No abstract

No abstract

The field of energy forecasting has attracted many researchers from different fields (e.g., meteorology, data sciences, mechanical or electrical engineering) over the last decade. Solar forecasting is a fast-growing subdomain of energy forecasting. Despite several previous attempts, the methods and measures used for verification of deterministic (also known as single-valued or point) solar forecasts are still far from being standardized, making forecast analysis and comparison difficult. To analyze and compare solar forecasts, the well-established Murphy-Winkler framework for distribution-oriented forecast verification is recommended as a standard practice. This framework examines aspects of forecast quality, such as reliability, resolution, association, or discrimination, and analyzes the joint distribution of forecasts and observations, which contains all time-independent information relevant to verification. To verify forecasts, one can use any graphical display or mathematical/statistical measure to provide insights and summarize the aspects of forecast quality. The majority of graphical methods and accuracy measures known to solar forecasters are specific methods under this general framework. Additionally, measuring the overall skillfulness of forecasters is also of general interest. The use of the root mean square error (RMSE) skill score based on the optimal convex combination of climatology and persistence methods is highly recommended. By standardizing the accuracy measure and reference forecasting method, the RMSE skill score allows-with appropriate caveats-comparison of forecasts made using different models, across different locations and time periods.

No abstract

This paper presents a study into the effect of aggregation of customers and an increasing share of photovoltaic (PV) power in the net load on prediction intervals (PIs) of probabilistic forecasting methods applied to distribution grid customers during winter and spring. These seasons are shown to represent challenging cases due to the increased variability of electricity consumption during winter and the increased variability in PV power production during spring. We employ a dynamic Gaussian process (GP) and quantile regression (QR) to produce probabilistic forecasts on data from 300 de-identified customers in the metropolitan area of Sydney, Australia. In case of the dynamic GP, we also optimize the training window width and show that it produces sharp and reliable PIs with a training set of up to 3 weeks. In case of aggregation, the results indicate that the aggregation of a modest number of PV systems improves both the sharpness and the reliability of PIs due to the smoothing effect, and that this positive effect propagates into the net load forecasts, especially for low levels of aggregation. Finally, we show that increasing the share of PV power in the net load actually increases the sharpness and reliability of PIs for aggregations of 30 and 210 customers, most likely due to the added benefit of the smoothing effect.