• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2000.04a
• /
• pp.347-356
• /
• 2000

This paper compares four models of artificial neural networks (ANN) supported by genetic algorithms the prediction of stock price index. Previous research proposed many hybrid models of ANN and genetic algorithms(GA) in order to train the network, to select the feature subsets, and to optimize the network topologies. Most these studies, however, only used GA to improve a part of architectural factors of ANN. In this paper, GA simultaneously optimized multiple factors of ANN. Experimental results show that GA approach to simultaneous optimization for ANN (SOGANN3) outperforms the other approaches.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.700-702
• /
• 2000

The new design methodology of a hybrid fuzzy controller by means of the genetic algorithms is presented. In fuzzy controller which has been widely applied and used. in order to construct the best fuzzy rules that include adjustment of fuzzy sets, a highly skilled techniques using trial and error are required. To deal with such a problem, first, a hybrid fuzzy controller(HFC) related to the optimal estimation of control parameters is proposed. The HFC combined a PID controller with a fuzzy controller concurrently produces the better output performance than any other controller from each control output in steady state and transient state. Second, a auto-tuning algorithms is presented to automatically improve the performance of hybrid fuzzy controller, utilizing the simplified reasoning method and genetic algorithms. In addition, to obtain scaling factors and PID Parameters of HFC using GA, three kinds of estimation modes such as basic, contraction, and expansion mode are effectively utilized. The HFCs are applied to the first-order second-order process with time-delay and DC motor Computer simulations are conducted at step input and the performances of systems are evaluated and also discussed from performance indices.

• Journal of Intelligence and Information Systems
• /
• v.13 no.3
• /
• pp.63-82
• /
• 2007

In this paper, we propose an adaptive hybrid genetic algorithm (ahGA) approach for effectively solving multistage-based scheduling problems in flexible manufacturing system (FMS) environment. The proposed ahGA uses a neighborhood search technique for local search and an adaptive scheme for regulation of GA parameters in order to improve the solution of FMS scheduling problem and to enhance the performance of genetic search process, respectively. In numerical experiment, we present two types of multistage-based scheduling problems to compare the performances of the proposed ahGA with conventional competing algorithms. Experimental results show that the proposed ahGA outperforms the conventional algorithms.

• Journal of Power Electronics
• /
• v.4 no.3
• /
• pp.127-137
• /
• 2004

This paper introduces a hybrid induction motor control using a genetically optimized pseudo-on-line method. Optimization results from the use of a look-up table based on genetic algorithms to find the global optimum of an unconstrained optimization problem. The approach to induction motor control includes a pseudo-on-line procedure that optimally estimates parameters of a fuzzy PID (FPID) controller. The proposed hybrid genetic fuzzy PID (GFPID) controller is applied to speed control of a 3-phase induction motor and its computer simulation is carried out. Simulation results show that the proposed controller performs better than conventional FPID and PID controllers. The contribution of this paper is the introduction of a high performance hybrid form of induction motor control that makes on-line and real-time control of the drive system possible.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.9
• /
• pp.802-809
• /
• 2005

A hybrid genetic algorithm is proposed for the problem of generating laser torch paths to cut a stock plate nested with free-formed parts each having a set of sub-parts. In the problem, the total unproductive travel distance of the torch is minimized. The problem is shown to be formulated as a special case of the standard travelling salesman problem. The hybrid genetic algorithm for solving the problem is hierarchically structured: First, it uses a genetic algorithm to find the cutting path f3r the parts and then, based on the obtained cutting path, sequence of sub-parts and their piercing locations are optimally determined by using a combined genetic and heuristic algorithms. This process is repeated until any progress in the total unproductive travel distance is not achieved. Computational results are provided to illustrate the validity of the proposed algorithm.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.1
• /
• pp.76-86
• /
• 2002

In this paper, we suggest a hybrid genetic algorithm reinforced by a fuzzy logic controller (flc-HGA) to overcome weaknesses of conventional genetic algorithms: the problem of parameter fine-tuning, the lack of local search ability, and the convergence speed in searching process. In the proposed flc-HGA, a fuzzy logic controller is used to adaptively regulate the fine-tuning structure of genetic algorithm (GA) parameters and a local search technique is applied to find a better solution in GA loop. In numerical examples, we apply the proposed algorithm to a simple test problem and two complex combinatorial optimization problems. Experiment results show that the proposed algorithm outperforms conventional GAs and heuristics.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.708-711
• /
• 1997

This paper proposes the hybrid algorithm for the optimization of the structure and parameters of the fuzzy neural networks by genetic algorithms (GA) to improve the behaviour and the design of fuzzy neural networks. Fuzzy neural networks have a distinguishing feature in that they can possess the advantage of both neural networks and fuzzy systems. In this way, we can bring the low-level learning and computational power of neural networks into fuzzy systems and also high-level, human like IF-THEN rule thinking and reasoning of fuzzy systems into neural networks. As a result, there are many research works concerning the optimization of the structure and parameters of fuzzy neural networks. In this paper, we propose the hybrid algorithm that can optimize both the structure and parameters of fuzzy neural networks. Numerical example is provided to show the advantages of the proposed method.

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.146-148
• /
• 2005

In this paper, we propose a genetically optimized identification of information granulation(IG)-based fuzzy model. To optimally design the IG-based fuzzy model we exploit a hybrid identification through genetic alrogithms(GAs) and Hard C-Means (HCM) clustering. An initial structure of fuzzy model is identified by determining the number of input, the seleced input variables, the number of membership function, and the conclusion inference type by means of GAs. Granulation of information data with the aid of Hard C-Means(HCM) clustering algorithm help determine the initial paramters of fuzzy model such as the initial apexes of the membership functions and the initial values of polyminial functions being used in the premise and consequence part of the fuzzy rules. And the inital parameters are tuned effectively with the aid of the genetic algorithms and the least square method. And also, we exploite consecutive identification of fuzzy model in case of identification of structure and parameters. Numerical example is included to evaluate the performance of the proposed model.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2004.11a
• /
• pp.171-178
• /
• 2004

This study suggests a general paradigm enhancing genetic mutability. Mutability among heterogeneous members in a genetic population has been a major problem in application of genetic programming to diverse business problems. This suggested paradigm is implemented to developing new methods from existing methods. Within the evolutionary approach taken to designing new methods, a general representation scheme of the genetic programming framework, called a kernel, is introduced. The kernel is derived from the literature of algorithms and heuristics for combinatorial optimization problems. The commonality and differences among these methods have been identified and again combined by following the genetic inheritance merging them. The kernel was tested for selected methods in combinatorial optimization. It not only duplicates the methods in the literature, it also confirms that each of the possible solutions from the genetic mutation is in a valid form, a running program. This evolutionary method suggests diverse hybrid methods in the form of complete programs through evolutionary processes. It finally summarizes its findings from genetic simulation with insight.

• The KIPS Transactions:PartB
• /
• v.14B no.5
• /
• pp.369-376
• /
• 2007

This paper proposes a hybrid genetic algorithm(HGA) for the classifier ensemble selection. HGA is added a local search operation for increasing the fine-turning of local area. This paper apply hybrid and simple genetic algorithms(SGA) to the classifier ensemble selection problem in order to show the superiority of HGA. And this paper propose two methods(SSO: Sequential Search Operations, CSO: Combinational Search Operations) of local search operation of hybrid genetic algorithm. Experimental results show that the HGA has better searching capability than SGA. The experiments show that the CSO considering the correlation among classifiers is better than the SSO.