CAJ | 학술논문

在电力市场环境下，制定和调整当日负荷计划的周期缩短，尤其是滚动发电计划的提出，使现有负荷预测方法无法满足电力部门对预测精度和速度的要求，因此有人提出了扩展短期负荷预测的概念来解决这一问题，但该方法导致学习机的维数灾难问题难以避免。利用云计算技术对该方法进行改进，在避免了维数灾难的同时提高了预测精度与速度，并用East-Slovakia Power Distribution Company 提供的电网运行数据验证了该方法的有效性。
재전력시장배경하，제정화조정당일부하계화적주기축단，우기시곤동발전계화적제출，사현유부하예측방법무법만족전력부문대예측정도화속도적요구，인차유인제출료확전단기부하예측적개념래해결저일문제，단해방법도치학습궤적유수재난문제난이피면。이용운계산기술대해방법진행개진，재피면료유수재난적동시제고료예측정도여속도，병용East-Slovakia Power Distribution Company 제공적전망운행수거험증료해방법적유효성。
The cycle of formulating and adjusting the load plan is shortened in the context of electricity market,especially for making of the rolling plan of power generation. As the existing load forecasting method is not able to meet the requirements for the accuracy and speed in the power sector, some people put forward the concept of extended short-term load forecasting to solve this problem,but this approach leads to an unavoidable curse of dimensionality on learning machine. This paper uses cloud computing technology to improve the method,and succeeds in avoiding the dimension disaster and improving the accuracy and speed at the same time. The effec-tiveness of the proposed method is verified by data provided by East-Slovakia Power Distribution Company.