• Journal of the Korean Society of Visualization
• /
• v.14 no.1
• /
• pp.57-61
• /
• 2016

We report a finding of fast(exceeding 2,000 K/s) heating of polydimethylsiloxane(PDMS), one of the most commonly-used microchannel materials, under cyclic loadings at high(~MHz) frequencies. A microheater was created based on the finding. The heating mechanism utilized vibration damping of sound waves, which were generated and precisely manipulated using a conventional surface acoustic wave(SAW) microfluidic system, in PDMS. The penetration depths were measured to range from $210{\mu}m$ to $1290{\mu}m$, enough to cover most microchannel heights in microfluidic systems. The energy conversion efficiency was SAW frequency-dependent and measured to be the highest at around 30 MHz. Independent actuation of each interdigital transducer(IDT) enabled independent manipulation of SAWs, permitting spatiotemporal control of temperature on the microchip. All the advantages of this microheater facilitated a two-step continuous flow polymerase chain reaction(CFPCR) to achieve the billion-fold amplification of a 134 bp DNA amplicon in less than 3 min. In addition, a technique was developed for establishing dynamic free-form temperature gradients(TGs) in PDMS as well as in gases in contact with the PDMS.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.1
• /
• pp.32-42
• /
• 2000

This paper presents an on-line learning algorithm for sequential construction of generalized radial basis function networks (GRBFNs) to model nonlinear systems from empirical data. The GRBFN, an extended from of standard radial basis function (RBF) networks with constant weights, is an architecture capable of representing nonlinear systems by smoothly integrating local linear models. The proposed learning algorithm has a two-stage learning scheme that performs both structure learning and parameter learning. The structure learning stage constructs the GRBFN model using two construction criteria, based on both training error criterion and Mahalanobis distance criterion, to assign new hidden units and the linear local models for given empirical training data. In the parameter learning stage the network parameters are updated using the gradient descent rule. To evaluate the modeling performance of the proposed algorithm, simulations and their results applied to two well-known benchmarks are discussed.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.2
• /
• pp.194-202
• /
• 2003

In this paper, we introduce a category of Multi-FNN (Fuzzy-Neural Networks) models, analyze the underlying architectures and propose a comprehensive identification framework. The proposed Multi-FNNs dwell on a concept of fuzzy rule-based FNNs based on HCM clustering and evolutionary fuzzy granulation, and exploit linear inference being treated as a generic inference mechanism. By this nature, this FNN model is geared toward capturing relationships between information granules known as fuzzy sets. The form of the information granules themselves (in particular their distribution and a type of membership function) becomes an important design feature of the FNN model contributing to its structural as well as parametric optimization. The identification environment uses clustering techniques (Hard C - Means, HCM) and exploits genetic optimization as a vehicle of global optimization. The global optimization is augmented by more refined gradient-based learning mechanisms such as standard back-propagation. The HCM algorithm, whose role is to carry out preprocessing of the process data for system modeling, is utilized to determine the structure of Multi-FNNs. The detailed parameters of the Multi-FNN (such as apexes of membership functions, learning rates and momentum coefficients) are adjusted using genetic algorithms. An aggregate performance index with a weighting factor is proposed in order to achieve a sound balance between approximation and generalization (predictive) abilities of the model. To evaluate the performance of the proposed model, two numeric data sets are experimented with. One is the numerical data coming from a description of a certain nonlinear function and the other is NOx emission process data from a gas turbine power plant.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.6
• /
• pp.1381-1388
• /
• 1993

This paper presents a method for the simultaneous optimal design of structural and control systems. Sensitivities of performance index with respect to structural design variables are analyzed. The structural design variables are optimized to minimize the performance index by use of conjugate gradient method. The method is applied to a half model of an active vehicle suspension system with elastic body moving on a randomly profiled road. The suspension control force of an optimally controlled system in the presence of measurement errors are calculated by use of linear quadratic Gaussian control theory and Kalman filter theory. The performance index contains ride comfort, road holding and working space of suspension. The structural design variables taken are stiffness, daming properties and the position of the suspension system. The random road profile considered as colored noise is shaped from white noise by use of shaping filter. The performance of an optimal simultaneous structure/control system is compared with that of an optimal controlled system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.5
• /
• pp.592-601
• /
• 2016

The recurrent neural network (RNN) is a deep learning model which is suitable to sequential or length-variable data. The Long Short-Term Memory (LSTM) mitigates the vanishing gradient problem of RNNs so that LSTM can maintain the long-term dependency among the constituents of the given input sequence. In this paper, we propose a LSTM based language model which can predict following words of a given incomplete sentence to generate a complete sentence. To evaluate our method, we trained our model using multiple Korean corpora then generated the incomplete part of Korean sentences. The result shows that our language model was able to generate the fluent Korean sentences. We also show that the word based model generated better sentences compared to the other settings.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.893-898
• /
• 2005

Due to the merits that only a small amount of computation is needed for solutions and stochastic policies can be handled explicitly, the actor-critic algorithm, which is a class of reinforcement learning methods, has recently attracted a lot of interests in the area of artificial intelligence. The actor-critic network composes of tile actor network for selecting control inputs and the critic network for estimating value functions, and in its training stage, the actor and critic networks take the strategy, of changing their parameters adaptively in order to select excellent control inputs and yield accurate approximation for value functions as fast as possible. In this paper, we consider a new actor-critic algorithm employing an RLS(Recursive Least Square) method for critic learning, and policy gradients for actor learning. The applicability of the considered algorithm is illustrated with experiments on the two linked robot arm.

