• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.288-291
• /
• 2004

기존의 SOFPNN은 데이터 수가 적고 비선형 요소가 많은 시스템에 대한 체계적이고 효율적인 최적 모델 을 구축할 수 있었으며 각 층 노드의 선택 입력을 변화시킴으로써 네트워크 구조 전체의 적응능력을 향상 시켰다. SOFPNN의 구조는 퍼지 다항식 뉴론(FPN)들로 구성되어 있으며, 층이 진행하는 동안 모델 스스로 노드의 선택과 제거를 통해 최적의 네트워크 구조를 생성할 수 있는 유연성을 가지고 있다. 그러나, 노드의 입력변수의 수와 규칙 후반부 다항식 차수 그리고 입력변수는 설계자의 경험 또는 반복적인 학습을 통해 선호된 네트워크 구조를 선택하였으나, 최적의 네트워크 구조를 구축하는데는 어려옴이 내재되어 있었다. 본 논문에서는 자기구성 퍼지 다항식 뉴럴네트워크(Self-Organizing Fuzzy Polynomial Neural Networks： SOFPNN)을 최적화시키기 위해 유전자 알고리즘을 이용하여 자기구성 퍼지 다항식 뉴럴 네트워크의 입력변수의 수와 이에 해당되는 입력변수 그리고 규칙 후반부 다항식의 차수를 탐색하여 최적 의 자기구성 퍼지 다항식 뉴럴 네트워크를 구축한다. 따라서 모델 구축에 있어서 유연성과 정확성을 가지며 객관적이고 좀 더 정확한 예측 능력을 가진 SOFPNN 모델 구조를 구축할 수가 있다.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1887-1888
• /
• 2008

본 논문에서는 퍼지 추론 기반의 다항식 RBF 뉴럴네트워크(Polynomial Radial Basis Function Neural Network; pRBFNN)를 설계하고 PSO(Particle Swarm Optimization) 알고리즘을 이용하여 모델의 파라미터를 동정한다. 제안된 모델은 "IF-THEN" 형식으로 기술되는 퍼지 규칙에 의해 조건부, 결론부, 추론부의 기능적 모듈로 표현된다. 조건부의 입력공간 분할에는 HCM 클러스터링에 기반을 두어 구조가 결정되며, 기존에 주로 사용된 가우시안 함수를 RBF로 이용하고, 원뿔형태의 선형 함수를 제안한다. 또한 입력공간 분할시 데이터 집합의 특성을 반영하기 위해 분포상수를 각 입력마다 고려하여 설계함으로서 공간 분할의 정밀성을 높인다. 결론부에서는 기존 상수항의 연결가중치를 다항식 형태로 표현하는 pRBFNN을 제안한다. 제안한 모델의 성능을 평가하기 위해 Box와 Jenkins가 사용한 가스로 시계열 데이터를 적용하고, 기존 모델과의 근사화와 일반화 능력에 대하여 토의한다.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.2
• /
• pp.71-77
• /
• 2016

There has been unknown polynomial time algorithm for cellular formation problem (CFP) that is one of the NP-hard problem. Therefore metaheuristic method has been applied this problem to obtain approximated solution. This paper shows the existence of polynomial-time heuristic algorithm in CFP. The proposed algorithm performs coarse-grained and fine-grained cell formation process. In coarse-grained cell formation process, the cell can be formed in accordance with machine frequently used that is the number of other products use same machine with special product. As a result, the machine can be assigned to most used cell. In fine-grained process, the product and machine are moved into other cell that has a improved grouping efficiency. For 35 experimental data, this heuristic algorithm performs better grouping efficiency for 12 data than best known of meta-heuristic methods.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.3 no.1
• /
• pp.7-12
• /
• 2003

In recent years, many attempts have been made to predict the behavior of bonds, currencies, stock or other economic markets. Most previous experiments used the neural network models for the stock market forecasting. The KOSPI 200 (Korea Composite Stock Price Index 200) is modeled by using different neural networks and fuzzy logic. In this paper, the neural network, the dynamic polynomial neural network (DPNN) and the fuzzy logic employed for the prediction of the KOSPI 200. The prediction results are compared by the root mean squared error (RMSE) and scatter plot, respectively. The results show that the performance of the fuzzy system is little bit worse than that of the DPNN but better than that of the neural network. We can develop the desired fuzzy system by optimization methods.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.12 no.7
• /
• pp.1102-1109
• /
• 2001

This paper presents the new architecture of an adaptive predistortion linearizer using the polynomial type predistorter. In the proposed linearizer, most of the processes, including the predistortion, are performed with a digital signal processor(DSP). The recursive least squares(RLS) algorithm is employed for the optimization process to minimize the errors between the predistorter and postdistorter output signals. Simulation results demonstrate that the adjacent channel power ratio(ACPR) is improved by greater than 40 dB at the band edge with linearization. The convergence and reconvergence performance of the linearizer is also satisfactory.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.18-31
• /
• 2002

