• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.361-366
• /
• 2013

Prediction of process parameters is very important in parameter design. If predictions are fairly accurate, the quality improvement process will be useful to save time and reduce cost. The concept of dual response approach based on response surface methodology has widely been investigated. Dual response approach may take advantages of optimization modeling for finding optimum setting of input factor by separately modeling mean and variance responses. This study proposes an alternative dual response approach based on machine learning techniques instead of statistical analysis tools. A hybrid neural network-genetic algorithm has been proposed for the purpose of parameter design. A neural network is first constructed to model the relationship between responses and input factors. Mean and variance responses correspond to output nodes while input factors are used for input nodes. Using empirical process data, process parameters can be predicted without performing real experimentations. A genetic algorithm is then applied to find the optimum settings of input factors, where the neural network is used to evaluate the mean and variance response. A drug formulation example from pharmaceutical industry has been studied to demonstrate the procedures and applicability of the proposed approach.

• Structural Engineering and Mechanics
• /
• v.44 no.5
• /
• pp.681-704
• /
• 2012

The dynamic response of Euler-Bernoulli beams to resonant harmonic moving loads is analysed. The non-dimensional form of the motion equation of a beam crossed by a moving harmonic load is solved through a perturbation technique based on a two-scale temporal expansion, which permits a straightforward interpretation of the analytical solution. The dynamic response is expressed through a harmonic function slowly modulated in time, and the maximum dynamic response is identified with the maximum of the slow-varying amplitude. In case of ideal Euler-Bernoulli beams with elastic rotational springs at the support points, starting from analytical expressions for eigenfunctions, closed form solutions for the time-history of the dynamic response and for its maximum value are provided. Two dynamic factors are discussed: the Dynamic Amplification Factor, function of the non-dimensional speed parameter and of the structural damping ratio, and the Transition Deamplification Factor, function of the sole ratio between the two non-dimensional parameters. The influence of the involved parameters on the dynamic amplification is discussed within a general framework. The proposed procedure appears effective also in assessing the maximum response of real bridges characterized by numerically-estimated mode shapes, without requiring burdensome step-by-step dynamic analyses.

• Journal of Welding and Joining
• /
• v.21 no.2
• /
• pp.76-81
• /
• 2003

Due to the environmental problem, automotive companies are trying to reduce the weight of car body. Therefore, WP(Transformation Induced Plasticity) steels, which are hish strength and ductility have been developed. The application of TRIP steel to the members has been reported to increase the energy absorption capability. Welding process is a complex process; therefore deciding the optimal welding conditions is an effective method on the basis of the experimental data. However, using a trial-and-error method from the beginning in such a wide area, in order to decide the optimal conditions requires too many numbers of experiments. To overcome these problems and to decide the optimal conditions, response surface methodology was used. Response surface methodology is a collection of mathematical and statistical techniques that are for the modeling and analysis of problems in which a response of interest is influenced by several variables and the objective is to optimize this response. The introduced method was applied to the resistance spot welding process of the TRIP steel and the welding parameters were optimized. (Received December 6, 2002)

• Earthquakes and Structures
• /
• v.8 no.6
• /
• pp.1349-1362
• /
• 2015

This paper presents an analytical study aimed at evaluating the effectiveness of using buckling-restrained braces (BRBs) in mitigating the seismic response of a case study 6 storey reinforced concrete (RC) building. In the design of the BRBs with non-prismatic cross-sections, twelve combinations of ${\alpha}$ and ${\beta}$ design parameters that influence the strength and stiffness of the BRBs, respectively, were considered. The response of the structure with and without BRBs under earthquake ground accelerations were evaluated through nonlinear dynamic analysis. Two sets of ground motions representative of the design earthquake with 10% and 50% exceedance probability in fifty years were taken into account. By comparing the structural performance of the original and buckling restrained braced structures, it was observed that the use of the BRBs were very effective in mitigating the seismic response as a retrofit scheme. However, the selection of the strength and stiffness parameters of the BRBs had considerable effect on the response characteristics of RC structures. For instance, by increasing the value of ${\alpha}$ and by decreasing the value of ${\beta}$ of the buckling-restrained braces, the maximum deformation demand of the structures increased.

• Earthquakes and Structures
• /
• v.2 no.2
• /
• pp.149-169
• /
• 2011

Dynamic response of two adjacent single degree-of-freedom (SDOF) structures connected with friction damper under base excitation is investigated. The base excitation is modeled as a stationary white-noise random process. As the force-deformation behavior of friction damper is non linear, the dynamic response of connected structures is obtained using the equivalent linearization technique. It is observed that there exists an optimum value of the limiting frictional force of the damper for which the mean square displacement and the mean square absolute acceleration responses of the connected structures attains the minimum value. The close form expressions for the optimum value of damper frictional force and corresponding mean square responses of the coupled undamped structures are derived. These expressions can be used for initial optimal design of the friction damper for connected structures. A parametric study is also carried out to investigate the influence of system parameters such as frequency ratio and mass ratio on the response of the coupled structures. It has been observed that the frequency ratio has significant effect on the performance of the friction damper, whereas the effects of mass ratio are marginal. Finally, the verification of the derived close from expressions is made by correlating the response of connected structures under real earthquake excitations.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2002.10a
• /
• pp.527-532
• /
• 2002

