• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.331-334
• /
• 1996

It is a well-known problem that the adaptive control has a poor transient response. In order to improve this problem, the scheme that model-reference adaptive control (MRAC) uses the genetic algorithm (GA) in the search for parameters is proposed. Use genetic algorithm (GA) in the searching for controller's parameters set and conventional gradient method for fine tuning. And show the reduction of the oscillations in transient response comparing with the conventional MRAC.

• Structural Engineering and Mechanics
• /
• v.62 no.2
• /
• pp.139-142
• /
• 2017

For the conventional computational methods for structural reliability analysis, the common limitations are long computational time, large number of iteration and low accuracy. Thus, a new novel method for structural reliability analysis has been proposed in this paper based on response surface method incorporated with an improved genetic algorithm. The genetic algorithm is first improved from the conventional genetic algorithm. Then, it is used to produce the response surface and the structural reliability is finally computed using the proposed method. The proposed method can be used to compute structural reliability easily whether the limit state function is explicit or implicit. It has been verified by two practical engineering cases that the algorithm is simple, robust, high accuracy and fast computation.

• Journal of Information Technology Application
• /
• v.3 no.3
• /
• pp.1-24
• /
• 2001

This paper introduces the genetic programming algorithm(GP), which can approximate highly nonlinear functions, as a tool for the modeling of response surfaces. When the response surfaces is approximated, the very small or minimal teaming set should be used, and thus it is almost certain that GP trees will show overfilling that must be avoided at all costs. We present a novel method, calledDDBS(DirectionalDerivative-Based Smoothering), which very effectively eliminates the unwanted behaviors of GP trees such as large peaks, oscillations, and also overfitting. Four illustrative numerical examples are given to demonstrate the performance of the genetic programming algorithm that adopts DDBS.

• Journal of Information Technology Application
• /
• v.3 no.3
• /
• pp.25-40
• /
• 2001

This paper deals with the problem of generating optimal polynomials using Genetic Programming(GP). The polynomial should approximate nonlinear response surfaces. Also, there should be a consideration regarding the size of the polynomial, It is not desirable if the polynomial is too large. To build small or medium size of polynomials that enable to model nonlinear response surfaces, we use the low order Tailor series in the function set of GP, and put the constrain on generating GP tree during the evolving process in order to prevent GP trees from becoming too large size of polynomials. Also, GAGPT(Group of Additive Genetic Programming Trees) is adopted to help achieving such purpose. Two examples are given to demonstrate our method.

• Asian-Australasian Journal of Animal Sciences
• /
• v.15 no.12
• /
• pp.1690-1694
• /
• 2002

Genetic parameters were estimated for birth weight and weaning weight from three year (1991-1993) data totalling 1100 records of 25 rams to 205 ewes of Indigenous Sabi flock maintained at Grasslands Research Station in Zimbabwe. AIREML procedures were used fitting an Animal Model. The statistical model included the fixed effects of year of lambing, sex of lamb, birth type and the random effect of ewe. Weight of ewe when first joined with ram was included as a covariate. Direct heritability estimates of 0.27 and 0.38, and maternal heritability estimates of 0.24 and 0.09, were obtained for birth weight and weaning weight, respectively. The total heritability estimates were 0.69 and 0.77 for birth weight and weaning weight, respectively. Direct-aternal genetic correlations were high and positive. The corresponding genetic covariance estimates between direct and maternal effects were positive and low, 0.25 and 0.18 for birth weight and weaning weight, respectively. Responses to selection were 0.8 kg and 0.14 kg for birth weight and weaning weight, respectively. The estimated expected correlated response to selection for birth weight by directly selecting for weaning weight was 0.26. Direct heritabilities were moderate; as a result selection for any of these traits should be successful. Maternal heritabilities were low for weaning weight and should have less effect on selection response. Indirect selection can give lower response than direct selection.

