人工神经网络
期限(时间)
图层(电子)
计算机科学
人工智能
电力系统
功率(物理)
数据挖掘
短时记忆
机器学习
循环神经网络
物理
量子力学
化学
有机化学
作者
Bo‐Sung Kwon,Rae‐Jun Park,Kyung‐Bin Song
标识
DOI:10.1007/s42835-020-00424-7
摘要
Short-term load forecasting (STLF) is essential for power system operation. STLF based on deep neural network using LSTM layer is proposed. In order to apply the forecasting method to STLF, the input features are separated into historical and prediction data. Historical data are input to long short-term memory (LSTM) layer to model the relationships between past observed data. The outputs of the LSTM layer are incorporated with outputs of fully-connected layer in which prediction data, for instance weather information for forecasting day, are input. The optimal parameters of the proposed forecasting method are selected following several experiment. The proposed method is expected to contribute to stable power system operation by providing a precise load forecasting.
科研通智能强力驱动
Strongly Powered by AbleSci AI