• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.5
• /
• pp.660-665
• /
• 2004

In this study, a new architecture and comprehensive design methodology of genetically optimized Multi-layer Fuzzy Neural Networks (gMFNN) are introduced and a series of numeric experiments are carried out. The gMFNN architecture results from a synergistic usage of the hybrid system generated by combining Fuzzy Neural Networks (FNN) with Polynomial Neural Networks (PNN). FNN contributes to the formation of the premise part of the overall network structure of the gMFNN. The consequence part of the gMFNN is designed using PNN. The optimization of the FNN is realized with the aid of a standard back-propagation learning algorithm and genetic optimization. The development of the PNN dwells on the extended Group Method of Data Handling (GMDH) method and Genetic Algorithms (GAs). To evaluate the performance of the gMFNN, the models are experimented with the use of a numerical example.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1996.10a
• /
• pp.234-238
• /
• 1996

In this paper, Plasma-Enhanced Chemical Vapor Deposition (PECVD) modeling using Polynomial Neural Networks (PNN) has been introduced. The deposition of SiO2 was characterized via a 25-1 fractional factorial experiment, was used to train PNNs using predicted squared error (PSE). The optimal neural network structure and learning parameters were determined by means of a second fractional factorial experiment. The optimized networks minimized both learning and prediction error. From these PNN process models, the effect of deposition conditions on film properties has been studied. The deposition experiments were carried out in a Plasma Therm 700 series PECVD system. The models obtained will ultimately be used for several other manufacturing applications, including recipe synthesis and process control.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.55 no.6
• /
• pp.264-273
• /
• 2006

In this paper, we propose a new architecture of Self-Organizing Fuzzy Polynomial Neural Networks (SOFPNN) by means of consecutive optimization and also discuss its comprehensive design methodology involving mechanisms of genetic optimization. The network is based on a structurally as well as parametrically optimized fuzzy polynomial neurons (FPNs) conducted with the aid of information granulation and genetic algorithms. In structurally identification of FPN, the design procedure applied in the construction of each layer of a SOFPNN deals with its structural optimization involving the selection of preferred nodes (or FPNs) with specific local characteristics and addresses specific aspects of parametric optimization. In addition, the fuzzy rules used in the networks exploit the notion of information granules defined over system's variables and formed through the process of information granulation. That is, we determine the initial location (apexes) of membership functions and initial values of polynomial function being used in the premised and consequence part of the fuzzy rules respectively. This granulation is realized with the aid of the hard c-menas clustering method (HCM). For the parametric identification, we obtained the effective model that the axes of MFs are identified by GA to reflect characteristic of given data. Especially, the genetically dynamic search method is introduced in the identification of parameter. It helps lead to rapidly optimal convergence over a limited region or a boundary condition. To evaluate the performance of the proposed model, the model is experimented with using two time series data(gas furnace process, nonlinear system data, and NOx process data).

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.12
• /
• pp.3063-3072
• /
• 1993

The layout is an important and difficult problem in industrial applications like sheet metal manufacturing, garment making, circuit layout, plant layout, and land development. The module layout problem is known to be non-deterministic polynomial time complete(NP-complete). To efficiently find an optimal layout from a large number of candidate layout configuration a heuristic algorithm could be used. In recent years, a number of researchers have investigated the combinatorial optimization problems by using neural network principles such as traveling salesman problem, placement and routing in circuit design. This paper describes the application of Self-organizing Feature Maps(SOM) of the Kohonen network and Simulated Annealing Algorithm(SAA) to the layout problem of the two-dimensional rectangular shapes.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.846-851
• /
• 2005

In this paper, we propose Genetic Algorithms(GAs)-based Optimized Polynomial Neural Networks(PNN). The proposed algorithm is based on Group Method of Data Handling(GMDH) method and its structure is similar to feedforward Neural Networks. But the structure of PNN is not fixed like in conventional neural networks and can be generated in a dynamic manner. As each node of PNN structure, we use several types of high-order polynomial such as linear, quadratic and modified quadratic, and it is connected as various kinds of multi-variable inputs. The conventional PNN depends on the experience of a designer that select the number of input variables, input variable and polynomial type. Therefore it is very difficult to organize optimized network. The proposed algorithm leads to identify and select the number of input variables, input variable and polynomial type by using Genetic Algorithms(GAs). The aggregate performance index with weighting factor is proposed as well. The study is illustrated with tile NOx omission process data of gas turbine power plant for application to nonlinear process. In the sequel the proposed model shows not only superb predictability but also high accuracy in comparison to the existing intelligent models.

