• 자동차안전학회지
• /
• 제15권1호
• /
• pp.55-62
• /
• 2023

In this study, a chest deflection is predicted by introducing a deep learning technique with the results of the frontal impact of the USNCAP conducted for 110 car models from MY2018 to MY2020. The 120 data are divided into training data and test data, and the training data is divided into training data and validation data to determine the hyperparameters. In this process, the deceleration data of each vehicle is averaged in units of 10 ms from crash pulses measured up to 100 ms. The performance of the deep learning model is measured by the indices of the mean squared error and the mean absolute error on the test data. A DNN (Deep Neural Network) model can give different predictions for the same hyperparameter values at every run. Considering this, the mean and standard deviation of the MSE (Mean Squared Error) and the MAE (Mean Absolute Error) are calculated. In addition, the deep learning model performance according to the inclusion of CVW (Curb Vehicle Weight) is also reviewed.

• International Journal of Contents
• /
• 제13권4호
• /
• pp.23-28
• /
• 2017

Of the total economic loss caused by disasters, 40% are due to floods and floods have a severe impact on human health and life. So, it is important to monitor the water level of a river and to issue a flood warning during unfavorable circumstances. In this paper, we propose a modified error function to improve a hydrological modeling using a multi-layer perceptron (MLP) neural network. When MLP's are trained to minimize the conventional mean-squared error function, the prediction performance is poor because MLP's are highly tunned to training data. Our goal is achieved by preventing overspecialization to training data, which is the main reason for performance degradation for rare or test data. Based on the modified error function, an MLP is trained to predict the water level with rainfall data at upper reaches. Through simulations to predict the water level of Nakdong River near a UNESCO World Heritage Site "Hahoe Village," we verified that the prediction performance of MLP with the modified error function is superior to that with the conventional mean-squared error function, especially maximum error of 40.85cm vs. 55.51cm.

• Journal of the Korean Data and Information Science Society
• /
• 제27권6호
• /
• pp.1645-1651
• /
• 2016

A lot of data, particularly in the medical field, contain variables that have a measurement error such as blood pressure and body mass index. On the other hand, recently smoothing methods are often used to solve a complex scientific problem. In this paper, we study a Bayesian curve-fitting under functional measurement error model. Especially, we extend our previous model by incorporating covariates free of measurement error. In this paper, we consider penalized splines for non-linear pattern. We employ a hierarchical Bayesian framework based on Markov Chain Monte Carlo methodology for fitting the model and estimating parameters. For application we use the data from the fifth wave (2012) of the Korea National Health and Nutrition Examination Survey data, a national population-based data. To examine the convergence of MCMC sampling, potential scale reduction factors are used and we also confirm a model selection criteria to check the performance.

• Journal of Electrical Engineering and Technology
• /
• 제10권3호
• /
• pp.1201-1211
• /
• 2015

WSN acts as an effective tool for tracking the large scale environments. In such environment, the battery life of the sensor networks is limited due to collection of the data, usage of sensing, computation and communication. To resolve this, a mobile robot is presented to identify the data present in the partitioned sensor networks and passed onto the sink. In novel data collection algorithm, the performance of the data collecting operation is reduced because mobile robot can be used only within the limited range. To enhance the data collection in a changing environment, Non Linear Error Identifier (NLEI) algorithm has been developed and presented in this paper to configure the robot by means of error models which are non-linear. Experimental evaluation has been conducted to estimate the performance of the proposed NLEI and it has been observed that the proposed NLEI algorithm increases the error correction rate upto 42% and efficiency upto 60%.

• 대한전기학회논문지:전력기술부문A
• /
• 제52권10호
• /
• pp.571-576
• /
• 2003

XML Web Service based on internet can cause problems on transmission speed and data error. Also system analysis results simulated by several different research groups can hardly have reliability because of error data that come from improperly managed files. In order to solve this problems, algorithm sever using XML Web Service is shared on the internet so widely that various application programs based on basic analysis module with a united IO can be developed. And also XML Dataset DB is interacted with XML Web Service, which prevents propagation of error data. It causes to improve reliabilityon the load flow analysis result and solve the problems on data error or transmission speed that can possibly come from internet.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제8권6호
• /
• pp.736-745
• /
• 2002

