• International Journal of Fluid Machinery and Systems
• /
• v.8 no.3
• /
• pp.209-219
• /
• 2015

A robust multi-fidelity optimization methodology has been developed, focusing on efficiently handling industrial runner design of hydraulic Francis turbines. The computational task is split between low- and high-fidelity phases in order to properly balance the CFD cost and required accuracy in different design stages. In the low-fidelity phase, a physics-based surrogate optimization loop manages a large number of iterative optimization evaluations. Two derivative-free optimization methods use an inviscid flow solver as a physics-based surrogate to obtain the main characteristics of a good design in a relatively fast iterative process. The case study of a runner design for a low-head Francis turbine indicates advantages of integrating two derivative-free optimization algorithms with different local- and global search capabilities.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.5
• /
• pp.748-754
• /
• 2008

The general capacitor placement problem is a combinatorial optimization problem having an objective function composed of power losses and capacitor installation costs subject to bus voltage constraints. In this paper, a global optimization technique, which employing the chaos search algorithm, is applied to solve optimal capacitor placement problem with reducing computational effort and enhancing global optimality of the solution. Chaos method in optimization problem searches the global optimal solution on the regularity of chaotic motions and easily escapes from local or near optimal solution than stochastic optimization algorithms. The chaos optimization method is tested on 9 buses and 69 buses system to illustrate the effectiveness of the proposed method.

• Structural Engineering and Mechanics
• /
• v.52 no.3
• /
• pp.445-468
• /
• 2014

In this paper a new particle swarm ray optimization algorithm is proposed for truss shape and size optimization with natural frequency constraints. These problems are believed to represent nonlinear and non-convex search spaces with several local optima and therefore are suitable for examining the capabilities of new algorithms. The proposed algorithm can be viewed as a hybridization of Particle Swarm Optimization (PSO) and the recently proposed Ray Optimization (RO) algorithms. In fact the exploration capabilities of the PSO are tried to be promoted using some concepts of the RO. Five numerical examples are examined in order to inspect the viability of the proposed algorithm. The results are compared with those of the PSO and some other existing algorithms. It is shown that the proposed algorithm obtains lighter structures in comparison to other methods most of the time. As will be discussed, the algorithm's performance can be attributed to its appropriate exploration/exploitation balance.

• Journal of Electrical Engineering and Technology
• /
• v.6 no.6
• /
• pp.750-759
• /
• 2011

This paper focuses on the control of bidirectional power flow in the electric shipboard power systems, especially in the Medium-Voltage Direct Current (MVDC) shipboard power system. Bidirectional power control between the main MVDC bus and the local zones can improve the energy efficiency and control flexibility of electric ship systems. However, since the MVDC system contains various nonlinear loads such as pulsed power load and radar in various subsystems, the voltage of the MVDC and the local zones varies significantly. This voltage variation affects the control performance of the bidirectional DC-DC converters as exogenous disturbances. To improve the control performance regardless of uncertainties and disturbances, this paper proposes a novel controller design method of the bidirectional DC-DC converters using $L_1$ control theory and intelligent optimization algorithm. The performance of the proposed method is verified via large-scale real-time digital simulation of a notional shipboard MVDC power system.

• Journal of Applied Biological Chemistry
• /
• v.56 no.2
• /
• pp.119-129
• /
• 2013

Thirteen different Streptomyces isolates were evaluated for their ability to produce keratinase using chicken feather as a sole carbon and nitrogen sources under solid state fermentation (SSF). Streptomyces sp. NRC 13S produced the highest keratinase activity [1,792 U/g fermented substrate (fs)]. The phenotypic characterization and analysis of 16S rDNA sequencing of the isolate were studied. Optimization of SSF medium for keratinase production by the local isolate, Streptomyces sp. NRC13S, was carried out using the one-variable-at-a-time and the statistical approaches. In the first optimization step, the effect of incubation period, initial moisture content, initial pH value of the fermentation medium, and supplementation of some agro-industrial by-products on keratinase production were evaluated. The strain produced about 2,310 U/gfs when it grew on chicken feather with moisture content of 75% (w/w), feather: fodder yeast ratio of 70:30 (w/w), and initial pH 7 using phosphate buffer after 8 days. Based on these results, the Box-Behnken design and response surface methodology were applied to find out the optimal conditions for the enzyme production. The corresponding maximal production of keratinase was about 2,569.38 U/gfs.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.44 no.7
• /
• pp.1-13
• /
• 2002

