A predictive preference model for maintenance of a heating ventilating and air conditioning system

Abstract Predicting next failure of the filter's differential pressure of heating ventilating and air conditioning (HVAC) system provides for a higher performance of the system. There exist various fluctuating parameters that contribute in this paramount prediction. In the current study, the traditional method of linear regression and artificial neural network are applied as means of prediction, and it is shown that the performance is improved when supplemented with a decision tree approach. The outcome reveals which one can more effectively predict trends and behavioral patterns as well as maintenance requirement of such systems with limited considered attributes. Hence, the empirical data is retrieved and a new method for predictive maintenance illustrated using HVAC system of Ecole de technologie superieure (ETS).

[1]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[2]  Michel Cabassud,et al.  Modelling of ultrafiltration fouling by neural network , 1998 .

[3]  Kuriakose Athappilly,et al.  A comparative predictive analysis of neural networks (NNs), nonlinear regression and classification and regression tree (CART) models , 2005, Expert Syst. Appl..

[4]  Manas Ranjan Patra,et al.  NETWORK INTRUSION DETECTION USING NAÏVE BAYES , 2007 .

[5]  Wright-Patterson Afb,et al.  Feature Selection Using a Multilayer Perceptron , 1990 .

[6]  Aditya Parida,et al.  An integrated approach for open e-Maintenance: Opportunities and challenges , 2010 .

[7]  Xin Yao,et al.  A constructive algorithm for training cooperative neural network ensembles , 2003, IEEE Trans. Neural Networks.

[8]  S. J. Press,et al.  Choosing between Logistic Regression and Discriminant Analysis , 1978 .

[9]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[10]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[11]  D. J. Greenwood,et al.  Statistical methods for evaluating a crop nitrogen simulation model, N_ABLE , 2000 .

[12]  Richard C. Daniel,et al.  Filtration Understanding: FY10 Testing Results and Filtration Model Update , 2011 .

[13]  David J. Evans,et al.  Application of a real-time predictive maintenance system to a production machine system , 2005 .

[14]  Jacek M. Zurada,et al.  Training neural network classifiers for medical decision making: The effects of imbalanced datasets on classification performance , 2008, Neural Networks.

[15]  Wen Wang.  Stochasticity, nonlinearity and forecasting of streamflow processes , 2006 .

[16]  Thierry Denoeux,et al.  A k-nearest neighbor classification rule based on Dempster-Shafer theory , 1995, IEEE Trans. Syst. Man Cybern..

[17]  S. Menard.  Applied Logistic Regression Analysis , 1996 .

[18]  Brad Warner,et al.  Understanding Neural Networks as Statistical Tools , 1996 .

[19]  R. O. Gilbert.  Statistical Methods for Environmental Pollution Monitoring , 1987 .

[20]  Janet M. Twomey,et al.  Validation and Verification , 1997 .