• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.270-272
• /
• 2006

In this paper, we discuss optimal design of Fuzzy Polynomial Neural Networks by means of Genetic Algorithms(GAs) using symbolic coding for non-linear data. One of the major subject of genetic algorithms is representation of chromosomes. The proposed model optimized by the means genetic algorithms which used symbolic code to represent chromosomes. The proposed gFPNN used a triangle and a Gaussian-like membership function in premise part of rules and design the consequent structure by constant and regression polynomial (linear, quadratic and modified quadratic) function between input and output variables. The performance of the proposed model is quantified through experimentation that exploits standard data already used in fuzzy modeling. These results reveal superiority of the proposed networks over the existing fuzzy and neural models.

• Proceedings of the KIEE Conference
• /
• 2001.07e
• /
• pp.141-145
• /
• 2001

In this paper, model reference PID genetic controller was proposed in order to overcome the difficulty of reflecting control performance required in the overall control system and defects of the adaptation performance in the PID genetic controller. The proposed controller comprised Inner feedback loop consisting of the PID controller and plant, and outer loop consisting of an genetic algorithm which was designed for tuning a parameter of the controller. A reference model was used for design criteria of a PID controller which characterizes and quantizes the control performance required in the overall control system. Tuning parameter of the controller is performed by the genetic algorithm. The performance of proposed algorithm was verified through experiment for the DC servo motor.

• Asian-Australasian Journal of Animal Sciences
• /
• v.30 no.8
• /
• pp.1074-1080
• /
• 2017

Objective: The growth, carcass and retail cut yield records on 1,428 Hanwoo steers obtained through progeny testing were analyzed in this study, and their heritability and genetic relationships among the traits were estimated using animal models. Methods: Two different models were compared in this study. Each model was fitted for different fixed class effects, date of slaughter for carcass traits and batch of progeny test live measurement traits, and a choice of covariates (carcass weight in Model 1 or backfat thickness in Model 2) for carcass traits. Results: The differences in body composition among individuals were deemed being unaffected by their age at slaughter, except for carcass weight and backfat thickness. Heritability estimates of body size measurements were 0.21 to 0.36. Heritability estimates of retail cut percentage were high (0.56 from Model 1 and 0.47 from Model 2). And the heritability estimates for loin muscle percentage were 0.36 from Model 1 and 0.42 from Model 2, which were high enough to consider direct selection on carcass cutability traits as effective. The genetic correlations between body size measurements and retail cut ratio (RCR) were close to zero. But, some negative genetic correlations were found with chest girths measured at yearling (Model 1) or at 24 months of age or with chest widths. Loin muscle ratio (LMR) was genetically negatively correlated with body weights or body size measurements, in general in Model 1. These relationships were low close to zero but positive in Model 2. Phenotypic correlation between cutability traits (RCR, LMR) and live body size measurements were moderate and negative in Model 1 while those in Model 2 were all close to zero. Conclusion: Therefore, the body weights or linear body measurements at an earlier age may not be the most desirable selection traits for exploitation of correlated responses to improve loin muscle or lean meat yield.

• Asian-Australasian Journal of Animal Sciences
• /
• v.14 no.1
• /
• pp.1-6
• /
• 2001

Additive and dominance genetic variances were estimated for purebred Landrace selected with line breeding from 1989 to 1995 at Miyazaki Livestock Experiment Station, Kawaminami Branch. Ten body measurements, two reproductive traits and fifteen carcass traits were analyzed with single-trait mixed model analysis. The estimates of narrow-sense heritabilities by additive model were in the range of 0.07 to 0.46 for body measurements, 0.05 to 0.14 for reproductive traits, and 0.05 to 0.68 for carcass traits. The additive model tended to slightly overestimate the narrow-sense heritabilities as compared to the additive and dominance model. The proportion of the dominance variance to total genetic variance ranged from 0.11 to 0.91 for body measurements, 0.00 to 0.65 for reproductive traits, and 0.00 to 0.86 for carcass traits. Large differences among traits were found in the ratio of dominance to total genetic variance. These results suggested that dominance effect would affect the expression of all ten body measurements, one reproductive trait, and nine carcass traits. It is justified to consider the dominance effects in genetic evaluation of the selected lines for those traits.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.6-8
• /
• 2005

In this paper, we discuss model identification of nonlinear data using GAs-based Fuzzy Polynomial Neural Networks(GAs-FPNN). Fuzzy Polynomial Neural Networks(FPNN) is proposed model based Group Method Data Handling(GMDH) and Neural Networks(NNs). Each node of FPNN is expressed Fuzzy Polynomial Neuron(FPN). Network structure of nonlinear data is created using Genetic Algorithms(GAs) of optimal search method. Accordingly, GAs-FPNN have more inflexible than the existing models (in)from structure selecting. The proposed model select and identify its for optimal search of Genetic Algorithms that are no. of input variables, input variable numbers and consequence structures. The GAs-FPNN model is select tuning to input variable number, number of input variable and the last part structure through optimal search of Genetic Algorithms. It is shown that nonlinear data model design using Genetic Algorithms based FPNN is more usefulness and effectiveness than the existing models.