• Journal of the Korean earth science society
• /
• v.32 no.6
• /
• pp.665-673
• /
• 2011

In order to estimate the hydrocarbon reserves, the porosity of the reservoir must be determined. The porosity of the area without a well is generally calculated by extrapolating the porosity logs measured at wells. However, if not only well logs but also seismic data exist on the same site, the more accurate pseudo porosity log can be obtained through artificial neural network technique by extracting the relations between the seismic data and well logs at the site. In this study, we have developed a module which creates pseudo porosity logs by using the polynomial neural network method. In order to obtain more accurate pseudo porosity logs, we selected the seismic attributes which have high correlation values in the correlation analysis between the seismic attributes and the porosity logs. Through the training procedure between selected seismic attributes and well logs, our module produces the correlation weights which can be used to generate the pseudo porosity log in the well free area. To verify the reliability and the applicability of the developed module, we have applied the module to the field data acquired from F3 Block in the North Sea and compared the results to those from the probabilistic neural network method in a commercial program. We could confirm the reliability of our module because both results showed similar trend. Moreover, since the pseudo porosity logs from polynomial neural network method are closer to the true porosity logs at the wells than those from probabilistic method, we concluded that the polynomial neural network method is effective for the data sets with insufficient wells such as F3 Block in the North Sea.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.4
• /
• pp.832-839
• /
• 2009

In this paper, we introduce an design of fuzzy relation-based fuzzy neural networks with multi-output. Fuzzy relation-based fuzzy neural networks comprise the network structure generated by dividing the entire input space. The premise part of the fuzzy rules of the network reflects the relation of the division space for the entire input space and the consequent part of the fuzzy rules expresses three types of polynomial functions such as constant, linear, and modified quadratic. For the multi-output structure the neurons in the output layer were connected with connection weights. The learning of fuzzy neural networks is realized by adjusting connections of the neurons both in the consequent part of the fuzzy rules and in the output layer, and it follows a back-propagation algorithm. In addition, in order to optimize the network, the parameters of the network such as apexes of membership functions, learning rate and momentum coefficient are automatically optimized by using real-coded genetic algorithm. Two examples are included to evaluate the performance of the proposed network.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.3
• /
• pp.970-977
• /
• 2014

In this paper, the artificial neural network is used to predict the junction temperature of the IGBT power module, by measuring the temperature sensitive electrical parameters (TSEP) of the module. An experiment circuit is built to measure saturation voltage drop and collector current under different temperature. In order to solve the nonlinear problem of TSEP approach as a junction temperature evaluation method, a Back Propagation (BP) neural network prediction model is established by using the Matlab. With the advantages of non-contact, high sensitivity, and without package open, the proposed method is also potentially promising for on-line junction temperature measurement. The Matlab simulation results show that BP neural network gives a more accuracy results, compared with the method of polynomial fitting.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1406-1420
• /
• 2017

Artificial neural networks (ANNs) play an important role in the fields of function approximation, prediction, and classification. ANN performance is critically dependent on the input parameters, including the number of neurons in each layer, and the optimal values of weights and biases assigned to each neuron. In this study, we apply the particle swarm optimization method, a popular optimization algorithm for determining the optimal values of weights and biases for every neuron in different layers of the ANN. Several regression models, including general linear regression, Fourier regression, smoothing spline, and polynomial regression, are conducted to evaluate the proposed method's prediction power compared to multiple linear regression (MLR) methods. In addition, residual analysis is conducted to evaluate the optimized ANN accuracy for both training and test datasets. The experimental results demonstrate that the proposed method can effectively determine optimal values for neuron weights and biases, and high accuracy results are obtained for prediction applications. Evaluations of the proposed method reveal that it can be used for prediction and estimation purposes, with a high accuracy ratio, and the designed model provides a reliable technique for optimization. The simulation results show that the optimized ANN exhibits superior performance to MLR for prediction purposes.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.861-868
• /
• 2023

One of the key technologies in providing data analysis in the deep learning while maintaining security is fully homomorphic encryption. Due to constraints in operations on fully homomorphically encrypted data, non-arithmetic functions used in deep learning must be approximated by polynomials. Until now, the degrees of approximation polynomials with composite minimax polynomials have been uniformly set across layers, which poses challenges for effective network designs on fully homomorphic encryption. This study theoretically proves that setting different degrees of approximation polynomials constructed by composite minimax polynomial in each layer does not pose any issues in the inference on convolutional neural networks.