• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.61 no.1
• /
• pp.135-142
• /
• 2012

In this paper, we proposed a new architecture called radial basis function-based polynomial neural networks classifier that consists of heterogeneous neural networks such as radial basis function neural networks and polynomial neural networks. The underlying architecture of the proposed model equals to polynomial neural networks(PNNs) while polynomial neurons in PNNs are composed of Fuzzy-c means-based radial basis function neural networks(FCM-based RBFNNs) instead of the conventional polynomial function. We consider PNNs to find the optimal local models and use RBFNNs to cover the high dimensionality problems. Also, in the hidden layer of RBFNNs, FCM algorithm is used to produce some clusters based on the similarity of given dataset. The proposed model depends on some parameters such as the number of input variables in PNNs, the number of clusters and fuzzification coefficient in FCM and polynomial type in RBFNNs. A multiobjective particle swarm optimization using crowding distance (MoPSO-CD) is exploited in order to carry out both structural and parametric optimization of the proposed networks. MoPSO is introduced for not only the performance of model but also complexity and interpretability. The usefulness of the proposed model as a classifier is evaluated with the aid of some benchmark datasets such as iris and liver.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.26 no.6
• /
• pp.987-992
• /
• 2002

In an attempt to solve multiobjective optimization problems, many traditional methods scalarize the objective vector into a single objective. In those cases, the obtained solution is highly sensitive to the weight vector used in the scalarization process and demands the user to have knowledge about the underlying problem. Moreover, in solving multiobjective problems, designers may be interested in a set of Pareto-optimal points, instead of a single point. In this paper, pareto-based Continuous Evolutionary Algorithms for Multiobjective Optimization problems having continuous search space are introduced. This algorithm is based on Continuous Evolutionary Algorithms to solve single objective optimization problems with a continuous function and continuous search space efficiently. For multiobjective optimization, a progressive reproduction operator and a niche-formation method fur fitness sharing and a storing process for elitism are implemented in the algorithm. The operator and the niche formulation allow the solution set to be distributed widely over the Pareto-optimal tradeoff surface. Finally, the validity of this method has been demonstrated through a numerical example.

• Journal of applied mathematics & informatics
• /
• v.26 no.3_4
• /
• pp.549-561
• /
• 2008

In this paper, we construct a pair of Wolfe type second order symmetric dual problems, in which each component of the objective function contains support function and is, therefore, nondifferentiable. For this problem, we validate weak, strong and converse duality theorems under bonvexity - boncavity assumptions. A second order self duality theorem is also proved under additional appropriate conditions.

• Journal of Electrical Engineering and Technology
• /
• v.7 no.4
• /
• pp.636-645
• /
• 2012

This study introduces an information granular-based fuzzy radial basis function neural networks (FRBFNN) based on multiobjective optimization and weighted least square (WLS). An improved multiobjective space search algorithm (IMSSA) is proposed to optimize the FRBFNN. In the design of FRBFNN, the premise part of the rules is constructed with the aid of Fuzzy C-Means (FCM) clustering while the consequent part of the fuzzy rules is developed by using four types of polynomials, namely constant, linear, quadratic, and modified quadratic. Information granulation realized with C-Means clustering helps determine the initial values of the apex parameters of the membership function of the fuzzy neural network. To enhance the flexibility of neural network, we use the WLS learning to estimate the coefficients of the polynomials. In comparison with ordinary least square commonly used in the design of fuzzy radial basis function neural networks, WLS could come with a different type of the local model in each rule when dealing with the FRBFNN. Since the performance of the FRBFNN model is directly affected by some parameters such as e.g., the fuzzification coefficient used in the FCM, the number of rules and the orders of the polynomials present in the consequent parts of the rules, we carry out both structural as well as parametric optimization of the network. The proposed IMSSA that aims at the simultaneous minimization of complexity and the maximization of accuracy is exploited here to optimize the parameters of the model. Experimental results illustrate that the proposed neural network leads to better performance in comparison with some existing neurofuzzy models encountered in the literature.

