• 자원ㆍ환경경제연구
• /
• 제28권3호
• /
• pp.437-465
• /
• 2019

본 연구에서는 장기 전력 수요와 GDP 사이의 소득계수를 시간과 GDP의 값에 따라 변화하도록 모형화한 Chang et al.(2016)에 기반을 두어 장기 에너지 수요의 예측에 관련된 새로운 방법을 제안한다. 본 논문에서는 장기 에너지와 GDP 사이의 소득계수를 함수로 표현하고, 함수 주성분 분석(Functional Principal Component Analysis)을 통하여 함수계수(Functional Coefficient)를 예측하고 이를 장기 에너지 수요 예측에 적용한다. 또한 함수계수를 비모수적으로 추정할 때 너비띠 모수를 예측 실험 오차를 최소화하도록 설정하는 방식을 제안하였고 개별 국가의 함수계수 변화 패턴을 반영하여 개별 국가의 특수성을 반영하는 예측 방법도 제시한다. 실증분석에서는 전 세계 에너지 데이터를 이용하여 한국의 장기 에너지 수요 예측을 본 논문에서 제시한 방법으로 예측하고, 기존의 방법들 보다 안정적인 장기 에너지 수요 예측이 가능함을 보였다.

• 전기학회논문지
• /
• 제64권9호
• /
• pp.1276-1280
• /
• 2015

Among the various statistical factors for South Korea, the population has been steadily decreased by lower birthrate. Nevertheless, the number of household is constantly increasing amid population aging and single life style. In general, residential electricity use is more the result of the number of household than the population. Therefore, residential electricity consumption is expected to be far higher for decades to come. The existing long-term load forecasting, however, do not necessarily reflect the growth of single and two-member households. In this respect, this paper proposes the long-term load forecasting for residential users considering the effect of changes of the housing type, and in the case study the changes of the residential load pattern is analyzed for accurate long-term load forecasting.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부A
• /
• pp.292-294
• /
• 1998

It is very important for electric utility to expand generating facilities and transmission equipments in accordance with the increase of electricity demand. Regional electricity demand forecasting is among the most important step for long-term investment and power supply planning. The main objectives of this paper are to develop the methodologies for forecasting regional load demand. The Model consists of four models, regional economy, regional electricity energy demand, areal electricity energy demand. and areal peak load demand. This paper mainly suggests regional electricity energy demand model and areal peak load demand. A case study is also presented.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1990년도 하계학술대회 논문집
• /
• pp.87-91
• /
• 1990

Regional electricity demand forecasting is among the most important step for lone-term investment and power supply planning. This study presents a regional electricity forecasting model for Korean power system. The model consists of three submodels, regional economy, regional electricity energy demand, and regional peak load submodels. A case study is presented.

• Communications for Statistical Applications and Methods
• /
• 제22권6호
• /
• pp.675-683
• /
• 2015

In the 21st century, we now face the serious problems of the enormous consumption of the energy resources. Depending on the power consumption increases, both China and South Korea face a reduction in available resources. This paper considers the regression models and time-series models to compare the performance of the forecasting accuracy based on Mean Absolute Percentage Error (MAPE) in order to forecast the electricity demand accurately on the short-term period (68 months) data in Northeast China and find the relationship with Korea. Among the models the support vector regression (SVR) model shows superior performance than time-series models for the short-term period data and the time-series models show similar results with the SVR model when we use long-term period data.

• 한국인공지능학회지
• /
• 제12권1호
• /
• pp.17-24
• /
• 2024

This paper introduces a novel approach to time-series estimation for energy load forecasting within Virtual Power Plant (VPP) systems, leveraging advanced artificial intelligence (AI) algorithms, namely Long Short-Term Memory (LSTM) and Seasonal Autoregressive Integrated Moving Average (SARIMA). Virtual power plants, which integrate diverse microgrids managed by Energy Management Systems (EMS), require precise forecasting techniques to balance energy supply and demand efficiently. The paper introduces a hybrid-method forecasting model combining a parametric-based statistical technique and an AI algorithm. The LSTM algorithm is particularly employed to discern pattern correlations over fixed intervals, crucial for predicting accurate future energy loads. SARIMA is applied to generate time-series forecasts, accounting for non-stationary and seasonal variations. The forecasting model incorporates a broad spectrum of distributed energy resources, including renewable energy sources and conventional power plants. Data spanning a decade, sourced from the Korea Power Exchange (KPX) Electrical Power Statistical Information System (EPSIS), were utilized to validate the model. The proposed hybrid LSTM-SARIMA model with parameter sets (1, 1, 1, 12) and (2, 1, 1, 12) demonstrated a high fidelity to the actual observed data. Thus, it is concluded that the optimized system notably surpasses traditional forecasting methods, indicating that this model offers a viable solution for EMS to enhance short-term load forecasting.

