A data-driven approach to extract operational signatures of HVAC systems and analyze impact on electricity consumption

Abstract The electricity consumption of Heating Ventilating and Air Conditioning (HVAC) systems has a significant share in the energy consumption of buildings, which account for 75% of total electricity produced in the US. Therefore, improving the energy efficiency in HVAC systems is an essential goal in facility management (FM) industry. Building Automation Systems (BASs) deployed in buildings provide an enormous amount of data on HVAC operations, which can be leveraged to extract hidden knowledge and insights about operational signatures of these systems (i.e., parameter-value pairs set for running the equipment) and their relationship to energy profiles. This study aims to identify critical parameters of HVAC systems that drive the changes in the building energy-use profiles and develop an automated approach for identifying HVAC operational signatures and their energy profiles in buildings. The approach relies on data-driven methodologies and is composed of three major steps: data preprocessing, feature selection, and signature discovery and analysis. The approach was tested on four air handling units (AHUs) in different buildings. The results showed that it is possible to define operational signatures for facility operators to run AHUs at these custom settings and achieve about 30% saving in electric power, given the profiles across the operational signatures.

[1]  Zhongheng Zhang,et al.  Missing data imputation: focusing on single imputation. , 2016, Annals of translational medicine.

[2]  Shengwei Wang,et al.  An intelligent chiller fault detection and diagnosis methodology using Bayesian belief network , 2013 .

[3]  Yan Lu,et al.  Optimal building energy management using intelligent optimization , 2013, 2013 IEEE International Conference on Automation Science and Engineering (CASE).

[4]  Arno Schlueter,et al.  A review of unsupervised statistical learning and visual analytics techniques applied to performance analysis of non-residential buildings , 2018 .

[5]  Luis Pérez-Lombard,et al.  A review on buildings energy consumption information , 2008 .

[6]  Lh Batista,et al.  Avaliação da amplitude articular do joelho: correlação entre as medidas realizadas com o goniômetro universal e no dinamômetro isocinético , 2006 .

[7]  Manish Marwah,et al.  Sustainable operation and management of data center chillers using temporal data mining , 2009, KDD.

[8]  Andrew Kusiak,et al.  Optimization of an HVAC system with a strength multi-objective particle-swarm algorithm , 2011 .

[9]  Henrik Madsen,et al.  Temporal knowledge discovery in big BAS data for building energy management , 2015 .

[10]  Martin Fischer,et al.  A method to compare simulated and measured data to assess building energy performance , 2012 .

[11]  Ryozo Ooka,et al.  Optimization of the HVAC system design to minimize primary energy demand , 2014 .

[12]  N.D. Hatziargyriou,et al.  Two-Stage Pattern Recognition of Load Curves for Classification of Electricity Customers , 2007, IEEE Transactions on Power Systems.

[13]  K. P. Singh,et al.  Support vector machines in water quality management. , 2011, Analytica chimica acta.

[14]  Zhimin Du,et al.  Fault detection and diagnosis based on improved PCA with JAA method in VAV systems , 2007 .

[15]  Pieter de Wilde,et al.  The gap between predicted and measured energy performance of buildings: A framework for investigation , 2014 .

[16]  Constantine Kontokosta,et al.  Grading buildings on energy performance using city benchmarking data , 2019, Applied Energy.

[17]  Frédéric Magoulès,et al.  Feature Selection for Predicting Building Energy Consumption Based on Statistical Learning Method , 2012 .

[18]  Lihua Xie,et al.  HVAC system optimization—in-building section , 2005 .

[19]  Chih-Fong Tsai,et al.  Feature selection in bankruptcy prediction , 2009, Knowl. Based Syst..

[20]  Luis Pérez-Lombard,et al.  The map of energy flow in HVAC systems , 2011 .

[21]  J. Zico Kolter,et al.  A Large-Scale Study on Predicting and Contextualizing Building Energy Usage , 2011, AAAI.

[22]  Fu Xiao,et al.  Data mining in building automation system for improving building operational performance , 2014 .

[23]  John E. Seem,et al.  Pattern recognition algorithm for determining days of the week with similar energy consumption profiles , 2005 .

[24]  R. Seidl,et al.  Trend Analysis for Commissioning , 2006 .

[25]  Shengwei Wang,et al.  Sensor-fault detection, diagnosis and estimation for centrifugal chiller systems using principal-component analysis method , 2005 .

[26]  Fu Xiao,et al.  Bayesian network based FDD strategy for variable air volume terminals , 2014 .

[27]  Zyad Shaaban,et al.  Normalization as a Preprocessing Engine for Data Mining and the Approach of Preference Matrix , 2006, 2006 International Conference on Dependability of Computer Systems.

[28]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[29]  Pawel Lewicki,et al.  Statistics : methods and applications : a comprehensive reference for science, industry, and data mining , 2006 .

[30]  Ron Shamir,et al.  Scoring clustering solutions by their biological relevance , 2003, Bioinform..

[31]  Thomas R. Kurfess,et al.  Detection of Wheel and Workpiece Contact/Release in Reciprocating Surface Grinding , 2003 .

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

[33]  Burcin Becerik-Gerber,et al.  Infrared thermography of human face for monitoring thermoregulation performance and estimating personal thermal comfort , 2016 .

[34]  Cheng Zhou,et al.  Chiller sensor fault detection using a self-Adaptive Principal Component Analysis method , 2012 .

[35]  Kuei-Peng Lee,et al.  A simulation–optimization approach for energy efficiency of chilled water system , 2012 .

[36]  Theodoros Damoulas,et al.  Towards data-driven energy consumption forecasting of multi-family residential buildings: Feature selection via the Lasso , 2014 .

[37]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[38]  T. Rajendran,et al.  Feature subset selection for irrelevant data removal using Decision Tree Algorithm , 2013, 2013 Fifth International Conference on Advanced Computing (ICoAC).

[39]  Arno Schlueter,et al.  Automated daily pattern filtering of measured building performance data , 2015 .

[40]  Erik Strumbelj,et al.  Explaining prediction models and individual predictions with feature contributions , 2014, Knowledge and Information Systems.

[41]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[42]  F. Gubina,et al.  Consumers' load profile determination based on different classification methods , 2003, 2003 IEEE Power Engineering Society General Meeting (IEEE Cat. No.03CH37491).

[43]  Fu Xiao,et al.  AHU sensor fault diagnosis using principal component analysis method , 2004 .

[44]  Andrew Kusiak,et al.  Modeling and optimization of HVAC energy consumption , 2010 .

[45]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[46]  P. Postolache,et al.  Load pattern-based classification of electricity customers , 2004, IEEE Transactions on Power Systems.

[47]  Robert McDowall,et al.  Fundamentals of HVAC systems , 2007 .

[48]  Burcin Becerik-Gerber,et al.  Real-time activity recognition for energy efficiency in buildings , 2018 .

[49]  Tianqi Chen,et al.  XGBoost: A Scalable Tree Boosting System , 2016, KDD.

[50]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[51]  Michael E. Webber,et al.  Clustering analysis of residential electricity demand profiles , 2014 .

[52]  Olga Lyapina,et al.  Cluster analysis of European surface ozone observations for evaluation of MACC reanalysis data , 2016 .