• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.25 no.12
• /
• pp.888-894
• /
• 2015

A new method for the simulation of the vehicle's interior road noise is proposed in the present study. The road noise model can synthesize road noise of a vehicle for varying driving speed within a range. In the proposed method, interior road noise is considered as a stochastic time-series, and is modeled by a nonstationary parametric model via two steps. First, each interior road noise signal, obtained from constant speed driving tests performed within a range of speed, is modeled as an autoregressive model whose parameters are estimated by using a standard method. Finally, the parameters obtained for different driving speeds are interpolated based on the varying driving speed to yield a time-varying autoregressive model. To model a full band road noise, audible frequency range is divided into an octave band using a wavelet filter bank, and the road noise in each octave band is modeled.

• Proceedings of the KIEE Conference
• /
• 2008.10b
• /
• pp.489-490
• /
• 2008

본 고에서는 체내 심실신호를 농하여 신경학적 분석 및 다형성의 측면에서 심실세동이 일어나는 것을 예측하는 분석 알고리즘을 설계하였다. 신경학적 측면에서는 시계열 신호의 Peak to Peak Interval을 예측법과 0.15Hz를 기준으로 HRV 신호의 AR Burg 모델링을 통하여 고주파성과 저주파성을 나누어 교감신경과 부교감신경의 활동성 통한 신경학적 예측법을 제시하였으며 또한 체내 심실신호의 비선형적 특성을 고려한 Fractal Dimension을 생성시킴으로서 주기성의 특성과 다형성 통한 예측법을 제시하였다. 체내 심전도를 기반으로 Simulation 하였으며 각 분석별 조합을 통하여 최적의 예측 구조를 찾고자 하였다. 의학적 의미가 있는 민감도와 특이도를 판별하였으며 예측을 위한 수행시간을 실험하였다. 이를 통하여 자율신경 활성도와 다형성 판별을 조합한 방법이 심실세동 예측을 위한 민감도의 측면에서 가장 우수함을 나타내었고 시뮬레이션을 위만 시뮬레이터(Simulator) UI(User Interface)를 제시하였다.

• Journal of IKEEE
• /
• v.24 no.2
• /
• pp.523-528
• /
• 2020

With a widespread of sensor-rich mobile devices, the analysis of human activities becomes more general and simpler than ever before. In this paper, we propose two deep neural networks that efficiently and accurately perform human activity recognition (HAR) using tri-axial accelerometers. In combination with powerful modern deep learning techniques like batch normalization and LSTM networks, our model outperforms baseline approaches and establishes state-of-the-art results on WISDM dataset.

• Journal of the Society of Naval Architects of Korea
• /
• v.47 no.2
• /
• pp.242-247
• /
• 2010

To consider effects of essential parameters of water impact pressure on dynamic structural responses of bow bottom structures, a parametric study for a ship bottom panel is carried out. The idealized pressure time history models were assumed by triangular and rectangular shapes in time domain. The main loading parameters are duration time and peak pressure value maintaining the same impulse value. The structural models for local bottom stiffened panels of a container ship are analysed. The natural frequency analysis and transient dynamic response analysis are performed using MSC/NASTRAN. Added mass effects of contacting water are considered and the pressure distributions are assumed to be uniform in the whole water contacting surface. The effects of loading parameters on the structural responses, especially maximum displacements, are considered. Besides the peak pressure value, effects of duration time correlated with natural frequencies are thought to be the important parameters.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.54 no.5
• /
• pp.91-102
• /
• 2012

강우의 관측망 확장과 위성 자료 및 기후 모델을 이용한 격자 단위자료가 개발 및 보급됨에 따라 다양한 자료의 분야별 활용성에 대한 연구의 필요성이 제기되고 있다. 본 연구에서는 지역 기후 모델 산출물의 오차 보정을 위한 격자 관측자료의 활용성을 평가하였다. 또한 통합 분포형 수문모델을 이용하여, 보정한 기후모델 결과의 수문 모의를 위한 기후 입력 자료로써의 적합성을 검토하였다. 보정된 결과는 각 관측자료의 월별 평균 강우량과 공간 분포를 비교적 잘 재현하였다. 한편 연강우량 시계열에 있어 그 양상은 잘 재현된 가운데 보정되지 않은 오차를 일부 포함하는 것으로 나타났다. 이는 점 관측자료로부터 추정된 시험 지역내 172개 소유역에 대한 일평균 강우량 자료와 비교해 볼 때 관측자료의 형식이나 정확성보다 기후모델의 불확실성에 기인하는 것으로 판단된다. 수문 모의 결과, 격자 자료를 이용하여 보정한 강우 입력자료는 수문 모델의 검보정에 이용된 소유역 단위 강우 자료를 이용한 결과에 상응하는 활용성을 보여주었다. 또한 강우의 공간 분포를 고려하지 않고, 시험유역 전체에 대한 평균 강우량을 입력 자료로 이용한 결과를 통해 기후 자료의 공간 분포와 관측 밀도의 중요성을 확인하였다.

