AutoPV: Automatically Design Your Photovoltaic Power Forecasting Model
This work addresses the problem of reducing labor costs and cross-domain expertise needed for PVPF model design, benefiting non-experts and industries in solar energy, though it is incremental as it applies NAS to a specific domain.
The paper tackles the challenge of constructing optimal predictive architectures for photovoltaic power forecasting (PVPF) by introducing AutoPV, a framework that uses neural architecture search (NAS) to automate model design, achieving superior performance compared to state-of-the-art predefined models in experiments on a dataset from the Daqing Photovoltaic Station in China.
Photovoltaic power forecasting (PVPF) is a critical area in time series forecasting (TSF), enabling the efficient utilization of solar energy. With advancements in machine learning and deep learning, various models have been applied to PVPF tasks. However, constructing an optimal predictive architecture for specific PVPF tasks remains challenging, as it requires cross-domain knowledge and significant labor costs. To address this challenge, we introduce AutoPV, a novel framework for the automated search and construction of PVPF models based on neural architecture search (NAS) technology. We develop a brand new NAS search space that incorporates various data processing techniques from state-of-the-art (SOTA) TSF models and typical PVPF deep learning models. The effectiveness of AutoPV is evaluated on diverse PVPF tasks using a dataset from the Daqing Photovoltaic Station in China. Experimental results demonstrate that AutoPV can complete the predictive architecture construction process in a relatively short time, and the newly constructed architecture is superior to SOTA predefined models. This work bridges the gap in applying NAS to TSF problems, assisting non-experts and industries in automatically designing effective PVPF models.