• International Journal of Automotive Technology
• /
• 제6권2호
• /
• pp.95-105
• /
• 2005

Diesel engines have low specific fuel consumption, but high particulate emissions, mainly soot. Diesel soot is suspected to have significant effects on the health of living beings and might also affect global warming. Hence stringent measures have been put in place in a number of countries and will be even stronger in the near future. Diesel engines require either advanced integrated exhaust after treatment systems or modified engine models to meet the statutory norms. Experimental analysis to study the emission characteristics is a time consuming affair. In such situations, the real picture of engine control can be obtained by the modeling of trend prediction. In this article, an effort has been made to predict emissions smoke and NO$_{x}$ using cylinder combustion derived parameters and diesel particulate filter data, with artificial neural network techniques in MATLAB environment. The model is based on three layer neural network with a back propagation learning algorithm. The training and test data of emissions were collected from experimental set up in the laboratory for different loads. The network is trained to predict the values of emission with training values. Regression analysis between test and predicted value from neural network shows least error. This approach helps in the reduction of the experimentation required to determine the smoke and NO$_{x}$ for the catalyst coated filters.

• 한국전자통신학회논문지
• /
• 제14권5호
• /
• pp.943-950
• /
• 2019

머신 러닝은 인력을 대체함으로써 업무 효율성을 크게 높일 수 있다. 특히 4차 산업혁명 시대의 요청에 따라 인공지능을 포함한 머신 러닝의 중요성은 점점 커지고 있다. 본 논문은 MLP, RNN, LSTM, ANFIS 신경망 알고리즘 이용하여, 월별 전력 거래량을 예측한다. 본 논문에서는 통계청에서 제공하는 월별 전력 거래량과 월별 전력 거래금액, 최종에너지 소비량, 자동차용 경유 가격에 대한 2001~2017년까지의 공공 데이터를 사용하였다. 본 논문은 제시하는 각각의 알고리즘들을 학습시키고, 알고리즘이 예측하는 시계열 그래프를 이용하여 예측 결과를 보여주고 RMSE를 이용하여 이들 중에서 가장 우수한 알고리즘 제시한다.

• 한국대기환경학회지
• /
• 제6권1호
• /
• pp.43-50
• /
• 1990

In the estimation of grid type area source emissions, the factors by which emissions are distributed spatially must be considered comprehensively. The estimation process can be approached systematically by connecting the emission activity of air pollutants with land use pattern and the various impacts can be taken into account in the process by using such area information as land use, population, distribution of households, types of housing units, fuel consumption and so on. The estimation procedure was computerized in order to calculate area source emissions by grid information system. This computerization helped to augment effectiveness and efficiency not only by managing enormous data but also by performing diverse functions.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.497-499
• /
• 2020

자율주행 차량이 많아지고 차량의 ECU가 고도화되면서 정확한 차량의 데이터를 획득하고 분석하여 활용하는 것이 중요해지고 있다. 현재에는 내연 기관 차량의 ECU 데이터를 얻기 위해서 OBDII 포트(규격)에 기반한 CAN동선을 주로 이용하고 있다. 하지만 OBDII 규격을 통해서 연비와 같은 중요한 차량 정보를 얻는 경우, 변환식 (MAF 센서(흡입 공기량 센서)와 공기/연료 비율을 이용)의 오차 범위가 커서 데이터의 정확도가 낮다. 본 연구에서는 머신 러닝 기법 중에 하나인 회귀 기법을 통해서 기존의 계산보디 더 정확한 연비를 구할 수 있는 모델을 개발하였다. 이러한 모델 개발을 통하여 차량의 RAW 데이터를 기반으로 필요한 차량 데이터를 정확하게 구할 수 있게 되었으며 20회가 넘는 실 도로주행을 통해서 본 모델의 정확도를 검증하였다.

