• Journal of Korean Association for Spatial Structures
• /
• v.20 no.2
• /
• pp.51-58
• /
• 2020

Recently, deep learning that is the most popular and effective class of machine learning algorithms is widely applied to various industrial areas. A number of research on various topics about structural engineering was performed by using artificial neural networks, such as structural design optimization, vibration control and system identification etc. When nonlinear semi-active structural control devices are applied to building structure, a lot of computational effort is required to predict dynamic structural responses of finite element method (FEM) model for development of control algorithm. To solve this problem, an artificial neural network model was developed in this study. Among various deep learning algorithms, a recurrent neural network (RNN) was used to make the time history response prediction model. An RNN can retain state from one iteration to the next by using its own output as input for the next step. An eleven-story building structure with semi-active tuned mass damper (TMD) was used as an example structure. The semi-active TMD was composed of magnetorheological damper. Five historical earthquakes and five artificial ground motions were used as ground excitations for training of an RNN model. Another artificial ground motion that was not used for training was used for verification of the developed RNN model. Parametric studies on various hyper-parameters including number of hidden layers, sequence length, number of LSTM cells, etc. After appropriate training iteration of the RNN model with proper hyper-parameters, the RNN model for prediction of seismic responses of the building structure with semi-active TMD was developed. The developed RNN model can effectively provide very accurate seismic responses compared to the FEM model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.7
• /
• pp.862-868
• /
• 2007

The planar motion of the index finger in general human hands is usually implemented by the actuation of three joints. This task requires a technique to determine the joint combination for each fingertip position which is well-known as the inverse kinematics problem in robotics. Especially, it is an essential work for grasping and manipulation tasks by robotic and humanoid fingers. In this paper, an intelligent neural learning scheme for solving such inverse kinematics is presented. Specifically, a multi-layered neural network is utilized for effective inverse kinematics, where a dynamic neural learning algorithm is employed for fast learning. Also, a bio-mimetic feature of general human fingers is incorporated to the learning scheme. The usefulness of the proposed approach is verified by simulations.

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.254-257
• /
• 1993

This paper presents a new method to implement Hebbian learning method on artificial neural network. In hebbian learning algorithm, complexity in terms of multiplications is high. To save the chip area, we consider a new learning circuit. By calculating similarity, or correlation between $X_i$ and $O_i$, large portion of circuits commonly used in conventional neural networks is not necessary for this new hebbian learning circuit named COR. The output signals of COR is applied to weight storage capacitors for direct control the voltages of the capacitors. The weighted sum, ${\Sigma}W_{ij}O_j$, is realized by multipliers, whose output currents are summed up in one line which goes to learning circuit or output circuit. The drain current of the multiplier can produce positive or negative synaptic weights. The pass transistor selects eight learning mode or recall mode. The layout of an learnable six-neuron fully connected Hopfield neural network is designed, and is simulated using PSPICE. The network memorizes, and retrieves the patterns correctly under the existence of minor noises.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.266-272
• /
• 1994

An essential ingredient in precision machining is a positioning system that responds quickly and precisely to very small input signal. In this paper, two different positioning systems were presented fot the precision positioning control. The one is a friction drive system, the other is a ballscrew system. The friction drive system was composed of an air sliding guide and a friction drive. The ballscrew system was made of a ballscrew and a linear guide. Nonlinear behaviors of the given systems tend to make the system inaccurate. The paper looked at the phenomena that has caused the positioning error. These apparently nonlinear phenomena can be attributed mainly to the presence of the nonlinear friction and slip effect plus the dynamic change from the microdynamic to the macrodynamic and form the macrodynamic to the microdynamic. For the control of the positioning system, the control algorithm based on a neural network is suggested. The FEL(Feedback Error Learning) controller can learn the inverse dynamics of a nonlinear system by using the neural network controller, and stabilize the system by a linear controller. In the experiment, PTP control is implemented withen the maximum error of 0.05 .mu.m ~0.1 .mu. m when i .mu.m step reference input is applied and that of maximum 1 .mu. m when 100 .mu.m step reference input is given. Sinusoidal inputs with the amplitude of 1 .mu.m and 100 .mu. m are used for the tracking control of the positioning system. Experimental results of the proposed algorithm are shown to be superior to those of conventional PD controls.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.5
• /
• pp.414-421
• /
• 2007