• Journal of the Korean Society for Precision Engineering
• /
• v.11 no.1
• /
• pp.138-149
• /
• 1994

This paper introduces a new monitoring technique which utilizes an adaptive signal processing for feature generation, coupled with a multilayered merual network for pattern recognition. The cutting force signal in face milling operation was modeled by a low order discrete autoregressive model, shere parameters were estimated recursively at each sampling instant using a parameter adaptation algorithm based on an RLS(recursive least square) method with discounted measurements. The influences of the adaptation algorithm parameters as well as some considerations for modeling on the estimation results are discussed. The sensitivity of the extimated model parameters to the tool state(new and worn tool)is presented, and the application of a multilayered neural network to tool state monitoring using the previously generated features is also demonstrated with a high success rate. The methodology turned out to be quite suitable for in-process tool wear monitoring in the sense that the model parameters are effective as tool state features in milling operation and that the classifier successfully maps the sensors data to correct output decision.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.135-135
• /
• 2020

우리나라의 기후 자료는 일반적으로 기상청에서 발표하는 종관기상관측(ASOS)과 방재기상관측(AWS), 그리고 북한이 세계기상기구(WMO, World Meteorogical Organization)의 기상통신망(GTS)을 통해 보낸 북한기상관측(NKO)을 사용 할 수 있다. 그러나 이 중 40년 이상의 완전한 관측 자료를 얻을 수 있는 건 ASOS가 유일하지만 공간적인 표현에 한계를 갖고 있다. AWS는 관측소가 많다는 장점이 있지만 관측 기간이 길지 않고 이용 가능한 기간에도 관측이 연속적이지 못한 경우가 많다. NKO는 비록 27개의 관측소가 있지만 많은 데이터가 누락되어 일별 기후자료의 사용에 한계를 갖고 있다. 이러한 미관측 기간이나 관측 자료의 누락은 연속적인 시계열 자료분석을 기반으로 하는 수자원 모델링에 있어서 문제를 야기한다. 본 연구는 1973년부터 2019년까지 47년의 신뢰도 높은 한반도 일일 기후 자료를 구축하기 위해 다양한 방법론을 비교하였다. 추정에 사용한 방법은 총 7개로 EM algorithm for probabilistic principal components (PPCA-EM), Inverse distance weight method (IDWM), Nearest neighbor method (NNM), Multivariate normal copulas (Copula), Elastic net model (Elastic), Ordinary kriging (OK), Regularized principal components with EM algorithm (RPCA-EM)를 살펴보았다. 다양한 형태의 결측치를 가정하여 그 결과값을 비교하였고 이는 Root mean squared error(RMSE), Kling-Gupta efficiency(KGE), Nash-Sutcliffe efficiency(NSE)를 통해 평가하였다. 최종 선택된 방법론을 통하여 한반도 전역을 그리드 기반의 강수 및 최저온도/최고온도의 일별자료로 생성하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.150-152
• /
• 2022

In this work, we propose a model that considers the behavior and synaptic plasticity of sensory neurons based on Liquid Time-constant Network (LTC). The neuron connection structure was experimented with four types: the increasing number of neurons, the decreasing number, the decreasing number, and the decreasing number. In this study, we experimented using a time series prediction dataset to see if the performance of the changed model improved compared to LTC. Experimental results show that the application of modeling of sensory neurons does not always bring about performance improvements, but improves performance through proper selection of learning rules depending on the type of dataset. In addition, the connective structure of neurons showed improved performance when it was less than four layers.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.3
• /
• pp.216-225
• /
• 2020

In this paper, broadband underwater propagation channel modeling based on eigenray analysis is discussed. Underwater channels are often formulated in frequency domain time-harmonic signals, which are impractical for simulating broadband signals in time domain. In this regard, time domain modeling of the underwater propagation channel is required for the simulation of broadband signals, for which the eigenray analysis based on ray tracing, resulting in multipath propagation delays in time-domain, is used in this paper. For discrete time system application, the phase, frequency-dependent loss and non-integer sample delays for each eigenray, are approximated by the finite impulse response of the broadband propagation channel.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.809-817
• /
• 2020

In the design and development process of Small-Sat power distribution and transmission module, the stability of dynamic resources was evaluated by a deep learning algorithm. The requirements for the stability evaluation consisted of the power distribution function of the power distribution module and demand module to the SAR radar in Small-Sat. To verify the performance of the switching power components constituting the power module PDM, the reliability was verified using a dynamic neural network. The adoption material of deep learning for reliability verification is the power distribution function of the payload to the power supplied from the small satellite main body. Modeling targets for verifying the performance of this function are output voltage (slew rate control), voltage error, and load power characteristics. First, to this end, the Coefficient Structure area was defined by modeling, and PCB modules were fabricated to compare stability and reliability. Second, Levenberg-Marquare based Two-Way NARX neural network Sigmoid Transfer was used as a deep learning algorithm.