Near field ground motions contain distinct and large amplitude pulses in both velocity and displacement. This paper investigates characteristics of near field earthquakes and their effects on seismic demands. 20 EQGMs were selected for this purpose that satisfied 5 conditions for Near field motion. Among them ten EQGMs have one distinct peak velocity pulse in the velocity time history. In this study the responsed are Linear Elastic Response Spectrum(LERS), Response Modification Factor(R) and Inelastic Response Spectrum(IRS). The effect of the selection of Near field EQGMs on these response parameters are investigated.

• Water for future
• /
• v.28 no.4
• /
• pp.185-194
• /
• 1995

The objective of this study is to explore scale problem and to analyze the relations between scale and geomorphologic parameters of the rainfall-runoff model. Generally, measurement and calculation of geomorphologic parameters rely on and are sensitive to the resolution of source information available. Therefore, rainfall-runoff models using geomorphologic parameters should take account of the effects of the map scale used in their development. The derived rainfall-runoff model considering scale problem in this research is the GIUH type model, that is a basin IUH consisting of the channel network response and hillslope response. The cannel network response is computed by means of the diffusion analogy transformed from linearized St. Venant equation and hillslope response is calculated by 2-parameter gamma distribution function. Representing geomorphologic structure of the channel network and initial distribution of its response is width function. This width function is derived by fractal theory and Melton's law to consider scale problems and is weighted by the source location function (SLF) proposed in this research to increase the applicability.

• Smart Structures and Systems
• /
• v.3 no.3
• /
• pp.245-261
• /
• 2007

Nonlinear time-domain system identification (SI) algorithm is proposed to assess damage in a shear building by synchronously estimating time-varying stiffness and damping parameters using measured acceleration data. Mass properties have been assumed as the a priori known information. Viscous damping was utilized for the current research. To chase possible nonlinear dynamic behavior under severe vibration, an incremental governing equation of vibrational motion has been utilized. Stiffness and damping parameters are estimated at each time step by minimizing the response error between measured and computed acceleration increments at the measured degrees-of-freedom. To solve a nonlinear constrained optimization problem for optimal structural parameters, sensitivities of acceleration increment were formulated with respect to stiffness and damping parameters, respectively. Incremental state vectors of vibrational motion were computed numerically by Newmark-${\beta}$ method. No model is pre-defined in the proposed algorithm for recovering the nonlinear response. A time-window scheme together with Monte Carlo iterations was utilized to estimate parameters with noise polluted sparse measured acceleration. A moving average scheme was applied to estimate the time-varying trend of structural parameters in all the examples. To examine the proposed SI algorithm, simulation studies were carried out intensively with sample shear buildings under earthquake excitations. In addition, the algorithm was applied to assess damage with laboratory test data obtained from free vibration on a three-story shear building model.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1006-1011
• /
• 2003

In this study, a novel systematic design procedure by Genetic Algorithm of a two stage relief valve is proposed. First of all, a mathematical model describing the dynamics of a balanced piston type relief valve has been derived. Governing equations such as dynamic equations for the main spool and the pilot spool and flow equations for each orifice are established. The mathematical model is verified by comparing the results of simulation with that of experiments. Furthermore, influences of the parameters on the dynamic characteristics of a relief valve have been investigated by simulation of the proposed model. Major design parameters on the valve response are determined, which affect the system response significantly. And then, using the determined parameters, the optimization of the two stage relief valve by Genetic Algorithm, which is a random search algorithm can find the global optimum without converging local optimum, is performed. The optimal design process of a two stage relief valve is presented to determine the major design parameters. Fitness function reflects the changing pressure according to parameters. It is shown that the genetic algorithms satisfactorily optimized the major design parameters of the two stage relief valve.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.14 no.6
• /
• pp.1464-1473
• /
• 1990

Optimal design parameters are estimated from the sensitivity function and performance index variation. Suspension design modification for performance improvement and basic materials for practical applications are presented. The linear quarter model of a vehicle suspension is analyzed in order to represent the utilities of sensitivity analysis, and sensitivity function is determined in the frequency domain. The change of frequency response function is predicted, which depends on the design parameter variation and the property is verified by computer simulation. As an investigation results of sensitivity function for the vibrational amplitude of sprung mass to road profile input, it is shown that the most sensitive parameters are the suspension damping and the suspension stiffness. In order to identify the effects of these two parameters to the performance of suspension system, the performance index variation according to the changes of parameters is considered and then optimal design parameters are determined. It is verified that the system response is improved noticeably in the both of frequency and time domain after the design modification with the optimal parameters.