Short-term load forecasting in non-residential Buildings

Abstract

Abstract-Short-term load forecasting (STLF) has become an essential tool in the electricity sector. It has been object of vast research since energy load is known to be non-linear and, therefore, very difficult to predict with accuracy. We focus here on non-residential building STLF, an special case of STLF where weather shows smaller influence on the load than in normal scenarios and forecast models, contrary to those on the literature, are required to be simple, avoiding dull and complicated trialand-error parametrisation or setting-up processes. Under these premises, we have used a two-step methodology comprising a classification and a adjustment steps. Since the non-linearity of the load is associated to the activity in the building, we have demonstrated that the best way to deal with it is using the work day schedule as day-type classifier. Moreover, we have evaluated a number of statistical methods and Artificial Intelligence methods to adjust the typical hourly consumption curve, concluding that an autoregressive time series suffices to fulfil the requirements, even in a 5 day-ahead horizon.

Short-term load forecasting in air-conditioned non-residential Buildings

“…In a previous work [3], we focused on a very special case of STLF, namely non-residential Buildings STLF. For non-residential buildings we understand schools, universities, public buildings and companies' facilities.…”

Short-term load forecasting in smart meters with sliding window-based ARIMA algorithms

“…Penya et al [9] present short-term load forecasting models for non-residential buildings. According to the authors, this special domain presents different characteristics: there is no consumption at night, or it is negligible, and anyway, there exists a notable gap between idle and activity times.…”

Building-level power demand forecasting framework using building specific inputs: Development and applications

“…A review of short term load forecasting (STLF) for non-residential buildings is provided in [11]. The authors note that load forecasting is frequently performed at the grid level (for a review, see [26]) but rarely for individual buildings, in spite of the potential usefulness of such information.…”

“…The authors note that load forecasting is frequently performed at the grid level (for a review, see [26]) but rarely for individual buildings, in spite of the potential usefulness of such information. In comparing several modeling approaches (neural networks, bayesian, auto-regressive (AR) and, auto-regressive integrated moving average (ARIMA)), the superiority and practicality of the AR modeling approach is highlighted [11]. AR models estimate the current power consumption based on previous power measurements only (for another example, see [27]).…”

“…We consider a time series based Auto Regressive eXogenous (ARX) modeling approach. However there are many variations, like ARIMAX and other data driven models or the use of Kalman Filter for online parameter estimation [11][12][13][14][15], which can be implemented for improved modeling accuracy. Rather than focus on the model quality, this work has a broader scope.…”