In this study, we introduce a concept of advanced neurofuzzy polynomial networks(ANFPN), a hybrid modeling architecture combining neurofuzzy networks(NFN) and polynomial neural networks(PNN). These networks are highly nonlinear rule-based models. The development of the ANFPN dwells on the technologies of Computational Intelligence(Cl), namely fuzzy sets, neural networks and genetic algorithms. NFN contributes to the formation of the premise part of the rule-based structure of the ANFPN. The consequence part of the ANFPN is designed using PNN. At the premise part of the ANFPN, NFN uses both the simplified fuzzy inference and error back-propagation learning rule. The parameters of the membership functions, learning rates and momentum coefficients are adjusted with the use of genetic optimization. As the consequence structure of ANFPN, PNN is a flexible network architecture whose structure(topology) is developed through learning. In particular, the number of layers and nodes of the PNN are not fixed in advance but is generated in a dynamic way. In this study, we introduce two kinds of ANFPN architectures, namely the basic and the modified one. Here the basic and the modified architecture depend on the number of input variables and the order of polynomial in each layer of PNN structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the ANFPN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed ANFPN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.2
• /
• pp.58-65
• /
• 2021

Meta-heuristic algorithms have been developed to efficiently solve difficult problems and obtain a global optimal solution. A common feature mimics phenomenon occurring in nature and reliably improves the solution through repetition. And at the same time, the probability is used to deviate from the regional optimal solution and approach the global optimal solution. This study compares the algorithm created based on the above common points with existed SA and HS to show advantages in time and accuracy of results. Existing algorithms have problems of low accuracy, high memory, long runtime, and ignorance. In a two-variable polynomial, the existing algorithms show that the memory increases and the accuracy decrease. In order to improve the accuracy, the new algorithm increases the number of initial inputs and increases the efficiency of the search by introducing a direction using vectors. And, in order to solve the optimization problem, the results of the last experiment were learned to show the learning effect in the next experiment. The new algorithm found a solution in a short time under the experimental conditions of long iteration counts using a two-variable polynomial and showed high accuracy. And, it shows that the learning effect is effective in repeated experiments.

• Journal of Electrical Engineering and Technology
• /
• v.3 no.1
• /
• pp.101-107
• /
• 2008

We introduce an optimization of fuzzy set-fuzzy systems based on IG (Information Granules). The proposed fuzzy model implements system structure and parameter identification by means of IG and GAs. The concept of information granulation was coped with to enhance the abilities of structural optimization of the fuzzy model. Granulation of information realized with C-Means clustering helps determine the initial parameters of the fuzzy model such as the initial apexes of the membership functions in the premise part and the initial values of polynomial functions in the consequence part of the fuzzy rules. The initial parameters are adjusted effectively with the help of the GAs and the standard least square method. To optimally identify the structure and the parameters of the fuzzy model we exploit GAs with successive tuning method to simultaneously search the structure and the parameters within one individual. We also consider the variant generation-based evolution to adjust the rate of identification of the structure and the parameters in successive tuning method. The proposed model is evaluated with the performance of the conventional fuzzy model.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.13 no.4
• /
• pp.81-89
• /
• 2005

The response surface method is the statistical method which can be applied to the non-sensitivity based optimization. The response surface which is constructed by the least square method contains only the polynomial terms so that the global maximum and minimum points are easily obtained. In this paper, this response surface method is utilized to optimize the crashworthiness of auto-body members. As the first step, the thickness of a simple circular tube is optimized to confirm the application of the response surface method to the crashworthiness. Optimization of the thickness on the front side member is, then, performed with the constructed response surface of the absorbed energy and deformation. Optimization results demonstrate that the absorbed energy and the deformation pattern of the front side member is improved in the viewpoint of enhancement of the crashworthiness.

• Steel and Composite Structures
• /
• v.32 no.4
• /
• pp.455-466
• /
• 2019

Despite the rapid advancement in computing resources, many real-life design and optimization problems in structural engineering involve huge computation costs. To counter such challenges, approximate models are often used as surrogates for the highly accurate but time intensive finite element models. In this paper, surrogates for first-order shear deformation based finite element models are built using a polynomial regression approach. Using statistical techniques like Box-Cox transformation and ANOVA, the effectiveness of the surrogates is enhanced. The accuracy of the surrogate models is evaluated using statistical metrics like $R^2$, $R^2{_{adj}}$, $R^2{_{pred}}$ and $Q^2{_{F3}}$. By combining these surrogates with nature-inspired multi-criteria decision-making algorithms, namely multi-objective genetic algorithm (MOGA) and multi-objective particle swarm optimization (MOPSO), the optimal combination of various design variables to simultaneously maximize fundamental frequency and frequency separation is predicted. It is seen that the proposed approach is simple, effective and good at inexpensively producing a host of optimal solutions.