Time Series Electricity Consumption Analysis using Non-negative Matrix Factorization

For developing a sustainable society, energy management systems are utilized in many organizations. Chiba University of Commerce (CUC) is one of the organizations that has completely switched to renewable energy-sourced electricity for the first time in Japan. In the campus, energy consumption due to air conditioning, lightning and so on at each room is monitored. These monitoring data are stored on a data server via smart meters. In order to promote awareness to reduce electricity consumption, we need to summarize a vast amount of data so that we can interpret the data easily, and find out where we can afford to save electricity consumption. In this paper, we employ non-negative matrix factorization (NMF) for summarizing time-series electricity consumption patterns to analyze the electricity consumption data over time. Through the data analysis, we show that the visualization of factor matrices by dimensionality reduction enables us easily to interpret the low level electricity consumption data, and it gives us some awareness on energy saving.

[1]  Yoshiyuki Shimoda,et al.  A STUDY ON DISAGGREGATING THE SMART METER DATA INTO ELECTRICITY ENERGY END-USES , 2018 .

[2]  Michael W. Berry,et al.  Algorithms and applications for approximate nonnegative matrix factorization , 2007, Comput. Stat. Data Anal..

[3]  P. Smaragdis,et al.  Non-negative matrix factorization for polyphonic music transcription , 2003, 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (IEEE Cat. No.03TH8684).

[4]  Krzysztof Gajowniczek,et al.  Data Mining Techniques for Detecting Household Characteristics Based on Smart Meter Data , 2015 .

[5]  Gianfranco Chicco,et al.  Overview and performance assessment of the clustering methods for electrical load pattern grouping , 2012 .

[6]  Ram Rajagopal,et al.  Smart Meter Driven Segmentation: What Your Consumption Says About You , 2013, IEEE Transactions on Power Systems.

[7]  Mohammad Yusri Hassan,et al.  A review on applications of ANN and SVM for building electrical energy consumption forecasting , 2014 .

[8]  H. Ochiai.  Power data management on the Internet space: Green ICT projects in Japan , 2012, 2012 IEEE Colombian Communications Conference (COLCOM).

[9]  Michael Conlon,et al.  A clustering approach to domestic electricity load profile characterisation using smart metering data , 2015 .

[10]  Raul Kompass,et al.  A Generalized Divergence Measure for Nonnegative Matrix Factorization , 2007, Neural Computation.

[11]  Christoph Flath,et al.  Cluster Analysis of Smart Metering Data , 2012, Business & Information Systems Engineering.

[12]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[13]  Jérôme Idier,et al.  Algorithms for Nonnegative Matrix Factorization with the β-Divergence , 2010, Neural Computation.

[14]  Benjamin C. M. Fung,et al.  A systematic procedure to study the influence of occupant behavior on building energy consumption , 2011 .

[15]  M. N. Assimakopoulos,et al.  Using intelligent clustering techniques to classify the energy performance of school buildings , 2007 .

[16]  Lai Guan Ng,et al.  Dimensionality reduction for visualizing single-cell data using UMAP , 2018, Nature Biotechnology.