The proposed automated metalearning framework plays a crucial role in the smart grid community by addressing the challenges associated with limited historical data and the need to optimize hyperparameters for time-series forecasting (TSF) in microgrid scenarios. This is the first time that a bilevel framework has been applied to an automated metalearning approach in the smart grid community. In remote or off-grid areas, it helps overcome data limitation by employing a lower level metalearner that mitigates the challenge. By leveraging the knowledge acquired from various energy sources, this approach enables more robust forecasting even with limited data, enhancing the reliability and efficiency of the smart grid system. Manually selecting suitable hyperparameters can be time consuming and suboptimal. The proposed algorithm saves effort and ensures that the TSF models are fine-tuned for optimal performance, leading to more accurate predictions and improved decision-making. Moreover, it ...

Microgrid can improve greenhouse gas emissions and reduce operational costs. To forecast both energy generation and load demand, time-series prediction has been a key tool in real-time control and optimization. Developing an adequate predictive model is difficult when there is a lack of historical data. Moreover, hyperparameters have a tangible impact on the performance of machine learning models. Bearing these considerations in mind, this article develops a BiLO-Auto-TSF/ML framework that automates the optimal design of a few-shot learning pipeline from a bilevel programming perspective. Specifically, a lower level metalearner helps mitigate the small data challenge, whereas an upper level optimization optimizes both hyperparameters for lower and upper level learners. Note that the proposed framework is designed to accommodate a wide range of machine learning methods, allowing for easy integration through a plug-in mechanism. Comprehensive experiments demonstrate the effectiveness of ...