Day-ahead forecasting approach for energy consumption of an office building using support vector machines

Abstract

This paper presents a Support Vector Machine (SVM) based approach for energy consumption forecasting. The proposed approach includes the combination of both the historic log of past consumption data and the history of contextual information. By combining variables that influence the electrical energy consumption, such as the temperature, luminosity, seasonality, with the log of consumption data, it is possible for the proposed method by find patterns and correlations between the different sources of data and therefore improves the forecasting performance. A case study based on real data from a pilot microgrid located at the GECAD campus in the Polytechnic of Porto is presented. Data from the pilot buildings are used, and the results are compared to those achieved by several states of the art forecasting approaches. Results show that the proposed method can reach lower forecasting errors than the other considered methods.

Energy Consumption Forecasting Using Ensemble Learning Algorithms

“…By comparing the results of the AR2 method to the results of the SVM and some fuzzy rule-based methods, it is possible to conclude that it provides a more accurate consumption forecast. Additionally, the results presented in this paper in comparison to the results of the fuzzy rule-based methods presented in [16], namely SVM, HyFIS and WM, tend to forecast more reliable values and all the calculated errors are closer to the average error with exception to SVM that outperforms RF and GBR. The use of environmental variables such as a humidity and temperature proved to be useful in this experiment as well as the use of lagged features of previous hours.…”

“…The study presented in [16] addresses the electricity consumption forecast based on fuzzy rules methods of the same location as this study, namely using HyFIS, WM and SVM. In order to compare the results of the ensemble methods with the ones addressed in [16] we trained our models using the third training strategy and forecasted the 24 3.…”

“…The study presented in [16] addresses the electricity consumption forecast based on fuzzy rules methods of the same location as this study, namely using HyFIS, WM and SVM. In order to compare the results of the ensemble methods with the ones addressed in [16] we trained our models using the third training strategy and forecasted the 24 3. The comparison between the fuzzy rule based methods and the ensemble methods in Figure 2 and Table 3 shows that the AR2 method has the lower average forecasting error between these methods and can achieve better forecasting results than the other methods.…”

Demand Response Optimization Using Particle Swarm Algorithm Considering Optimum Battery Energy Storage Schedule in a Residential House

“…Each time that user update the forecast can perform a new optimization. Regarding the influence of the forecasting results on optimization, in the case that the presented day-ahead forecasting strategies in References [30,31] are considered, the forecasting error, using Supporter Vector Machine algorithms to predict the values for the next 24 h, will be 9.11%. Figure 3 with different colors.…”

IoH: A Platform for the Intelligence of Home with a Context Awareness and Ambient Intelligence Approach

“…The HyFIS, seen in Figure 13, receives a combination of four effective variables on the energy consumption value from past seven days as train data to create the forecasting model and based on the same variables from the current hour estimates the consumption value for the next hour. In the following of this process, a support vector machine (SVM) [35] classification algorithm has been used to identify the occupancy of the room. This classification method receives the estimated consumption value from HyFIS.…”