Optimal generation scheduling and operating reserve management for PV generation using RNN-based forecasting models for stand-alone microgrids

Photovoltaic (PV) generation can drop sharply under reductions in solar irradiance due to cloud cover, resulting in adverse impacts on the performance of stand-alone microgrids. In addition, due to the stochastic nature of solar irradiance, it has remained challenging to forecast PV generation accurately. Therefore, a sufficient amount of operating reserve must be allocated for the variability and uncertainty of PV generation. However, excess operating reserve can decrease the economic efficiency of microgrids. This paper proposes an optimal generation scheduling and operating reserve management scheme for stand-alone PV-integrated microgrids, which comprises three core models. First, a two-stage recurrent neural network-based model is proposed for solar irradiance and cloud cover forecasting. Second, the operating reserve required for PV is managed more effectively based on weather conditions. Finally, an optimal generation scheduling model is applied based on the forecasting and operating reserve management models. The proposed scheme is applied to representative case studies to validate its advantages. The forecasting accuracy of the proposed model outperforms a conventional model. Under clearsky conditions, the operating reserve for PV can be reduced down to 30% of PV output. Subsequently, the optimal operation scheduled by the proposed scheme can produce substantial savings of more than 5.5%. (C) 2022 Elsevier Ltd. All rights reserved.

Automated summary

This paper proposes an optimal generation scheduling and operating reserve management scheme for stand-alone PV-integrated microgrids. It utilizes a recurrent neural network-based model for accurate solar irradiance and cloud cover forecasting, and effectively manages the operating reserve required for PV based on weather conditions. The proposed scheme demonstrates improved forecasting accuracy and substantial cost savings.