AMI Intelligence Connected Building Energy Modeling

VEIC proposes to demonstrate a building simulation modeling tool that links backwards looking data intelligence with forward projecting Open Studio model simulations. The tool would use Advanced Metering Infrastructure (AMI) data and regression methods to identify customer specifics about when and how they use their facility and then populate the what-if scenario building energy modeling simulations with those insights. Output would not only be efficiency measure suggestions, but also energy savings customized to the customer's specific building situation and build through energy modeling. VEIC plans to demonstrate this technology on grocery stores, K-12 schools, and non-hospital health facilities, but the tool could be applicable to many common building types. Total quantity of prior usage AMI data necessary would ideally be twelve months; however, the analytical methods will work with significantly less. VEIC needs to do more testing to understand how the confidence with which the tool may make recommendations might adjust based on the depth of the available input information. VEIC will leverage the ASHRAE Guideline 14 to define a confidence metric for the model virtualization. The tools will rely on a small and simple subset of building specific information as being provided by the user, as well as providing an AMI data file. The details are expected to include size, age, location, function, and some HVAC particulars. The interface will then use these inputs to enter into the automation portion of the workflow where additional site specifics are inferred from regression analytics performed on the AMI meter.   The efficiencies identified by this automated energy analyst are expected to be in the range of 5 kBtu/sqft/year which may provide a range of 5-20% overall improvement, depending upon the specific customer. While tool usage won’t directly deliver savings, the customer will realize savings by acting upon the recommendations delivered by the tool.  The scope is to support control-based strategies for energy savings; however, once a model is constructed it would be possible to add in options for additional types of energy efficiency strategies, measures like equipment upgrades and weatherization. The expectation is that the ease of use, customized savings analysis, and persistent presence of the tools becomes a catalyst for users to continually make efficiency enhancements beyond the original suggested control-based strategies.