• Journal of Integrative Natural Science
• /
• v.13 no.4
• /
• pp.147-151
• /
• 2020

In this paper, we propose a method to detect concept drift by applying Convolutional Neural Network (CNN) in a data stream environment. Since the conventional method compares only the final output value of the CNN and detects it as a concept drift if there is a difference, there is a problem in that the actual input value of the data stream reacts sensitively even if there is no significant difference and is incorrectly detected as a concept drift. Therefore, in this paper, in order to reduce such errors, not only the output value of CNN but also the probability vector are used. First, the data entered into the data stream is patterned to learn from the neural network model, and the difference between the output value and probability vector of the current data and the historical data of these learned neural network models is compared to detect the concept drift. The proposed method confirmed that only CNN output values could be used to reduce detection errors compared to how concept drift were detected.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.22 no.3
• /
• pp.114-122
• /
• 2014

Sensor fusion is the one of the main research topics. It offers the highly reliable estimation of vehicle movement by processing and mixing several sensor outputs. But unfortunately, every sensor has drift which degrades the performance of sensor. It means a single degraded sensor output may affect whole sensor fusion system. Drift in most research is ideally assumed to be zero because it's usually a nonlinear model and has sample variation. Plus, it's very difficult for the acceleration to separate drift from the output signal since it contains many contributors such as vehicle acceleration, slope angle, pitch angle, surface condition and so on. In this paper, modified spline interpolation is introduced as a dynamic temperature compensation method covering sample variation. Using the last known output and the first initial output is suggested to build and update compensation factor. When the system has more compensation data, the system will have better performance of compensated output because of the regression compensation model. The performance of the dynamic temperature compensation system is evaluated by measuring offset drift between with and without the compensation.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.9B
• /
• pp.1683-1689
• /
• 1999

We have implemented Frequency bandwidth expander with frequency upconverting by VCO drift canceller and comb generator. Te output of the low frequency synthesizer which the output frequency is 220~280MHz(Resolution : 5MHz) is expanded to 1660~2140MHz by this system. The phase noise of this system only depends on the phase noise of comb generator and low frequency synthesizer. The phase noise of VCO don’t influence at the frequency expander because the drift of VCO cancel out. When we control the output of VCO, the output frequency of this system is varied by 60MHz x N as filter banker. The switching time and the spurious of the frequency expander is below 3usec, -55dBc respectively. This system easily expands bandwidth additively by expanding the output bandwidth of the VCO. We can apply the frequency expander to very wide band microwave synthesizer which has fast switching time.

• Journal of the Korean Solar Energy Society
• /
• v.30 no.1
• /
• pp.7-12
• /
• 2010

As many grid-connected photovoltaic power conditioning systems (PVPCS)are installed in many residential areas simultaneously, these have raised potential problems of network protection on electrical power system. One of the numerous problems is an Islanding phenomenon. In this paper, active frequency drift (AFD) method, one of the anti-islanding methods which is analyzed by current magnitude compensation and calculation of the fundamental component. Both harmonic component and RMS value of the output current for THD analysis are provided and verified by simulation.

• Journal of the Korean Solar Energy Society
• /
• v.29 no.6
• /
• pp.75-80
• /
• 2009

Recently, as the grid-connected photovoltaic power conditioning systems (PVPCS) are installed in many residential areas, these have raised potential problems of network protection on electrical power system. One of the numerous problems is an Islanding phenomenon. In this paper, active frequency drift (AFD) method, one of the anti-islanding methods which is analyzed by current magnitude compensation and calculation of RMS value of the output power is proposed and verified by simulation.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.16 no.2
• /
• pp.145-151
• /
• 2011

In this paper, among the active islanding detection techniques, the modified active frequency drift method was analyzed, which is relatively easy to apply to the single-phase grid-connected PV PCS. The existing designs for turbulences in these applications were empirically conducted, and do not have sufficient reliability and performance. Therefore, three application forms of the modified active frequency drift technique were modeled, based on which the proper magnitude of turbulence, which is the frequency acceleration component, was calculated. Using the results, the magnitude of and injection method for turbulence for ensuring the islanding detection performance and improving the output power quality were proposed, and they were verified via simulations and experiment to prove that the reliable islanding detection technique can be developed merely by measuring the basic output power quality, without the need for expensive islanding simulation equipment.

• Journal of the Korean Magnetics Society
• /
• v.15 no.6
• /
• pp.332-335
• /
• 2005

Output voltage of the flux-meter has always drift due to the input bias current of non-ideal operational amplifier. In this study we have employed a digital sample and hold amplifier which has no voltage drop to compensate drift of the flux-meter automatically. The drift of the developed flux-meter was smaller than $5{\times}10^{-8}\;Wb/s$ for the integration time constant of $RC=10^{-3}$ s.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.07a
• /
• pp.985-988
• /
• 2001

A piezo-buzzer being used for the purpose of generation of audible frequency, which is a electric-acoustic transducer utilizing the inverse piezoelectric effect. Also it can be used for a pressure sensor according to the piezoelectric effect. But the output of a piezo-buzzer is a differential signal. In this study, we've made a system that can measure a real pressure by integration of output signal. According to our results, it could be found a possibility of application for pressure sensor by measurement of output characteristics when a piezo-buzzer was pressurized and depressurized, and by measuring of an error by means of the drift current of OP-Amp, etc..

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.301-304
• /
• 2002

This paper has been studied an A/D conversion system for precision weighing signal process In weighing system. A/D conversion has some problem.; offset drift voltage with environment situation and nonzero value of initial output voltage. The Offset voltage in analog circuit produces a drift of an output voltage before A/D conversion stage. This paper suggested the method of reducing the offset voltage by switching analog chopping circuit and making the initial output close to zero to enhance the swing range by D/A converter. Also, we have designed active filter and digital filter with Auto Zero Tracking algorithm for better signal process of the weighing system.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.5
• /
• pp.927-932
• /
• 2022

Instrument drift is caused by the passage of time, environmental changes, normal wear and tear, debris buildup, sudden shock, vibrations, electromagnetic fields, and improper use. Since it is inappropriate to directly determine the change of the output value as drift during the limited test period, a new algorithm that reflects both zero drift and span drift by giving changes over time to the calibration method of the instrument was proposed. The temperature drift was calculated to be 0.49% for about 60 minutes at 1-minute intervals in the nine-step constant temperature environment through the warming and cooling process.