• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.139.4-139
• /
• 2001

The thesis propose the sensorless vector control method that estimates the rotor speed using stator current. The estimated speed is used as feedback in a vector control system. The conventional MRAS structure has a problem the error output is decreasing as estimated speed error is increasing and the estimation performance is not robust when mutual inductance has been changed. In the proposed method, error output is proportional to estimated speed error. The described technique is less complex, robust to variations of mutual inductance. This new method can achieve much wider bandwidth speed control than that of the conventional MRAS structure.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.34.2-34
• /
• 2001

There are many control methods guaranteeing the robustness of systems. Among them, H$\infty$ control, sliding mode control and disturbance observer have been used widely. Especially, disturbance observer(DOB) is one of the most popular methods because it is easy to apply and the cost to pay is low due to its simplicity. Conventional DOB utilizes output signal as a feedback signal. But in Optical Disk Drive(ODD) systems, the Position Error Signal(PES) is the only available one. So conventional DOB is not applicable. If we use error signal in stead of output signal, another form of DOB is made. We call it as Error based Disturbance Observer(EDOB). We show in this paper the difference between two systems, namely conventional DOB system and the EDOB system, and also show the effectiveness of EDOB through experiment.

• Proceedings of the KIEE Conference
• /
• 1989.07a
• /
• pp.527-530
• /
• 1989

In order to reduce harmonic contents in the inverter output, various PWM ( pulse with modulation ) techniques have been utilized. This paper suggests a single phase inverter with 4-level output and its PWM control scheme. The proposed 4-level output inverter shows better performence than the 3-level output one in respects of the current error and the switching frequency.

• ETRI Journal
• /
• v.15 no.2
• /
• pp.35-51
• /
• 1993

This paper presents function approximation based on nonparametric estimation. As an estimation model of function approximation, a three layered network composed of input, hidden and output layers is considered. The input and output layers have linear activation units while the hidden layer has nonlinear activation units or kernel functions which have the characteristics of bounds and locality. Using this type of network, a many-to-one function is synthesized over the domain of the input space by a number of kernel functions. In this network, we have to estimate the necessary number of kernel functions as well as the parameters associated with kernel functions. For this purpose, a new method of parameter estimation in which linear learning rule is applied between hidden and output layers while nonlinear (piecewise-linear) learning rule is applied between input and hidden layers, is considered. The linear learning rule updates the output weights between hidden and output layers based on the Linear Minimization of Mean Square Error (LMMSE) sense in the space of kernel functions while the nonlinear learning rule updates the parameters of kernel functions based on the gradient of the actual output of network with respect to the parameters (especially, the shape) of kernel functions. This approach of parameter adaptation provides near optimal values of the parameters associated with kernel functions in the sense of minimizing mean square error. As a result, the suggested nonparametric estimation provides an efficient way of function approximation from the view point of the number of kernel functions as well as learning speed.

• Journal of Power Electronics
• /
• v.18 no.2
• /
• pp.533-546
• /
• 2018

This study proposes a comprehensive mathematical model that includes coil-system circuit and loss models for power converters in wireless power transfer (WPT) systems. The proposed model helps in understanding the performance of WPT systems in terms of coil-to-coil efficiency, overall efficiency, and output power capacity and facilitates system performance optimization. Three methods to achieve constant output power for variable-load systems are presented based on system performance analysis. An optimal method can be selected for a specific WPT system by comparing the efficiencies of the three methods calculated with the proposed model. A two-coil 1 kW WPT system is built to verify the proposed mathematical model and constant output power control methods. Experimental results show that when the load resistance varies between 5 and $25{\Omega}$, the system output power can be maintained at 1 kW with a maximum error of 6.75% and an average error of 4%. Coil-to-coil and overall efficiencies can be maintained at above 90% and 85%, respectively, with the selected optimal control method.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.2
• /
• pp.437-447
• /
• 2006

Error Correcting Output Coding(ECOC) is used to solve multi-class problem. It is known that it improves the classification accuracy. In this paper, we compared various criteria to design code matrix while encoding. In addition. we prorpose an ensemble which uses the ability of each classifier while decoding. We investigate the justification of the proposed method through real data and synthetic data.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.232-237
• /
• 1991

Digital Variable Structure Controller(DVSC) is proposed to control variable speed recirculating pump for hot-water heating control system. In this study, nonlinear sliding line is used beyond output error boundary layer and PID sliding line is used within the layer. For long dead time compensation, constraint is added to Smith predictor algorithm. Steady state error is eliminated by using the proposed sliding line in spite of heating load change. By decreasing sampling time, good sliding motion is yielded but system output noise bv flow dynamics is amplified.

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.4
• /
• pp.1219-1231
• /
• 2008

The main object of this paper is to develop a leave-one-out(LOO) bound of all pairwise comparison error correcting output codes (APC-ECOC). To avoid using classifiers whose corresponding target values are 0 in APC-ECOC and requiring pilot estimates we developed a bound based on mean misclassification probability(MMP). It can be used to tune kernel hyperparameters. Our empirical experiment using kernel mean squared estimate(KMSE) as the binary classifier indicates that the bound leads to good estimates of kernel hyperparameters.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.17-20
• /
• 2000

In a digital communication system, the transmission channel may introduce error into the digital signal being transmitted. It would be useful if a process could be devised so that the error could be removed in order to recover the transmitted digital signal. We design a corrective filter that is inverse filter, which will generate an output signal identical to the input signal. in order for two systems connected in cascade to produce an output which is identical to the input signal, the over-all unit sample response of the cascade connection must be a unit sample function.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1997.11a
• /
• pp.137-146
• /
• 1997

The simplest bit error detection scheme is to append a parity bit to the end of a bit sequence. In this paper an error correction technique with the use of a parity bit is proposed, and the performance of the proposed system is analyzed. The error probability of the proposed system is compared with the output of computer simulation of the proposed system. It is also compared with the error probability of error at BPSK system, and the signal-to-noise ratio gain is showed.