• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.20 no.6
• /
• pp.33-43
• /
• 2021

A60 class bulkhead penetration piece is a fire resistance system installed on a bulkhead compartment to protect lives and to prevent flame diffusion in a fire accident on a ship and offshore plant. This study focuses on the approximate optimization of the fire resistance design of the A60 class bulkhead penetration piece using a multi-island genetic algorithm. Transient heat transfer analysis was performed to evaluate the fire resistance design of the A60 class bulkhead penetration piece. For approximate optimization, the bulkhead penetration piece length, diameter, material type, and insulation density were considered discrete design variables; moreover, temperature, cost, and productivity were considered constraint functions. The approximate optimum design problem based on the meta-model was formulated by determining the discrete design variables by minimizing the weight of the A60 class bulkhead penetration piece subject to the constraint functions. The meta-models used for the approximate optimization were the Kriging model, response surface method, and radial basis function-based neural network. The results from the approximate optimization were compared to the actual results of the analysis to determine approximate accuracy. We conclude that the radial basis function-based neural network among the meta-models used in the approximate optimization generates the most accurate optimum design results for the fire resistance design of the A60 class bulkhead penetration piece.

• Journal of the Korean Institute of Telematics and Electronics T
• /
• v.35T no.2
• /
• pp.48-55
• /
• 1998

The main drawback of filtered-X LMS(FXLMS) algorithm for the ANC of broadband noises is its low convergence speed when the filtered reference signals are strongly correlated, producing a large eigenvalue ratio in correlation matrix. This correlation can be caused either by autocorrelation of the signals of the reference sensors, or by coupling between the error path which introduces intercorrelation in the filtered reference signals. In this paper, we introduce a transform domain FXLMS(TD-FXLMS) algorithm that has a high convergence speed by orthogonal transform's decorrelation properties.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.1
• /
• pp.87-91
• /
• 2005

The cascade-correlation (CC) learning algorithm of Fahlman and Lebiere is one of the most influential constructive algorithm in a neural network. Cascading the hidden neurons results in a network that can represent very strong nonlinearities. Although this power is in principle useful, it can be a disadvantage if such strong nonlinearity is not required to solve the problem. 3 models are presented and compared empirically. All of them are based on valiants of the cascade architecture and output neurons weights training of the CC algorithm. Empirical results indicate the followings： (1) In the pattern classification, the model that train only new hidden neuron to output layer connection weights shows the best predictive ability; (2) In the function approximation, the model that removed input-output connection and used sigmoid-linear activation function is better predictability than CasCor algorithm.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.4
• /
• pp.499-506
• /
• 2012

In this paper, we analyze the effects of phase noise and frequency offset that cause performance degradation. Basically, we like to propose reduced PNFS(Phase Noise and Frequency offset Suppression) algorithm. The OFDM system is seriously affected by ICI component such as phase noise, frequency offset and Doppler effects. Especially, complicated processing algorithm with high complexity was required it in order to compensate those ICI components. So, we propose PNFS algorithm that can decrease complexity and compensate ICI components. We propose a method decreased complexity by approximation of parameters that affect slightly performance change and compare the quantity of conventional and revised PNFS algorithm. Also, simulation shows that BER performance of revised PNFS algorithm can be improved slightly.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.1 s.33
• /
• pp.181-188
• /
• 2005

Chinese Postman Problem (CPP) is a problem that finds a shortest tour traversing all edges or arcs at least once in a given network. The Chinese Postman Problem on Mixed networks (MCPP) is a Practical generalization of the classical CPP and it has many real-world applications. The MCPP has been shown to be NP-complete. In this paper, we transform a mixed network into a symmetric network using virtual arcs that are shortest paths by Floyd's algorithm. With the transformed network, we propose a Genetic Algorithm (GA) that converges to a near optimal solution quickly by a multi-directional search technique. We study the chromosome structure used in the GA and it consists of a path string and an encoding string. An encoding method, a decoding method, and some genetic operators that are needed when the MCPP is solved using the Proposed GA are studied. . In addition, two scaling methods are used in proposed GA. We compare the performance of the GA with an existing Modified MDXED2 algorithm (Pearn et al. , 1995) In the simulation results, the proposed method is better than the existing methods in case the network has many edges, the Power Law scaling method is better than the Logarithmic scaling method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.6
• /
• pp.792-798
• /
• 2021