Two global optimization methods, the SCE-UA method and the Annealing-simplex (A-S) method for calibrating a daily rainfall-runoff model, a Tank model, was compared with that of the Downhill Simplex method. The performance of the four objective functions, DRMS (daily root mean square), HMLE (heteroscedastic maximum likelihood estimator), ABSERR (mean absolute error), and NS (Nash-Sutcliffe measure), was tested and synthetic data and historical data were used. In synthetic data study. 100% success rates for all objective functions were obtained from the A-S method, and the SCE-UA method was also consistently able to obtain good estimates. The downhill simplex method was unable to escape from local optimum, the worst among the methods, and converged to the true values only when the initial guess was close to the true values. In the historical data study, the A-S method and the SCE-UA method showed consistently good results regardless of objective function. An objective function was developed with combination of DRMS and NS, which putted more weight on the low flows.

• The KIPS Transactions:PartA
• /
• v.8A no.1
• /
• pp.48-55
• /
• 2001

Using BDD, we can represent Boolean functions uniquely and compactly, Hence, BDD have become widely used for CAD applications, such as logic synthesis, formal verification, and etc. The size of the BDD representation for a function is very sensitive to the choice of orderings on the input variables. Therefore, it is very important to find a good variable ordering which minimize the size of the BDD. Since finding an optimal ordering is NP-complete, several heuristic algorithms have been proposed to find good variable orderings. In this paper, we propose a variable ordering algorithm based on the $\mu$O(microcanonical optimization). $\mu$O consists of two distinct procedures that are alternately applied : Initialization and Sampling. The initialization phase is to executes a fast local search, the sampling phase leaves the local optimum obtained in the previous initialization while remaining close to that area of search space. The proposed algorithm has been experimented on well known benchmark circuits and shows superior performance compared to a algorithm based on simulated annealing.

• Journal of Digital Contents Society
• /
• v.18 no.5
• /
• pp.969-976
• /
• 2017

This paper proposes the learning method of an artificial neural network and a convolutional neural network using the WFSO algorithm developed as an optimization algorithm. Since the optimization algorithm searches based on a number of candidate solutions, it has a drawback in that it is generally slow, but it rarely falls into the local optimal solution and it is easy to parallelize. In addition, the artificial neural networks with non-differentiable activation functions can be trained and the structure and weights can be optimized at the same time. In this paper, we describe how to apply WFSO algorithm to artificial neural network learning and compare its performances with error back-propagation algorithm in multilayer artificial neural networks and convolutional neural networks.

• International Journal of Highway Engineering
• /
• v.8 no.4 s.30
• /
• pp.159-179
• /
• 2006

By using the current road design method that is based on the regulation about structure and facilities standard of the road and the route plan guide of a national road and the alignment optimization road design method which is studied in the inside and outside of country, this study operate the route plan of the sample study and compare and analysis the route character, consequently the current design method has local optimization that is formed the plan by the stage and the section. Alignment optimization road design has the system optimal route search. But cost function has limite that caused by construction parameter that is not included in cost function. So we design a road route included cost function in main fields. As a result, we obtain a realistic and economically road route. The alignment optimization road design model has to be made up some problems, like the change of vertical gradient in the tunnel section, though this defects it has a lot of merits as a geometric design tool, especially in the feasibility study and the scheme design.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.169.4-169
• /
• 2001

Conventional methods to solve the nonlinear programming problem range from augmented Lagrangian methods to sequential quadratic programming (SQP) methods. NPSOL, which is a SQP code, has been widely used to solve various optimization problems but is still subject to many numerical problems such as convergence to local optima, difficulties in initialization and in handling non-smooth cost functions. Recently, many evolutionary methods have been developed for constrained optimization. Among them, CEALM (Co-Evolutionary Augmented Lagrangian Method) shows excellent performance in the following aspects: global optimization capability, low sensitivity to the initial parameter guessing, and excellent constraint handling capability due to the benefit of the augmented Lagrangian function. This algorithm is ...