Energy demand prediction for the implementation of an energy tariff emulator to trigger demand response in buildings

Buildings are key actors of the electrical gird. As such they have an important role to play in grid stabilization, especially in a context where renewable energies are mandated to become an increasingly important part of the energy mix. Demand response provides a mechanism to reduce or displace electrical demand to better match electrical production. Buildings can be a pool of flexibility for the grid to operate more efficiently. One of the ways to obtain flexibility from building managers and building users is the introduction of variable energy prices which evolve depending on the expected load and energy generation. In the proposed scenario, the wholesale energy price of electricity, a load prediction, and the elasticity of consumers are used by an energy tariff emulator to predict prices to trigger end user flexibility. In this paper, a cluster analysis to classify users is performed and an aggregated energy prediction is realised using Random Forest machine learning algorithm.

Download Full-text

Energy Flexibility Prediction for Data Center Engagement in Demand Response Programs

Sustainability ◽

10.3390/su12041417 ◽

2020 ◽

Vol 12 (4) ◽

pp. 1417 ◽

Cited By ~ 2

Author(s):

Andreea Valeria Vesa ◽

Tudor Cioara ◽

Ionut Anghel ◽

Marcel Antal ◽

Claudia Pop ◽

...

Keyword(s):

Energy Consumption ◽

Demand Response ◽

Prediction Error ◽

Energy Demand ◽

Business Processes ◽

Prediction Models ◽

Percentage Error ◽

Energy Prediction ◽

Demand Prediction ◽

Optimal Ensemble

In this paper, we address the problem of the efficient and sustainable operation of data centers (DCs) from the perspective of their optimal integration with the local energy grid through active participation in demand response (DR) programs. For DCs’ successful participation in such programs and for minimizing the risks for their core business processes, their energy demand and potential flexibility must be accurately forecasted in advance. Therefore, in this paper, we propose an energy prediction model that uses a genetic heuristic to determine the optimal ensemble of a set of neural network prediction models to minimize the prediction error and the uncertainty concerning DR participation. The model considers short term time horizons (i.e., day-ahead and 4-h-ahead refinements) and different aspects such as the energy demand and potential energy flexibility (the latter being defined in relation with the baseline energy consumption). The obtained results, considering the hardware characteristics as well as the historical energy consumption data of a medium scale DC, show that the genetic-based heuristic improves the energy demand prediction accuracy while the intra-day prediction refinements further reduce the day-ahead prediction error. In relation to flexibility, the prediction of both above and below baseline energy flexibility curves provides good results for the mean absolute percentage error (MAPE), which is just above 6%, allowing for safe DC participation in DR programs.

Download Full-text

Energy Demand Prediction of the Building Sector Based on Induced Kernel Method and MESSAGEix Model

Chinese Journal of Urban and Environmental Studies ◽

10.1142/s2345748119500167 ◽

2019 ◽

Vol 07 (04) ◽

pp. 1950016

Author(s):

Xin TAN ◽

Zijian ZHAO ◽

Changyi LIU ◽

Shining ZHANG ◽

Xing CHEN ◽

...

Keyword(s):

Energy Demand ◽

Historical Data ◽

Kernel Method ◽

Prediction Methods ◽

Building Sector ◽

Related Research ◽

Energy Prediction ◽

Demand Prediction ◽

Final Energy ◽

Energy Demand Prediction

The building sector, including resident, commercial and public services, is one of the most energy-intensive sectors nowadays. The share of buildings’ energy consumption in the final energy dramatically increases in various scenarios. As the preliminary work of the final energy prediction, the prediction of useful energy demand of the building sector is essential in the fields of energy-related research, especially for the scenarios design. To this end, this paper presents the prediction of energy demand in the building sector based on the Induced Kernel Method (IKM) for the useful energy. First, similar to other learning-based prediction methods, a database is constructed for the training. Specifically, the database contains not only the historical data of the useful energy demand and related indicators, but also some development templates to induce the prediction. Second, the detailed process is mathematically deduced to predict the useful energy demand components of the building sector, including electricity and heating. Finally, using various countries as examples, prediction results of the useful energy are presented in the numerical analysis. Furthermore, by using useful energy prediction results as the input of the MESSAGEix model, the paper further predicts global final energy of the building sector.

Download Full-text