Developing portable Energy Benchmarking methods

Considerable effort is required to consolidate diverse data sets for carrying out energy benchmarking studies at a site level. This could range from non-availability of smart meter data, sparse metering data, ambiguous metering hierarchy to name a few. We have developed an energy reporting application which is deployable at scale, normalized for weather, and based on semantic models. This application has been deployed on multiple sites. Deployment of the application uses a transformational automated approach, an improvement over the typical process where analytics providers must manually configure their energy benchmarking applications for each site. This results in deployment of the model in a fraction of the time that would be normally required for a model that is hardwired to the building data. However, developing and using these semantic models comes with its own challenges. This paper will describe the steps involved and the challenges faced in developing and trialing a platform that embodies this process.