압축한 비디오 데이타를 전송할 때 인터넷과 같이 네트워크 채널이 불안정한 경우 패킷이 분실될 가능성이 높다. 패킷 분실은 연속적 비트 열에 오류가 발생하는 버스트 오류 형태로 일어난다. 본 논문에서는 버스트 오류를 은닉 처리하는데 효과적인 오류 내성 기법을 적용하는 동시에 데이타 숨김을 이용하여 디코더의 계산 복잡도를 줄인 빠른 오류 은닉 방법을 제안한다. 오류 은닉 효과를 높이기 위해, 인코더에서는 네트워크 채널의 버스트 오류에 강건하도록 비디오 데이타에 공간적, 시간적 영역에 대한 오류 내성 기법을 적용한다. 공간적 오류 내성 기법으로는 패킷 분실이 발생한 오류 블록을 분리하는데 효과적인 블록 셔플링을 적용하고 시간적 오류 내성 기법으로는 움직임 벡터의 프레임간 패리티 비트를 데이터 숨김 방법으로 내용 데이타에 삽입, 전송하여 디코더에서 분실된 패킷을 처리한다. 비디오 데이타는 전송 후 디코더에서 오류 은닉 처리하는데 디코더에서 주변 정보를 이용하여 오류 비디오 블록을 보간하는 것은 계산이 복잡하여 비용이 많이 든다. 따라서 본 연구에서는 비디오 인코딩 단계에서 비디오 블록의 에지 특징을 추출 후 이 특징 데이타를 원 데이타에 숨겨 전송하고 전송 시 비디오 데이타가 손상되면 디코더에서 숨겨 온 비디오 블록의 특징을 추출하여 쌍선형 보간법을 통해 전송 시 발생한 오류를 은닉 처리한다. 데이타 숨김을 이용하면 디코더의 계산 복잡도는 낮아진다. 본 논문의 실험 결과는 제안 방법이 비디오의 패킷 분실이 30%에 달하는 경우에도 이를 은닉 처리하여 인지 가능한 품질의 비디오 데이타를 보장한다.

• Ocean and Polar Research
• /
• 제30권2호
• /
• pp.149-158
• /
• 2008

The characteristic analysis of the estimated population parameters, i.e., standard deviation and error bound of coastal pollutant concentrations (hereafter PC, i.e., COD, TN, and TP concentrations), was carried out by using environmental data with different sampling frequency in Lake Shihwa and Incheon coastal zone. The results clearly show that standard deviation of the PC increases as its mean value increases. The error bounds of the annual mean values based on seasonally measured DO concentrations and PC data in Incheon coastal zone were estimated as ranges 2.26 mg/l, $0.68{\sim}0.86\;mg/l$, $0.62{\sim}0.80\;mg/l$, and $0.074{\sim}0.082\;mg/l$, respectively. In terms of annual mean of the DO concentration and PC in Lake Shihwa, the error bounds based on monthly measured data from 1997 to 2003 were also estimated as ranges 4.0 mg/l, 3.0 mg/l, $0.5{\sim}1.0\;mg/l$, and 0.05 mg/l, respectively. The error bound on the basis of real-time monitoring data is $7{\sim}13%$ only as compared to that of monthly measured data.

• 산업경영시스템학회지
• /
• 제43권2호
• /
• pp.1-13
• /
• 2020

There are several methods of peak-shaving, which reduces grid power demand, electricity bought from electricity utility, through lowering "demand spike" during On-Peak period. An optimization method using linear programming is proposed, which can be used to perform peak-shaving of grid power demand for grid-connected PV+ system. Proposed peak shaving method is based on the forecast data for electricity load and photovoltaic power generation. Results from proposed method are compared with those from On-Off and Real Time methods which do not need forecast data. The results also compared to those from ideal case, an optimization method which use measured data for forecast data, that is, error-free forecast data. To see the effects of forecast error 36 error scenarios are developed, which consider error types of forecast, nMAE (normalizes Mean Absolute Error) for photovoltaic power forecast and MAPE (Mean Absolute Percentage Error) for load demand forecast. And the effects of forecast error are investigated including critical error scenarios which provide worse results compared to those of other scenarios. It is shown that proposed peak shaving method are much better than On-Off and Real Time methods under almost all the scenario of forecast error. And it is also shown that the results from our method are not so bad compared to the ideal case using error-free forecast.

• 항공우주기술
• /
• 제11권2호
• /
• pp.96-102
• /
• 2012

최근 여러 개의 지상국에서 수신한 PCM 데이터를 병합하여 잡음에 의한 프레임 오류를 제거하는 방법에 관한 관심이 증가하고 있다. 단순 병합 방식은 이미 나로우주센터의 데이터처리 시스템에 적용되어, 나로호의 1, 2차 비행시험 데이터 분석에 사용된 바 있다. 본 논문은 단순 데이터 병합방식에 데이터 치환, 비트단위 투표 등의 오류교정 알고리즘 및 시간지연데이터를 이용한 교정알고리즘을 적용하여 오류율을 더 낮추는 데 초점을 맞추고 있다. 네 개 지상국에서 수신된 나로호 비행시험 데이터에 제안된 알고리즘을 적용한 결과 단순 병합방식보다 1.32%의 오류율이 개선된 것으로 나타났다. 제시된 알고리즘은 향후 다양한 텔레메트리 병합데이터 생성에 유용하게 사용될 수 있으리라 판단된다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.1144-1148
• /
• 1996

This paper presents WMR modelling and path tracking algorithm using Inertial Navigation System. The error models of gyroscope and accelerometers in INS are derived by Gauss-Newton method which is nonlinear regression model. Then, to test availability of error model, we pursue the fitness diagnosis about probability characteristic for real data and estimated data. Performance of inertial sensor with error model and Kalman filter is pursued by comparing with one without them. The computer simulation shows that position error remarkably decrease when error compensation is applied.