This paper presents a engine/brake integrated VDC(Vehicle Dynamic Control) system using neural network algorithm methods for wheel slip and yaw rate control. For stable performance of vehicle, not only is the lateral motion control(wheel slip control) important but the yaw motion control of the vehicle is crucial. The proposed NNPI(Neural Network Proportional-Integral) controller operates at throttle angle to improve the performance of wheel slip. Also, the suggested NNPID controller performs at brake system to improve steering performance. The proposed controller consists of multi-hidden layer neural network structure and PID control strategy for self-learning of gain scheduling. Computer Simulation have been performed to verify the proposed neural network based control scheme of 17 dof vehicle dynamic model which is implemented in MATLAB Simulink.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.442-445
• /
• 1997

We present a hybrid self-tuning method of fuzzy inference systems with hyper elliptic Gaussian membership functions using genetic algorithm(GA) and back-propagation algorithm. The proposed self-tuning method has two phases : one is the coarse tuning process based on GA and the other is the fine tuning process based on back-propagation. But the parameters which is obtained by a GA are near optimal solutions. In order to solve the problem in GA applications, it uses a back-propagation algorithm, which is one of learning algorithms in neural networks, to finely tune the parameters obtained by a GA. We provide Box-Jenkins time series to evaluate the advantage and effectiveness of the proposed approach and compare with the conventional method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.10a
• /
• pp.252-256
• /
• 1997

The success of a fuzzy neural network(FNN) control system solving any given problem critically depends on the architecture of the network. Various attempts have been made in optimizing its structure using genetic algorithm automated designs. This paper presents a new approach to structurally optimized designs of FNN models. A messy genetic algorithm is used to obtain structurally optimized FNN models. Structural optimization is regarded important before neural networks based learning is switched into. We have applied the method to the problem of a numerical approximation

• Proceedings of the IEEK Conference
• /
• 2002.06c
• /
• pp.9-12
• /
• 2002

This paper proposes a dynamic moment algorithm to control oscillaion before the convergence of the KR(Kernel Relaxation). The proposed dynamic moment algorithm can be controlled to convergence speed and performance according to the change of the dynamic moment by teaming training. we used SONAR data that is a neural network classifier standard evaluation data in order to do impartial performance evaluation. The proposed algorithm has been applied to the KP (kernel perceptron), KPM(kernel perceptron with margin) and KLMS(kernel lms) as the kernel method presented recently. The simulation results of proposed algorithm have better the convergence performance than those using none and static moment.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.11
• /
• pp.1074-1081
• /
• 2007

This paper presents a method for real-time face detection and tracking which combined Adaboost and Camshift algorithm. Adaboost algorithm is a method which selects an important feature called weak classifier among many possible image features by tuning weight of each feature from learning candidates. Even though excellent performance extracting the object, computing time of the algorithm is very high with window size of multi-scale to search image region. So direct application of the method is not easy for real-time tasks such as multi-task OS, robot, and mobile environment. But CAMshift method is an improvement of Mean-shift algorithm for the video streaming environment and track the interesting object at high speed based on hue value of the target region. The detection efficiency of the method is not good for environment of dynamic illumination. We propose a combined method of Adaboost and CAMshift to improve the computing speed with good face detection performance. The method was proved for real image sequences including single and more faces.

• Journal of Satellite, Information and Communications
• /
• v.6 no.2
• /
• pp.131-135
• /
• 2011

In this paper, we describe the design and implementation of wireless smart plug having AC power sensor and intelligent standby power control algorithm for energy sensor network. The adaptive standby power control algorithm has function to apply different threshold of standby power by using learning algorithm depending on electric equipments. As using the proposed algorithm, user convenience will be more better and power consumption can be more reduced. The implemented prototypes of wireless smart plug and wireless access point were tested to verify the required functions and performance. As a result, we confirmed practicality of wireless smart power sensor and satisfaction of given design specifications.