• Journal of the Society of Naval Architects of Korea
• /
• v.52 no.1
• /
• pp.1-7
• /
• 2015

The weight estimation of floating offshore structures such as FPSO, TLP, semi-Submersibles, Floating Offshore Wind Turbines etc. in the preliminary design, is one of direct measures of both construction cost and basic performance. Through both literature investigation and internet search, the weight data of floating offshore structures such as FPSO and TLP was collected. In this study, the weight estimation model with the genetic programming was suggested for FPSO. The weight estimation model using genetic programming was established by fixing the independent variables based on this data. In addition, the correlation analysis was performed to make up for the weak points of genetic programming; it is apt to induce over-fitting when the number of data is relatively smaller than that of independent variables. That is, by reducing the number of variables through the analysis of the correlation between the independent variables, the increasing effect in the number of weight data can be expected. The reliability of the developed weight estimation model was within 2% of error rate.

• KIEAE Journal
• /
• v.15 no.3
• /
• pp.29-36
• /
• 2015

Purpose: This study is as only basic research for the model Development of the New-Hanok Type Service Facilities in Apartment Housing, which is as a decisive factor used as a planning element for developing the model inherited tradition, There aimed at extracting the genetic factor of Korea's traditional architecture. Method: For this purpose, Consider the concept and regulations of the New-Hanok Type Service Facilities in Apartment Housing and examined the Domestic Application Status of the New-Hanok Type Service Facilities in Apartment Housing. It sets direction of the New-Hanok Type models development based on Expert advice and the literature, and was reviewed a primal reason system of Korea as an extraction base of genetic factors. Result: Then Through the framework of the vertical axis (the form), the horizontal axis (space), It extracted the genetic factors of the Korea Traditional Architecture, classified the genetic factors extracted as the structure(layout, construction, space), features, traditional beauty, investigated the content of the form representation and spatial meaning, and were characterized. Based on the result, It were comprehensive the genetic factors extracted as plan Elements for inheriting of the traditions.

• 한국생물공학회:학술대회논문집
• /
• 2000.04a
• /
• pp.177-180
• /
• 2000

An algorithm for modeling of yeast fermentation process using genetic-fuzzy algorithm is presented in this work. The algorithm involves developing the fuzzy modeling of the process and model update capability against the system change. The membership functions of state variables and specific rates and the decision table were generated using genetic algorithm. This algorithm could replace the complex mathematical model to simple fuzzy model and cope with the change of process characteristics well.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.1
• /
• pp.9-15
• /
• 2003

To improve the performance of PID controller of high order systems by model reduction, we proposed two model reduction methods. One, Original model with two point $({\angle}G(jw)=\;-{\pi}/2,\;-{\pi})$ in Nyquist curve used gradient base method and genetic algorithm. The other, Original model without two point$({\angle}G(jw)=\;-{\pi}/2,\;-{\pi})$in Nyquist curve used to add very small dead time. This method has annexed very small dead time on the base model for reduction, and we remove it after getting the reduced model, and , we improved Smith-predictor for a dead-time compensator using genetic algorithms. This method considered four points$({\angle}G(jw)=0,\;-\pi/2,\;-\pi,\;-3\pi/2)$ in the Nyquist curve to reduce steady state error between original and reduced model. It is shown that the proposed methods have more performance than the conventional method.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.5
• /
• pp.53-64
• /
• 2019

The term 're-use' here signifies the re-use of end-of-life products without changing their form after they have been thoroughly inspected and cleaned. In the re-use network model, the distributor determines the product order quantity on the network through which new products are received from the suppliers or products are supplied to the customers through re-use of the recovered products. In this paper, we propose a reusable network model for reusable products that considers the total logistics cost from the forward logistics to the reverse logistics. We also propose a reusable network model that considers the processing and disposal costs for reuse in an optimal inventory management environment. The authors employe Genetic Algorithm (GA), which is one of the optimization techniques, to verify the validity of the proposed model. And in order to investigate the effect of the parameters on the solution, the priority-based GA (priGA) under three different parameters and the modified Hybrid GA (mhGA), in which parameters are adjusted for each generation, were applied to four examples with varying sizes in the simulation.