A Tool for Efficient First Views of Commercial Building Energy Performance

This paper describes a proof of concept for a method of operational benchmarking. This benchmarking can be done with only daily weather data and the monthly energy bill information required for an Energy Star rating calculation, but the result is overtly functional, directed at improving operating efficiency. Several derived building attributes are compared with reference values to identify the general nature of major building inefficiencies. This “first view” can direct further investigation of the problem and potentially result in energy saving corrections. The underlying approach builds on past work on inverse modeling and energy signatures, discussed in detail herein, which shows useful performance insights proceeding from a monthly, temperature-correlated view of energy consumption. This process has produced promising anecdotal findings but has been difficult to scale up because it requires review by a reasonably experienced engineer or building operator. This new work standardizes and automates an enhanced regression process to support wide implementation, with pilot testing on a sample of 185 buildings across the country. The results demonstrate the feasibility of automating an energy balance-based Equivalent Analog Building Model that reproduces a building’s monthly energy usage in relation to outside temperature. For gas-heated office buildings, a relatively small set of eight key parameters has proved to produce robust and repeatable results. Automated messages based on indicators derived from the parameter values can suggest areas of particularly high efficiency or potential future improvement, categorized under broad areas of occupancy-related loads, facility structure and systems, and heating and cooling controls.