Sanitation and Analysis of Operation Data in Energy Systems

We present a workflow for data sanitation and analysis of operation data with the goal of increasing energy efficiency and reliability in the operation of building-related energy systems. The workflow makes use of machine learning algorithms and innovative visualizations. The environment, in which monitoring data for energy systems are created, requires low configuration effort for data analysis. Therefore the focus lies on methods that operate automatically and require little or no configuration. As a result a generic workflow is created that is applicable to various energy-related time series data; it starts with data accessibility, followed by automated detection of duty cycles where applicable. The detection of outliers in the data and the sanitation of gaps ensure that the data quality is sufficient for an analysis by domain experts, in our case the analysis of system energy efficiency. To prove the feasibility of the approach, the sanitation and analysis workflow is implemented and applied to the recorded data of a solar driven adsorption chiller.

[1]  Padhraic Smyth,et al.  Model selection for probabilistic clustering using cross-validated likelihood , 2000, Stat. Comput..

[2]  Srinivas Katipamula,et al.  Review Article: Methods for Fault Detection, Diagnostics, and Prognostics for Building Systems—A Review, Part I , 2005 .

[3]  Stefano Piva,et al.  EN 15316 Calculation Methods for the Generation Sub-system: The Influence of Input Data on the Results , 2014 .

[4]  Luís Torgo,et al.  Search-Based Class Discretization , 1997, ECML.

[5]  Gustaf Olsson,et al.  Instrumentation, Control and Automation in Wastewater Systems , 2015 .

[6]  Padhraic Smyth,et al.  Clustering Using Monte Carlo Cross-Validation , 1996, KDD.

[7]  Shengwei Wang,et al.  Development of prediction models for next-day building energy consumption and peak power demand using data mining techniques , 2014 .

[8]  Benjamin C. M. Fung,et al.  A novel methodology for knowledge discovery through mining associations between building operational data , 2012 .

[9]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[10]  Peter Palensky.  THE JEVIS SYSTEM-AN ADVANCED DATABASE FOR ENERGY-RELATED SERVICES , .

[11]  S. P. Lloyd,et al.  Least squares quantization in PCM , 1982, IEEE Trans. Inf. Theory.

[12]  Anil K. Jain.  Data clustering: 50 years beyond K-means , 2008, Pattern Recognit. Lett..

[13]  Eibe Frank,et al.  Conditional Density Estimation with Class Probability Estimators , 2009, ACML.

[14]  Steven T. Bushby,et al.  A rule-based fault detection method for air handling units , 2006 .

[15]  Richard Y. Wang,et al.  Data quality assessment , 2002, CACM.

[16]  F. W. Yu,et al.  Improved energy management of chiller systems by multivariate and data envelopment analyses , 2012 .

[17]  Jasmine A. Malinao,et al.  Improving energy efficiency of buildings using data mining technologies , 2014, 2014 IEEE 23rd International Symposium on Industrial Electronics (ISIE).

[18]  Padhraic Smyth,et al.  From Data Mining to Knowledge Discovery in Databases , 1996, AI Mag..

[19]  Le Yang,et al.  Data and analytics to inform energy retrofit of high performance buildings , 2014 .

[20]  Jian-Qiao Sun,et al.  Cross-level fault detection and diagnosis of building HVAC systems , 2011 .

[21]  Zoran Kapelan,et al.  Improved real-time data anomaly detection using context classification , 2011 .

[22]  M Mourad,et al.  A method for automatic validation of long time series of data in urban hydrology. , 2002, Water science and technology : a journal of the International Association on Water Pollution Research.