Energy Predictive Models with Limited Data using Transfer Learning
This addresses the challenge of insufficient training data for energy forecasting, which is incremental as it applies transfer learning to a known CNN approach.
The paper tackles the problem of developing predictive models for energy assets with limited historical data by proposing a transfer learning strategy on a convolutional neural network (CNN) model, resulting in significant improvement over existing forecasting methods for daily electricity demand.
In this paper, we consider the problem of developing predictive models with limited data for energy assets such as electricity loads, PV power generations, etc. We specifically investigate the cases where the amount of historical data is not sufficient to effectively train the prediction model. We first develop an energy predictive model based on convolutional neural network (CNN) which is well suited to capture the interaday, daily, and weekly cyclostationary patterns, trends and seasonalities in energy assets time series. A transfer learning strategy is then proposed to address the challenge of limited training data. We demonstrate our approach on a usecase of daily electricity demand forecasting. we show practicing the transfer learning strategy on the CNN model results in significant improvement to existing forecasting methods.