• Journal of Advanced Marine Engineering and Technology
• /
• 제8권1호
• /
• pp.17-36
• /
• 1984

Since 1973, the competition on the development of fuel saving type internal combustion engines has become severe by the two times oil shock, and new type engines are reported every several months. Whenever these new type engines are developed, new designs are required and they will be offered in the market after performing the endurance test for a long time. But the engine market is faced with a heavy burden of finance, as the developing of a new engine requires tremendous expenses. For this reason, the computer simulation method has been lately developed to cope with it. The computer simulation method can be available to perform the reasonable research works by the theoretical analysis before carrying out practical experiments. With these processes, the developing expenses are cut down and the period of development is curtailed. The object of this study is the development of simulation computer program for the small naturally aspirated four-stroke diesel engine which is intended to product by the original design of our country. The process of simulation is firstly investigated for the ideal engine cycle, and secondly for the real engine cycle. In the ideal engine cycle, each step of the cycle is simulated by the energy balance according to the first law of thermodynamics, and then the engine performance is calculated. In the real cycle imulation program, the injection rate, the preparation rate and the combustion rate of fuel and the heat transfer through the wall of combustion chamber are considered. In this case, the injection rate is supposed as constant through the crank angle interval of injection and the combustion rate is calculated by the Whitehouse-Way equation and the heat transfer is calculated by the Annand's equation. The simulated values are compared with measured values of the YANMAR NS90(C) engine and Mitsubishi 4D30 engine, and the following conclusions are drawn. 1. The heat loss by the exhaust gas is well agree with each other in the lower load, but the measured value is greater than the calculated value in the higher load. The maximum error rate is about 15% in the full load. 2. The calculated quantity of heat transfer to the cooling water is greater than the measured value. The maximum error rate is about 11.8%. 3. The mean effective pressure, the fuel consumption, the power and the torque are well agree with each other. The maximum error is occurred in the fuel consumption, and its error rate is about 7%. From the above remarks, it may be concluded that the prediction of the engine performance is possibly by using the developed program, although the program needs to reform by adding the simulation of intake and exhaust process and assumping more reliable mechanical efficiency, volumetric efficiency, preparation rate and combustion rate.

• 한국산학기술학회논문지
• /
• 제20권2호
• /
• pp.1-6
• /
• 2019

본 연구의 목적은 100kW급 연료전지 시스템의 열관리 성능을 실도로 운전조건에서 분석하여, 성능 해석 모델링을 개발하는 것이다. 개발된 모델을 적용하여, 열관리 시스템의 운전조건 변화에 따른 성능 변화를 고찰하고자 한다. 해석 모델링은 핵심부품들에 대한 성능 평가 데이터를 바탕으로, 성능에 영향을 주는 변수들로 개발하였다. 개발된 연료전지 열관리 시스템 해석 모델링으로 다양한 실차 운전조건에서의 최적 열관리 시스템에 대한 전력소비량을 분석하였다. 주요하게, 연료전지 열관리시스템 핵심부품(워터펌프, 냉각 팬, 3 Way Valve, 라디에이터)에 대한 성능 특성 분석 후 모델링을 진행하였다. 개발된 모델링으로 운전조건에 따른 유량 예측을 하였고, 실험값과 예측값과의 비교분석을 통하여서, 해석 모델링에 대한 검증을 진행하였다. 과도해석을 통하여서, 냉시동시 냉각수 온도가 특정온도까지의 소요시간을 예측하였다. 스택 운전조건에서 스택 입출구 온도가 적정 수준에서 움직이기 위한 열관리 시스템 운전조건에 대한 예측을 진행하였다. 그 결과를 바탕으로, 소모전력과 열방출량과의 비교분석을 하였다. 개발된 해석 모델링은 핵심부품들의 성능 변화시 연료전지 시스템 운전에 대한 영향도를 분석할 수 있도록 활용할 예정이다.

