Utilising the Cross Industry Standard Process for Data Mining to reduce uncertainty in the Measurement and Verification of energy savings

Loading...
Thumbnail Image
Files
Date
2016-06-14
Authors
Gallagher, Colm V.
Bruton, Ken
O'Sullivan, Dominic T. J.
Journal Title
Journal ISSN
Volume Title
Publisher
Springer International Publishing AG
Research Projects
Organizational Units
Journal Issue
Abstract
This paper investigates the application of Data Mining (DM) to predict baseline energy consumption for the improvement of energy savings estimation accuracy in Measurement and Verification (M&V). M&V is a requirement of a certified energy management system (EnMS). A critical stage of the M&V process is the normalisation of data post Energy Conservation Measure (ECM) to pre-ECM conditions. Traditional M&V approaches utilise simplistic modelling techniques, which dilute the power of the available data. DM enables the true power of the available energy data to be harnessed with complex modelling techniques. The methodology proposed incorporates DM into the M&V process to improve prediction accuracy. The application of multi-variate regression and artificial neural networks to predict compressed air energy consumption in a manufacturing facility is presented. Predictions made using DM were consistently more accurate than those found using traditional approaches when the training period was greater than two months.
Description
Keywords
Measurement and Verification , Data mining , Energy efficiency , Baseline energy modelling
Citation
Gallagher C. V., Bruton K. and O’Sullivan D. T. J. (2016) ‘Utilising the Cross Industry Standard Process for Data Mining to reduce uncertainty in the Measurement and Verification of energy savings’, in Tan Y. and Shi Y. (eds.) Data Mining and Big Data - DMBD 2016, Bali, Indonesia, 25-30 June. Lecture Notes in Computer Science, Vol. 9714. Springer International Publishing AG. doi:10.1007/978-3-319-40973-3_5
Copyright
© 2016, Springer International Publishing AG. The final publication is available at http://link.springer.com/chapter/10.1007%2F978-3-319-40973-3_5