(Big) data analytics in smart grids

E. Mocanu; H.P. Nguyen; M. Gibescu

(Big) data analytics in smart grids

E. Mocanu
, H.P. Nguyen
, M. Gibescu

Electrical Energy Systems

Research output: Contribution to conference › Poster

154 Downloads (Pure)

Abstract

Prediction of building energy consumption is a fundamental problem in the smart grid context. Unprecedented high volumes of data and information are available with the upward growth of the smart metering infrastructure. Therefore, we develop two deep learning methods. Firstly, the demand forecasting problem was solved at low aggregation levels (i.e. 1900 buildings) using factored conditional restricted Boltzmann machine. Secondly, we developed an unsupervised energy prediction method using reinforcement cross-building transfer able to accurately estimate the energy based on the information available in the neighborhood. Both methods have been successfully validated on real-world databases.

Original language	English
Publication status	Published - 29 Jun 2016
Event	European Data Forum 2016 (EDF 2016), June 29-30, 2016, Eindhoven, The Netherlands - Eindhoven, Netherlands Duration: 29 Jun 2016 → 30 Jun 2016 http://2016.data-forum.eu/

Conference

Conference	European Data Forum 2016 (EDF 2016), June 29-30, 2016, Eindhoven, The Netherlands
Abbreviated title	EDF 2016
Country/Territory	Netherlands
City	Eindhoven
Period	29/06/16 → 30/06/16
Other	"Scaling up the European data economy"
Internet address	http://2016.data-forum.eu/

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Access to Document

edf2016ElenaFinal author version , 1.44 MB

Cite this

@conference{da7a2f46fa834663b63f497ac3570392,

title = "(Big) data analytics in smart grids",

abstract = "Prediction of building energy consumption is a fundamental problem in the smart grid context. Unprecedented high volumes of data and information are available with the upward growth of the smart metering infrastructure. Therefore, we develop two deep learning methods. Firstly, the demand forecasting problem was solved at low aggregation levels (i.e. 1900 buildings) using factored conditional restricted Boltzmann machine. Secondly, we developed an unsupervised energy prediction method using reinforcement cross-building transfer able to accurately estimate the energy based on the information available in the neighborhood. Both methods have been successfully validated on real-world databases.",

author = "E. Mocanu and H.P. Nguyen and M. Gibescu",

year = "2016",

month = jun,

day = "29",

language = "English",

note = "European Data Forum 2016 (EDF 2016), June 29-30, 2016, Eindhoven, The Netherlands, EDF 2016 ; Conference date: 29-06-2016 Through 30-06-2016",

url = "http://2016.data-forum.eu/",

}

TY - CONF

T1 - (Big) data analytics in smart grids

AU - Mocanu, E.

AU - Nguyen, H.P.

AU - Gibescu, M.

PY - 2016/6/29

Y1 - 2016/6/29

N2 - Prediction of building energy consumption is a fundamental problem in the smart grid context. Unprecedented high volumes of data and information are available with the upward growth of the smart metering infrastructure. Therefore, we develop two deep learning methods. Firstly, the demand forecasting problem was solved at low aggregation levels (i.e. 1900 buildings) using factored conditional restricted Boltzmann machine. Secondly, we developed an unsupervised energy prediction method using reinforcement cross-building transfer able to accurately estimate the energy based on the information available in the neighborhood. Both methods have been successfully validated on real-world databases.

AB - Prediction of building energy consumption is a fundamental problem in the smart grid context. Unprecedented high volumes of data and information are available with the upward growth of the smart metering infrastructure. Therefore, we develop two deep learning methods. Firstly, the demand forecasting problem was solved at low aggregation levels (i.e. 1900 buildings) using factored conditional restricted Boltzmann machine. Secondly, we developed an unsupervised energy prediction method using reinforcement cross-building transfer able to accurately estimate the energy based on the information available in the neighborhood. Both methods have been successfully validated on real-world databases.

M3 - Poster

T2 - European Data Forum 2016 (EDF 2016), June 29-30, 2016, Eindhoven, The Netherlands

Y2 - 29 June 2016 through 30 June 2016

ER -

(Big) data analytics in smart grids

Abstract

Conference

UN SDGs

Access to Document

Fingerprint

Cite this