• Journal of Korean Society of Forest Science
• /
• v.85 no.2
• /
• pp.172-179
• /
• 1996

This paper described the application of fuzzy multiobjective linear programming to solving a multiple-use problem of forest resources management. At first the concepts of linear programming, fuzzy linear programming and fuzzy multiobjective linear programming were introduced briefly. In order to illustrate a role of fuzzy multiobjective linear programming in the process of multiple-use forest planning, the natural recreation forest in Mt. Yoomyung was selected for this study. A fuzzy multiobjective linear programming model is formulated with data obtained from this Mt. Yoomyumg natural recreation forest to solve the multiple-use management planning problem of forest resources. Finally, the results, which were obtained from the calculation of this model, were discussed. The maximal value of the membership function(${\lambda}$) was 0.29, when the timber production and the forest recreation function were optimized at the same time through the fuzzy multiobjective linear programming. The cutting area in each period was 102.7ha, while total cutting area was 410.8ha for 4 periods. During 4 periods $57,904m^3$ will be harvested from this natural recreation forest and at the same time total visitors were estimated to be about 8.6 millions persons.

• Journal of applied mathematics & informatics
• /
• v.29 no.5_6
• /
• pp.1157-1165
• /
• 2011

The Karush-Kuhn-Tucker (KKT) necessary optimality conditions for nonlinear differentiable programming problems are also sufficient under suitable convexity assumptions. The KKT conditions in multiobjective programming problems with interval-valued objective and constraint functions are derived in this paper. The main contribution of this paper is to obtain the Pareto optimal solutions by resorting to the sufficient optimality condition.

• Journal of Korean Society of Steel Construction
• /
• v.9 no.3 s.32
• /
• pp.457-465
• /
• 1997

The Shape optimization makes it possible to reduce the weight of structure and cost then member sizing optimization. A vast amount of imprecise information is existed in constraints of the optimum design. It is very difficult and sometimes confusing to describe and to deal with the several criteria which contain fuzzy degrees of relatives importance. This paper proposed weighting strategies in the multiobjective shape optimization of fuzzy structural system by ${\alpha}$-cut approach. The algorithm in this research is numerically tested for 2-bar truss structure. The result show that. the user can choose the one optimum solution in practices as obtaining the optimum solutions according to the ${\alpha}$-cut approach, weight of volume and displacement.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.61
• /
• pp.41-50
• /
• 2000

Cell formation(CF) Is to group parts with similar geometry, function, material and process into part families, and the corresponding machines into machine cells. Cell formation solutions often contain exceptional elements(EEs). Also, the following objective functions - minimizing the total costs of dealing with exceptional elements and maximizing total similarity coefficients between parts - have been used in CF modeling. Thus, multiobjective programming approach can be developed to model cell formation problems with two conflicting objective functions. This paper presents an effective cell formation method with fuzzy multiobjective nonlinear mixed-integer programming simultaneously to form machine cells and to minimize the cost of eliminating EEs.

• Journal of Advanced Marine Engineering and Technology
• /
• v.24 no.3
• /
• pp.99-105
• /
• 2000

For multi-body dynamic problems. especially coalescent eigenvalue problems with multiobjective optimization, the design sensitivity analysis is too much complicated mathematically and numerically. Therefore, this article proposes a new technique for structural dynamic modification using a mode modification and homologous structures design method with Genetic Algorithm(GA). In this work, the homologous structure of the resiliently mounted multi-body for marine diesel generator systems is studied and the problem is treated as a combinational optimization problem using the GA. In GA formulation, fitness is defined based on penalty function approach. That include homology, allowable stress and minimum weight of common plate.

• Journal of the Korean Society for Precision Engineering
• /
• v.17 no.5
• /
• pp.93-99
• /
• 2000

In many dynamic structural optimization problems, the goal is to reduce the total weight of the structure without causing the resonance. Up to now, gradient informations(i.e., design sensitivity) have been used to achieve the goal. For some class of dynamic problems, especially coalescent eigenvalue Problems with multiobjective optimization, the design sensitivity analysis is too much complicated mathematically and numerically. Therefore, this article proposes a new technique fur structural dynamic modification using a mode modification method with Genetic Algorithm(GA). In GA formulation, fitness is defined based on penalty function approach. Design variables are iteratively improved by using genetic algorithm. Two numerical examples are shown, (ⅰ) a cantilevered plate, and (ⅱ) H-shaped structure. The results demonstrate that the proposed method is highly efficient.