Probabilistic Load Forecasting Optimization for Building Energy Models via Day Characterization.

摘要: Accurate load forecasting in buildings plays an important role for grid operators, demand response aggregators, building energy managers, owners, customers, etc. Probabilistic (PLF) becomes essential to understand and manage the building’s energy-saving potential. This research explains a methodology optimize results of PLF using daily characterization forecast. The forecast provided by calibrated white-box model real weather was classified hierarchically selected perform kernel density estimation (KDE) only similar days from database characterized quantitatively qualitatively. A case study is presented show office located Pamplona, Spain. monitoring, both inside—thermal sensors—and outside—weather station—is key when implementing this optimization technique. showed that thanks characterization, it possible accuracy probabilistic forecasting, reaching values close 100% some cases. In addition, explained scalable can be used initial stages its implementation, improving obtained as increases with information each new day.