mechanical behaviours of the structure made from composite materials or the structure with periodic configurations depend not only on the macroscopic conditions of structure, but also on the detailed configurations. The Two-Scale Analysis (TSA) method for these structures, which couples the macroscopic characteristics of structure with its detailed configurations, is configurations, is presented for 2 or 3 dimensional case in this paper. And the finite element algorithms based on TSA are developed, and some results of numerical experiments are given. They show that TSA with its finite element algorithms is more effective.

• Smart Structures and Systems
• /
• v.10 no.1
• /
• pp.47-65
• /
• 2012

The presence of crack-like defects in mechanical and structural elements produces failures during their service life that in some cases can be catastrophic. So, the early detection of the fatigue cracks is particularly important because they grow rapidly, with a propagation velocity that increases exponentially, and may lead to long out-of-service periods, heavy damages of machines and severe economic consequences. In this work, a non-destructive method for the detection and identification of elliptical cracks in shafts based on stress wave propagation is proposed. The propagation of a stress wave in a cracked shaft has been numerically analyzed and numerical results have been used to detect and identify the crack through the genetic algorithm optimization method. The results obtained in this work allow the development of an on-line method for damage detection and identification for cracked shaft-like components using an easy and portable dynamic testing device.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.56
• /
• pp.9-18
• /
• 2000

This paper considers the problem of two-machine, M-parallel flow shop scheduling to minimize makespan, and proposes a series of heuristic algorithms and a branch and bound algorithm. Two processing times of each job at two machines on each line are identical on any line. Since each flow-shop line consists of two machines, Johnson's sequence is optimal for each flow-shop line. Heuristic algorithms are developed in this paper by combining a "list scheduling" method and a "local search with global evaluation" method. Numerical experiments show that the proposed heuristics can efficiently give optimal or near-optimal schedules with high accuracy. with high accuracy.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.12 no.1
• /
• pp.711-722
• /
• 2020

An optimization framework using genetic algorithms has been developed towards an automated parametric optimization of the Octabuoy semi-submersible design. Compared with deep draft production units, the design of the shallow draught Octabuoy semi-submersible provides a floating system with improved motion characteristics, being less susceptible to vortex induced motions in loop currents. The relatively large water plane area results in a decreased natural heave period, which locates the floater in the wave period range with more wave energy. Considering this, the hull design of Octabuoy semi-submersible has been optimized to improve the floater's motion performance. The optimization has been conducted with optimized parameters of the pontoon's rectangular cross section area, the cone shaped section's height and diameter. Through numerical evaluations of both the 1st-order and 2nd-order hydrodynamics, the optimization through genetic algorithms has been proven to provide improved hydrodynamic performance, in terms of heave and pitch motions. This work presents a meaningful framework as a reference in the process of floating system's design.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.3
• /
• pp.114-121
• /
• 1997

In this paper, the automated initial process planning for 3-axis NC machining of sculptured surfaces is persented. The solution algorithms determining three process planning functions, i.e. machining feasibility, setup orientation and feasible machine selection are developed. The machining feasibility is determined by means of BSM(Binary Spherical Map) which derives its solution quickly in algebraic form, and the setup orientation is determined so that the cutting force is minimized. Finally, the feasible machine is determined by computing the minimum motion ranges of each control axisl. The developed algorithms are tested by numerical simulations, convincing they can by readily implemented on the CAD/CAM system as a process planner.

• Structural Engineering and Mechanics
• /
• v.88 no.6
• /
• pp.535-549
• /
• 2023

Deep extreme learning machine (DELM) and multi-verse optimization algorithms (MVO) are hybridized for designing an optimal and adaptive control framework for uncertain buildings. In this approach, first, a robust model predictive control (RMPC) scheme is developed to handle the problem uncertainty. The optimality and adaptivity of the proposed controller are provided by the optimal determination of the tunning weights of the linear programming (LP) cost function for clustered external loads using the MVO. The final control policy is achieved by collecting the clustered data and training them by DELM. The efficiency of the introduced control scheme is demonstrated by the numerical simulation of a ten-story benchmark building subjected to earthquake excitations. The results represent the capability of the proposed framework compared to robust MPC (RMPC), conventional MPC (CMPC), and conventional DELM algorithms in structural motion control.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.9-15
• /
• 2024

Tree-based algorithms have been the dominant methods used build a prediction model for tabular data. This also includes personal credit data. However, they are limited to compatibility with categorical and numerical data only, and also do not capture information of the relationship between other features. In this work, we proposed an ensemble model using the Transformer architecture that includes text features and harness the self-attention mechanism to tackle the feature relationships limitation. We describe a text formatter module, that converts the original tabular data into sentence data that is fed into FinBERT along with other text features. Furthermore, we employed FT-Transformer that train with the original tabular data. We evaluate this multi-modal approach with two popular tree-based algorithms known as, Random Forest and Extreme Gradient Boosting, XGBoost and TabTransformer. Our proposed method shows superior Default Recall, F1 score and AUC results across two public data sets. Our results are significant for financial institutions to reduce the risk of financial loss regarding defaulters.