• Asian-Australasian Journal of Animal Sciences
• /
• v.18 no.1
• /
• pp.8-12
• /
• 2005

Response in a modified reciprocal recurrent selection scheme for egg production was evaluated considering variable family sizes and genetic correlation between purebred and crossbred half sisters. The criteria of selection of purebred breeders included pullet's own performance, purebred full and half sisters and crossbred half sister's performance. Heritability of egg production of crossbreds (aggregate genotype) and purebred's was assumed to be 0.2 and genetic correlation between purebred and crossbred half sisters ($r_{pc}$) as 0.1, 0.2, 0.3, 0.4, 0.5, 1.0, -0.1, -0.2, -0.3, -0.4, -0.5 and -1.0. Number of dams per sire to produce purebred and crossbred progenies assumed to be 5, 6, 7, 8, while number of purebred female progeny ($N_p$) and crossbred progeny ($N_c$) per dam were considered to be 3, 4, 5 and 6 in each case. Considering phenotypic variance as unity, selection indices were constructed for different combinations of dams and progeny for each value of $r_{pc}$. Following selection index theory, response in crossbred and purebred for egg production was computed. Results indicated that response in crossbreds depended mainly on crossbred family size and also on magnitude of$r_{pc}$ irrespective of its direction, and response was greater with large crossbred family size than the purebred families. Correlated response in purebreds depends both on magnitude and direction of $r_{pc}$ and was expected to be greater with large purebred family size only. Inclusion of purebred information increased the accuracy of selection for crossbred response for higher magnitude of$r_{pc}$ irrespective of its direction. Present results indicate that desirable response in both crossbred and purebred performance is a function of $r_{pc}$ and family sizes. The ratio of crossbred and purebred family sizes can be optimized depending on the objective of improving the performance of crossbreds and/or of purebreds.

• Journal of Electrical Engineering and Technology
• /
• v.1 no.4
• /
• pp.466-471
• /
• 2006

This raper presents a new optimization method combining the genetic algorithm with the response surface method for the optimum design of a Brushless Direct Current motor. The method utilizes a regression function approximating an objective function and the window moving and zoom-in method so as to complement disadvantages of both the genetic algorithm and response surface method. The results verify that the proposed method is powerful and effective in reducing cogging torque by optimizing only a few decisive design factors compared with the conventional stochastic methods.

• 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.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.2
• /
• pp.713-718
• /
• 2015

Background: A number of association studies have been carried out to investigate the relationship between genetic polymorphisms in DNA repair genes and response to radiotherapy-based multimodality treatment of patients with rectal cancer. However, their conclusions were inconsistent. The objective of the present study was to assess the role of DNA repair gene genetic polymorphisms in predicting genetic biomarkers of the response in rectal cancer patients treated with neoadjuvant chemoradiation. Materials and Methods: Studies were retrieved by searching the PubMed database, Cochrane Library, Embase, and ISI Web of Knowledge. We conducted a meta-analysis to evaluate the association between genetic polymorphisms and the response in rectal cancer treated with neoadjuvant chemoradiation by checking odds ratios (ORs) and 95% confidence intervals (CIs). Results: Data were extracted from 5 clinical studies for this meta-analysis. The results showed that XRCC1 RS25487, XRCC1 RS179978, XRCC3 RS861539, ERCC1 RS11615 and ERCC2 RS13181 were not associated with the response in the radiotherapy-based multimodality treatment of patients with rectal cancer (p>0.05). Conclusions: This study shows that DNA repair gene common genetic polymorphisms are not significantly correlated with the radiotherapy-based multimodality treatment in rectal cancer patients.

• Plant Breeding and Biotechnology
• /
• v.5 no.4
• /
• pp.363-370
• /
• 2017

In this study, Quercus acutissima seedlings were subjected to drought for 30 days then analyzed to determine their response to water deficit. The growth phenotype, chlorophyll fluorescence response, fresh weight, dry weight, photosynthetic pigment levels, soluble sugar content, and malondialdehyde (MDA) were measured to evaluate the effects of drought on plant growth and physiology. The growth phenotype was observed by infrared (IR) digital thermal imaging after 30 days of drought treatment. The maximum, average, and minimum temperatures of drought-treated plant leaves were $1-2^{\circ}C$ higher than those of the control. In contrast, the fresh and dry weights of the dehydrated leaves were generally lower than those of the control. There were no significant differences between treatments in terms of chlorophyll a, chlorophyll b, total chlorophyll, and carotenoid levels. Nevertheless, for the drought treatment, the $F_v/F_m$ and $F_v/F_o$ ratios (chlorophyll fluorescence response) were lower than those for the control. Therefore, photosynthetic activity was lower in the dehydrated plants than the control. The drought-stressed Q. acutissima S0536 had lower soluble sugar (glucose and fructose) and higher MDA levels than the controls. These findings may explain the early growth and physiological responses of Q. acutissima to dehydration and facilitate the selection of drought-resistant tree families.