In this paper, an algorithm for determining the coagulant input rate in the drug-injection tank during the process of the water purification plant was derived through big data analysis and prediction based on artificial intelligence. In addition, analysis of big data technology and AI algorithm application methods and existing academic and technical data were reviewed to analyze and review application cases in similar fields. Through this, the goal was to develop an algorithm for determining the coagulant input rate and to present the optimal input rate through autonomous driving simulator and pilot operation of the coagulant input process. Through this study, the coagulant injection rate, which is an output variable, is determined based on various input variables, and it is developed to simulate the relationship pattern between the input variable and the output variable and apply the learned pattern to the decision-making pattern of water plant operating workers.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.6
• /
• pp.511-516
• /
• 2002

The neural networks may have problem such that the amount of calculation for the network learning goes too big according to the dimension of the dimension. To overcome this problem, the wavelet neural networks(WNN) which use the orthogonal basis function in the hidden node are proposed. One can compose wavelet functions as activation functions in the WNN by determining the scale and center of wavelet function. In this paper, when we compose the WNN using wavelet functions, we set a single scale function as a node function together. We intend that one scale function approximates the target function roughly, the other wavelet functions approximate it finely During the determination of the parameters, the wavelet functions can be determined by the global search for solutions suitable for the suggested problem using the genetic algorithm and finally, we use the back-propagation algorithm in the learning of the weights.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.6
• /
• pp.47-52
• /
• 2002

The wavelet functions are originated from scaling functions and can be used as activation function in the hidden node of the network by deciding two parameters such as scale and center. In this paper, we would like to propose the mixed structure. When we compose the WNN using wavelet functions, we propose to set a single scale function as a node function together. The properties of the proposed structure is that while one scale function approximates the target function roughly, the other wavelet functions approximate it finely. During the determination of the parameters, the wavelet functions can be determined by the global search algorithm such as genetic algorithm to be suitable for the suggested problem. Finally, we use the back-propagation algorithm in the learning of the weights.

• Geophysics and Geophysical Exploration
• /
• v.10 no.2
• /
• pp.131-137
• /
• 2007

With consideration of the first arrival picking methodology and inherent errors in picking process, I propose, from the computerization point of view, a practical algorithm for picking and error computation. The proposed picking procedure consists of 2-step; 1) picking the first coherent peak or trough events, 2) derive a line which approximates the record in the interval prior to the pick, and set the intercept time of the line as the first break. The length of fitting interval used in experiment, is few samples less than 1/4 width of the arriving wavelet. A quantitative measure of the error involved in first arrival picking is defined as the time length that needed to determine if an event is the first arrival or not. The time length is expressed as a function of frequency bandwidth of the signal and the S/N ratio. For 3 sets of cross-well seismic data, first breaks are picked twice, by manually, and by the proposed method. And at the same time, the error bound for each trace is computed. Experiment results show that good performance of the proposed picking method, and the usefulness of the quantitative error measure in pick-quality evaluation.

• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1054-1061
• /
• 2003

Q-Learning, a representative algorithm of reinforcement learning, experiences repeatedly until estimation values about all state-action pairs of state space converge and achieve optimal policies. When the state space is high dimensional or continuous, complex reinforcement learning tasks involve very large state space and suffer from storing all individual state values in a single table. We introduce Q-Map that is new function approximation method to get classified policies. As an agent learns on-line, Q-Map groups states of similar situations and adapts to new experiences repeatedly. State-action pairs necessary for fine control are treated in the form of rule. As a result of experiment in maze environment and mountain car problem, we can achieve classified knowledge and extract easily rules from Q-Map