• International Journal of Aeronautical and Space Sciences
• /
• 제16권2호
• /
• pp.311-324
• /
• 2015

This study presents computational analyses for low-drag aerodynamic design that are applied to modify a long-endurance UAV. EAV-2 is a test-bed for a hybrid electric power system (fuel cell and solar cell) that was developed by the Korean Aerospace Research Institute (KARI) for use in future long-endurance UAVs. The computational investigation focuses on designing a wing with a reduced drag since this is the main contributor of the aerodynamic drag. The airfoil and wing aspect ratio of the least drag are defined, the fuselage configuration is modified, and raked wingtips are implemented to further reduce the profile and induced drag of EAV-2. The results indicate that the total drag was reduced by 54% relative to EAV-1, which was a small-sized version that was previously developed. In addition, static stabilities can be achieved in the longitudinal and lateral-directional by this low-drag configuration. A long-endurance flight test of 22 hours proves that the low-drag design for EAV-2 is effective and that the average power consumption is lower than the objective cruise powerof 200 Watts.

• 한국기계가공학회지
• /
• 제11권3호
• /
• pp.116-121
• /
• 2012

In recent years, world is faced with a transportation energy dilemma, and the transportation is dependent on a single fuel - petroleum. However, Hybrid Electric Vehicle(HEV) technology holds more advantages to reduce the demand for petroleum in the transportation by efficiency improvements of petroleum consumption. Therefore, there is a trend that lower gear noise levels are demanded in HEV for drivers to avoid annoyance and fatigue during operation. And meshing transmission error(T.E.) is the excitation that leads to the tonal noise known as gear whine, and radiated gear whine is also the dominant source of noise in the whole gearbox. This paper presents a method for the analysis of gear tooth profile and lead modification, and the predictions of transmission error and load distribution are shown under one loaded torque for the 1st gear pair of HEV gearbox. The test is also obtained before tooth micro-modification under the torque. At last, the appropriate tooth modification is used to minimize the transmission error and load distribution under the loaded torque. It is a good approach which the simulated result is used to improve the design in order to minimize the radiation gear whine noise.

• Journal of Advanced Marine Engineering and Technology
• /
• 제24권4호
• /
• pp.446-453
• /
• 2000

The closed cycle diesel system is operated in closed circuit system where there is non air breathing with working fluid consisted of the combination of oxygen, argon and recycled exhaust gas for obtaining underwater or underground power sources. this study has been carried out to analysis the performance of closed cycle system by means of investigation on the combustion characteristics of diesel engine MTU8V183TE52 operating in open, semi-closed, and closed cycle modes. The combustion in closed mode starts a little bit earlier than in open cycle mode. The oxygen concentration and fuel consumption at 240kW closed cycle running are 21∼24% by volume and 77∼79kg/h, respectively. The maximum cylinder pressure and ignition delay time are investigated 110bar and 8.9degree. Also, The combustion simulation program has been studied to predict whether or not combustion. The results from numerical prediction for the basic, cylinder averaged quantities such as the cylinder pressure and the heat release showed excellent with the experimental data.

• 설비공학논문집
• /
• 제10권6호
• /
• pp.695-701
• /
• 1998

Researches on unused energy are being continued because of the limited fossil fuel and the destruction of environment. Therefore this study was peformed as follows. The collectable amount of exhausted heat for an air-conditioning was calculated by the subway thermal environment prediction program. And the electric power needed by conventional heat source equipments was compared with one by unused heat source equipments when the exhausted heat was used by heat pump in heating and hot water supplying. The results are summarized as follows; 1) Forced ventilation should be conducted to keep optimal temperature in subway tunnel in summer as well as in winter. According to the simulation, temperature in tunnel was higher than that on the ground in summer when the forced ventilation was conducted only in winter. 2) Ventilating time should be calculated out to the optimal condition for not only saving power of ventilation fan but reusing exhausted heat. By the simulation, it is certain that the exhausted heat should be eliminated in air-conditioning time. 3) The use of exhausted heat source heat pump could save 8% of electric power per hour in comparison with existing heat pump. It was based on a present heat generation and traffic for ventilating time of general air-conditioning, but could be different by ventilating time. 4) As the traffic increases up to 1.5 or 2 times, electric power consumption of the conventional heat pump increases to 11% or 13.5% per mean hour in comparison with that of the exhausted heat source heat pump, though all-day ventilation.