Hybrid methodologies for electricity load forecasting: Entropy-based feature selection with machine learning and soft computing techniques

Abstract

Worldwide scientific community is currently doing a great effort of research in the area of Smart Grids because energy production, distribution, and consumption play a critical role in the sustainability of the planet. The main challenge lies in intelligently integrating the actions of all users connected to the grid. In this context, electricity load forecasting methodologies is a key component for demand-side management. In this research the accuracy of different Machine Learning methodologies is determined for the hourly energy forecasting in buildings. The main goal of this work is to demonstrate the performance of these models and their scalability for different consumption profiles. We propose a hybrid methodology that combines feature selection based on entropies with soft computing and machine learning approaches, i.e. Fuzzy Inductive Reasoning, Random Forest and Neural Networks. They are compared with the traditional statistical time's series forecasting technique ARIMA in order to justify the capability of hybrid methods. In addition, in contrast to the general approaches where offline modelling takes considerable time, the approaches discussed in this work generate fast and reliable models, with low computational costs. These approaches could be embedded, for instance, in a second generation of smart meters, where they could generate onsite electricity forecasting of the next hours, or even trade the excess of energy with other smart meters.

A Permutation Importance-Based Feature Selection Method for Short-Term Electricity Load Forecasting Using Random Forest

“…This historical load dataset contains a total of 366 days. According to Jurado et al, a total of 9% of these days (33 days) are randomly selected as the test set, and the remaining 91% (333 days) are used as the training set [20]. When the load data are sampled in this city, the sampling is 1 h. The load curve of the entire year is shown in Figure 2.…”

Modeling public holidays in load forecasting: a German case study

“…Especially in studies where the focus in on other modeling aspects this approach is still popular, [6,9,[12][13][14][15][16][17][18].…”

A Flexible Fuzzy Inductive Reasoning Approach for Load Modelling Able to Cope With Missing Data

“…Although its popularity is not comparable to other Artificial Intelligence (AI) techniques such as Random Forest (RF), Artificial Neural Networks (ANN) or Supported Vector Machine (SVM). This methodology has been proved to model load consumptions with high accuracy compared to other typical AI and statistical techniques [14] [20]. Nevertheless, it has several problems when missing data are presented in the input variables of the model.…”