• Journal of the Korean Solar Energy Society
• /
• v.32 no.3
• /
• pp.42-49
• /
• 2012

신경망방법은 공학, 경영 그리고 정보기술과 같이 다양한 분양에서 널리 사용되어지고 있다. 신경망방법은 기본적으로 예측, 제어, 식별과 같은 기능을 가지고 있는데, 본 논문에서는 신경망방법을 이용하여 C사의 모델 T의 히트펌프 전기부하를 예측하였다. 부하예측은 시스템을 더욱 효율적이고, 적절하게 만들기 위해 필요하다. 본 논문에서 사용된 히트펌프는 지열원 히트 펌프 시스템이다. 이 지열 히트 펌프의 부하는 사전에 미리 예측되어진 외기온도 및 건물 열부하에 따라 측정 학습된 전력 소비량으로 겨울에는 난방, 여름에는 냉방에 대한 전력 부하를 예측할 수 있다. 이 신경망방법은 신경망 학습 순서를 통해 부하 예측을 위해 히트펌프의 성능데이터를 필요로 한다. 이 부하 예측 인공지능망 방법으로 외기 온도별 건물 통합형 지열 히트 펌프 부하가 예측되어질 수 있다.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.13 no.8
• /
• pp.682-691
• /
• 2001

Most of radiant floor heating systems are operated in the intermittent heating mode in Korea. The application possibility of predictive suboptimal control for Koran residential house was investigated by computer simulation and experiment. For this study, TRNSYS program was used and an experimental facility consisting of tow rooms ($3\times4.4\times2.8 m$) identical in construction was built. The facility enabled simultaneous comparison of two different control method. And real multi residential hose was investigated. Results showed that outdoor air temperature prediction control was superior to the conventional control for radiant floor heating system operated in the intermittent heating mode. New control system resulted in good thermal environment and les energy consumption.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.11a
• /
• pp.225-226
• /
• 2014

본 연구에서는 회귀분석을 통해 오피스 건물의 에너지 사용량을 예측하는 시스템을 개발하고 이를 실증하는 실험적 연구를 수행하였다. 회귀분석의 파라미터로는 외기온도, 에너지 사용량 등이 사용되었으며 예측 정확도 향상을 위해 파라미터를 확장해서 실험하였다. 에너지 사용량 예측에 대한 검증을 위해서 실시간 데이터 수집과 분석을 위한 시스템을 개발하였으며, 해당 시스템을 이용해 수원 소재 오피스 건물에서 실증한 결과에 따르면 겨울철 에너지 사용량에 대한 예측 오차율이 10% 미만으로 나타났다.

• 전자공학회논문지 IE
• /
• v.49 no.2
• /
• pp.40-45
• /
• 2012

In this paper, cooling system of power transformers is proposed for temperature optimized control. We predict the peak temperature of power transformer coils using load factors and construct a cooling system using weighting function. For the optimized temperature control for power transformer, a correlation function based on the load factor of a load current and the each temperatures for winding coils, for air and for oil is presented to predict the winding-coil peak temperature. Also, the results controlled by applying the power transformer is presented.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.3
• /
• pp.1488-1494
• /
• 2011

In this paper, we propose an algorithm using linear regression model that forecasts the demand of heated water in winter. To supply heated water to apartments, stores and office buildings, Korea District Heating Corp.(KDHC) operates boilers including electric power generators. In order to operate facilities generating heated water economically, it is essential to forecast daily demand of heated water with accuracy. Analysis of history data of Kangnam Branch of KDHC in 2006 and 2007 reveals that heated water supply on previous day as well as temperature are the most important factors to forecast the daily demand of heated water. When calculated by the proposed regression model, mean absolute percentage error for the demand of heated water in winter of the year 2006 through 2009 does not exceed 3.87%.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.24 no.11
• /
• pp.759-766
• /
• 2012

To estimate the cooling load for the following day, outdoor temperature and humidity are needed in hourly base. But the meteorological administration forecasts only maximum and minimum temperature. New methodology is proposed for predicting hourly outdoor temperature and humidity by using the forecasted maximum and minimum temperature. The correlations for normalized outdoor temperature and specific humidity has been derived from the weather data for five years at Seoul, Daejeon and Pusan. The correlations for normalized temperature are independent of date, while the correlations for specific humidity are linearly dependent on date. The predicted results show fairly good agreement with the measured data. The prediction program is also developed for hourly outdoor dry bulb temperature, specific humidity, dew point, relative humidity, enthalpy and specific volume.

• Proceedings of the Korea Society for Energy Engineering kosee Conference
• /
• 1997.10a
• /
• pp.3-8
• /
• 1997

본 연구에서는 복합발전시스템의 외기온도 변화로 인한 출력저하 문제를 극복할 수 있는 LNG 냉열 이용 복합발전 시스템을 제안하였다. 본 연구에 의해 제안된 LNG 냉열 이용 복합발전 시스템의 타당성을 검토하기 위해 ASPEN과 GateCycle을 이용한 시뮬레이션 모델을 구성하였고, 모델에 의해 예측한 결과를 실제 발전소 성능시험결과와 비교하여, 본 시뮬레이션 방법의 예측정확도를 검증하였다. 본 시뮬레이션 방법을 토대로 LNG 냉열을 이용하여 가스터빈의 유입공기를 냉각시켰을 경우의 복합발전 시스템 성능변화를 분석하였다. 그 결과 LNG 냉열을 이용하여 유입 공기를 원하는 온도까지 냉각시켜 하절기에도 출력을 일정하게 유지시킬 수 있음을 확인할 수 있었고, 이를 위한 기스터빈과 LNG 간의 열교환기 설계기준도 제시하였다.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.19 no.5
• /
• pp.394-402
• /
• 2007

To estimate the cooling load for the following day, outdoor temperature and humidity are needed in hourly base. But the meteorological administration forecasts only maximum and minimum temperature. New methodology is proposed for predicting hourly outdoor temperature and humidity by using the forecasted maximum and minimum temperature. The correlations for normalized outdoor temperature and specific humidity has been derived from the weather data for five years from 2001 to 2005 at Seoul, Daejeon and Pusan. The correlations for normalized temperature are independent of date, while the correlations for specific humidity are linearly dependent on date. The predicted results show fairly good agreement with the measured data. The prediction program is also developed for hourly outdoor dry bulb temperature, specific humidity, dew point, relative humidity, enthalpy and specific volume.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.15 no.4
• /
• pp.287-293
• /
• 2003

This study examined the temperature-dependent regression model of energy consumption based on various measuring period. The methodology employed was to construct temperature-dependent linear regression model of daily energy consumption from one day to three months data-sets and to compare the annual heating energy consumption predicted by these models with actual annual heating energy consumption. Heating energy consumption from a building in Daejon was examined experimentally. From the results, predicted value based on one day experimental data can have error over 100%. But predicted value based on one week experimental data showed error over 30%. And predicted value based on over three months experimental data provides accurate prediction within 6% but it will be required very expensive.

• Journal of the Korean Society for Geothermal and Hydrothermal Energy
• /
• v.12 no.4
• /
• pp.8-14
• /
• 2016

In this study, the effects of prediction and reset control of outdoor air temperature on energy consumption for central heating system are researched by using TRNSYS program package, and the control performances with the suggested methods of prediction and reset control of outdoor air temperature are compared with the existing ones. As a result, the value of coefficient of determination $R^2$ for the predicted outdoor temperatures is improved and the suggested control method shows maximum 21.8% energy saving in comparison with existing control ones.