Abstract
Photovoltaic (PV) generation is potentially uncertain. Probabilistic PV generation forecasting methods have been proposed with prediction intervals (PIs) to evaluate the uncertainty quantitively. However, few studies have applied PIs to geographically distributed PVs in a specific area. In this study, a two-step probabilistic forecast scheme is proposed for geographically distributed PV generation forecasting. Each step of the proposed scheme adopts ensemble forecasting based on three different machine-learning methods. When individual PV generation is forecasted, the proposed scheme utilizes surrounding PVs’ past data to train the ensemble forecasting model. In this case study, the proposed scheme was compared with conventional non-multistep forecasting. The proposed scheme improved the reliability of the PIs and deterministic PV forecasting results through 30 days of continuous operation with real data in Japan.
Highlights
We introduce the idea of optical flow to data-driven methods, such as machine-learning-based methods, to improve existing probabilistic PV generation forecasting methods
The simulation result is evaluated based on four criteria: the cover rate of the prediction interval, the width of the prediction interval, mean average percentage error (MAPE), and root mean square error (RMSE)
The cover rate was improved from 72% to 100% using the multiple PV forecast model
Summary
Photovoltaic (PV) generation in a distribution network plays a key role in promoting clean energy production. The peak time and the amount of power flow depend on the demand and PV generation in the network. The peak is mitigated by energy storage systems (ESSs) operations, such as fixed batteries reported in [2,3]. The peak time and energy generated from the PVs must be forecasted to operate the ESS with the best efficiency. The PIs are evaluated using two fundamental but contradictory ideas: the coverage rate and the width of the intervals [4]. If the PIs cover all observations, the coverage rate is the best at 100%. As the PIs have a high coverage rate of observations and become narrower, the performance of the peak mitigation improves [3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