• Korean Journal of Animal Reproduction
• /
• v.17 no.4
• /
• pp.271-278
• /
• 1994

The aim of this study was to compare the two culture systems 1) co-culture with cumulus cells and 2) chemically defined medium supplemented with amino acids (CR1aa) and fetal calf serum (FCS) of in vitro produced bovine embryos from follicular oocytes in vitro. Bovine follicular oocytes were collected from ovaries of slaughtered cows and matured in TCM199 supplemented with 10% FCS and hormones (1$\mu\textrm{g}$/ml FSH-P and 1$\mu\textrm{g}$/ml oestradiol-17$\beta$)24 hours at 39$^{\circ}C$ under 5% CO2 in air. The capacitation of spermatozoa from ejaculated or frozen bull semen was induced by centrifugation through Percoll density gradient (45%, 90%). Then capacitated spermatozoa (1$\times$106/ml) were inseminated into 50${mu}ell$ droplet containing matured follicular oocytes and incubated for 40~42 hours. Cleaved embryos of 2~4cell stage were transferred to the co-culture with cumulus cells and/or CR1aa medium supplemented with FCS. In semen source, the developmental rates to the blastocyst and the hatched blastocyst stages were higher in ejaculated semen(27.6% and 14.9%) than those of frozen-thawed semen(18.3% and 11.8%), respectively. In two culture systems, the proportions of embryonic development upto the blastocysts and the hatched blastocysts were higher of CR1aa medium (22.1% and 12.1%) than those of cumulus cell co-culture (16.8% and 5.1%), respectively. The number of cells in exapnded blastocysts was slightly higher in cumulus cells co-culture (122.6$\pm$8.5) than that in CR1aa medium (117.9$\pm$5.9). The present results indicated that the early development of in vitro produced bovine embryos can be maintained efficiently in CR1aa medium as well as in co-culture with cumulus cells.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.457-466
• /
• 2005

In this paper, we propose a wavelet neural network(WNN) based predictive control method for path tracking of mobile robots with multi-input and multi-output. In our control method, we use a WNN as a state predictor which combines the capability of artificial neural networks in learning processes and the capability of wavelet decomposition. A WNN predictor is tuned to minimize errors between the WNN outputs and the states of mobile robot using the gradient descent rule. And control signals, linear velocity and angular velocity, are calculated to minimize the predefined cost function using errors between the reference states and the predicted states. Through a computer simulation for the tracking performance according to varied track, we demonstrate the efficiency and the feasibility of our predictive control system.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.2
• /
• pp.206-211
• /
• 2008

Trainee can enhance his perception of and interaction with the real world by displayed virtual objects in simulation using image processing technology. Therefore, it is essential for realistic simulation to determine the occlusion areas of the virtual object produces after registering real image and virtual object exactly. In this paper, we proposed the new method to solve occlusions which happens during virtual target moves according to the simulated route on real image using improved object contour extraction by neighboring edge search and picking algorithm. After we acquire the detailed contour of complex objects by proposed contour extraction algorithm, we extract the three dimensional information of the position happening occlusion by using MER for performance improvement. In the experiment, we compared proposed method with existed method and preyed the effectiveness in the environment which a partial occlusions happens.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.5
• /
• pp.431-437
• /
• 2006

In this paper vector code correlation(VCC) method and an algorithm to promote the image-processing performance in building an effective measurement system using cameras are described far automatically berthing and controlling the ship equipped with side-thrusters. In order to realize automatic ship berthing, it is indispensable that the berthing assistant system on the ship should continuously trace a target in the berth to measure the distance to the target and the ship attitude, such that we can make the ship move to the specified location. The considered system is made up of 4 apparatuses compounded from a CCD camera, a camera direction controller, a popular PC with a built-in image processing board and a signal conversion unit connected to parallel port of the PC. The object of this paper is to reduce the image-processing time so that the berthing system is able to ensure the safety schedule against risks during approaching to the berth. It could be achieved by composing the vector code image to utilize the gradient of an approximated plane found with the brightness of pixels forming a certain region in an image and verifying the effectiveness on a commonly used PC. From experimental results, it is clear that the proposed method can be applied to the measurement system for automatic ship berthing and has the image-processing time of fourfold as compared with the typical template matching method.