In recent decades, the evolution of loads and distributed energy resources has added complexity and uncertainty to distribution networks. A reliable characterization of the uncertainty associated with prediction is fundamental to the wide range of newly emerging applications in the low-voltage level grid.
In a paper published in IEEE Transactions on Smart Grid, engineers from EPE proposed a novel method to solve the short-term probabilistic load forecasting (STPLF) problem in distribution networks in which the loads are usually too volatile to be forecasted accurately. The approach employed a Dirichlet process mixture model (DPMM) to handle the uncertainty of load pattern. The DPMM representation of the load patterns was combined with a tree-based ensemble learning method to address the STPLF by solving a classification problem.
The results obtained demonstrated that the proposed approach outperformed the benchmark methods in STPLF at the given aggregation levels.
Visit IEEE’s website to read the study.