Short-Term Load Forecasting With Different Aggregation Strategies

Abstract

Effective short-term load forecasting (STLF) plays an important role in demand-side management and power system operations. In this paper, STLF with three aggregation strategies are developed, which are information aggregation (IA), model aggregation (MA), and hierarchy aggregation (HA). The IA, MA, and HA strategies aggregate inputs, models, and forecasts, respectively, at different stages in the forecasting process. To verify the effectiveness of the three aggregation STLF, a set of 10 models based on 4 machine learning algorithms, i.e., artificial neural network, support vector machine, gradient boosting machine, and random forest, are developed in each aggregation group to predict 1-hour-ahead load. Case studies based on 2-year of university campus data with 13 individual buildings showed that: (a) STLF with three aggregation strategies improves forecasting accuracy, compared with benchmarks without aggregation; (b) STLF-IA consistently presents superior behavior than STLF based on weather data and STLF based on individual load data; (c) MA reduces the occurrence of unsatisfactory single-algorithm STLF models, therefore enhancing the STLF robustness; (d) STLF-HA produces the most accurate forecasts in distinctive load pattern scenarios due to calendar effects.

An Unsupervised Clustering-Based Short-Term Solar Forecasting Methodology Using Multi-Model Machine Learning Blending

“…Therefore, it can be concluded that both UC and M3 have improved the short-term GHI forecasting accuracy significantly. 3) Calendar and weather effects: It is reported in the literature that the forecasting accuracy of power time series, such as solar and load, is influenced by calendar effects [37] and weather effects [8]. To further explore the calendar and weather effects on the developed method, the best model(s) in each group is(are) picked out to make comparisons, which are C opt , SVR 3 , and GBM 3 in the UC-M3 group ({C opt,cm , SVR 3,cm , GBM 3,cm }∈M l,cm ), C opt and SVR 2 in the UC-SAML group ({C opt,cs , SVR 2,cs }∈M l,cs ), SVR 2 in the AIO-M3 group (SVR 2,am ∈M l,am ), and 1HA persistence of cloudiness method in the AIO-SAML group (P as ∈M l,as ).…”

Reinforcement Learning based Dynamic Model Selection for Short-Term Load Forecasting

“…For instance, Alamaniotis et al [7] linearly combined six Gaussian processes (GPs), which outperformed individual GPs. Feng et al [8] aggregated multiple artificial neural network (ANN), SVR, random forest (RF), and gradient boosting machine (GBM) models to mitigate the risk of choosing unsatisfactory models. Second, load patterns are classified into clusters based on some similarities to select the best load forecasting model in each cluster.…”