• 전기전자학회논문지
• /
• 제23권1호
• /
• pp.134-142
• /
• 2019

에너지인터넷 기술의 발전과 다양한 전자기기의 보급으로 에너지소비량이 패턴이 다양해짐에 따라 수요예측에 대한 신뢰도가 감소하고 있어 발전량 최적화 및 전력공급 안정화에 문제를 야기하고 있다. 본 연구에서는 고신뢰성을 갖는 수요예측을 위해 딥러닝 기법인 Convolution neural network(CNN)과 Bidirectional Long Short-Term Memory(BLSTM)을 융합한 1Dimention-Convolution and Bidirectional LSTM(1D-ConvBLSTM)을 제안하고, 제안한 기법을 활용하여 시계열 에너지소비량대한 소비패턴을 효과적으로 추출한다. 실험 결과에서는 다양한 반복학습 횟수와 feature map에 대해서 수요를 예측하고 적은 반복학습 횟수로도 테스트 데이터의 그래프 개형을 예측하는 것을 검증한다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 A
• /
• pp.722-724
• /
• 2004

Demand estimates in electric power systems have traditionally consisted of time-series analyses over long time periods. The resulting database consisted of huge amounts of data that were then analyzed to create the various coefficients used to forecast power demand. In this research, we take advantage of universally used analysis techniques analysis, but we also use easily available data-mining techniques to analyze patterns of days and special days(holidays, etc.). We then present a new method for estimating and forecasting power flow using decision tree analysis. And because analyzing the relationship between the estimate and power system ceiling Trices currently set by the Korea Power Exchange. We included power system ceiling prices in our estimate coefficients and estimate method.

• 한국빅데이터학회지
• /
• 제4권1호
• /
• pp.63-72
• /
• 2019

전력 에너지의 경우 발전 및 송전 과정을 거쳐 사용자에게 제공된 이후에는 회수가 불가능하기 때문에 정확한 수요 예측에 기반한 최적 발전 및 송배전 계획이 필요하다. 전력 수요 예측의 실패는 2011년 9월에 발생한 대규모 정전사태와 같이 다양한 사회적·경제적 문제를 야기할 수 있다. 전력 수요 예측 관련 기존 연구에서는 ARIMA, 신경망모형 등 다양한 방법으로 개발이 되었다. 하지만 전국 단위의 평균 외기온도를 사용한다는 점과, 계절성을 구분하기 위한 획일적 기준을 적용하는 한계점으로 인해 데이터의 왜곡이나 예측모형의 성능 저하를 초래하고 있다. 이에 본 연구에서는 전력 수요 예측 모형의 성능을 향상하기 위해 전국을 5대 권역으로 구분하여 지역적 특성과 이동 기간 학습 기법을 통해 계절적 특성을 반영한 선형회귀모형과 신경망 모형의 장기적 전력 수요 예측 모형을 개발하였다. 이를 통해 중장기부터 단기에 이르기까지 다양한 범위의 수요 예측에 해당 모델을 활용할 수 있을 뿐만 아니라 특정 기간 중에 발생하는 다양한 이벤트와 예외 상황을 고려할 수 있을 것이다.

• ETRI Journal
• /
• 제46권3호
• /
• pp.379-391
• /
• 2024

To meet increasing traffic requirements in mobile networks, small base stations (SBSs) are densely deployed, overlapping existing network architecture and increasing system capacity. However, densely deployed SBSs increase energy consumption and interference. Although these problems already exist because of densely deployed SBSs, even more SBSs are needed to meet increasing traffic demands. Hence, base station (BS) switching operations have been used to minimize energy consumption while guaranteeing quality-of-service (QoS) for users. In this study, to optimize energy efficiency, we propose the use of deep reinforcement learning (DRL) to create a BS switching operation strategy with a traffic prediction model. First, a federated long short-term memory (LSTM) model is introduced to predict user traffic demands from user trajectory information. Next, the DRL-based BS switching operation scheme determines the switching operations for the SBSs using the predicted traffic demand. Experimental results confirm that the proposed scheme outperforms existing approaches in terms of energy efficiency, signal-to-interference noise ratio, handover metrics, and prediction performance.