Accurate savings estimations are important to promote energy efficiency projects and demonstrate their cost-effectiveness. The increasing presence of advanced metering infrastructure (AMI) in commercial buildings has resulted in a rising availability of high frequency interval data. These data can be used for a variety of energy efficiency applications such as demand response, fault detection and diagnosis, and heating, ventilation, and air conditioning (HVAC) optimization. This large amount of data has also opened the door to the use of advanced statistical learning models, which hold promise for providing accurate building baseline energy consumption predictions, and thus accurate saving estimations. The gradient boosting machine is a powerful machine learning algorithm that is gaining considerable traction in a wide range of data driven applications, such as ecology, computer vision, and biology. In the present work an energy consumption baseline modeling method based on a gradient boosting machine was proposed. To assess the performance of this method, a recently published testing procedure was used on a large dataset of 410 commercial buildings. The model training periods were varied and several prediction accuracy metrics were used to evaluate the model's performance. The results show that using the gradient boosting machine model improved the R-squared prediction accuracy and the CV(RMSE) in more than 80 percent of the cases, when compared to an industry best practice model that is based on piecewise linear regression, and to a random forest algorithm.

Trustworthy savings calculations are critical to convincing investors in energy efficiency projects of the benefit and cost-effectiveness of such investments and their ability to replace or defer supply-side capital investments. However, today's methods for measurement and verification (M&V) of energy savings constitute a significant portion of the total costs of efficiency projects. They also require time-consuming manual data acquisition and often do not deliver results until years after the program period has ended. The rising availability of "smart" meters, combined with new analytical approaches to quantifying savings, has opened the door to conducting M&V more quickly and at lower cost, with comparable or improved accuracy. These meter-and software-based approaches, increasingly referred to as "M&V 2.0", are the subject of surging industry interest, particularly in the context of utility energy efficiency programs. Program administrators, evaluators, and regulators are asking how M&V 2.0 compares with more traditional methods, how proprietary software can be transparently performance tested, how these techniques can be integrated into the next generation of whole-building focused efficiency programs. This paper expands recent analyses of public-domain whole-building M&V methods, focusing on more novel M&V2.0 modeling approaches that are used in commercial technologies, as well as approaches that are documented in the literature, and/or developed by the academic building research community. We present a testing procedure and metrics to assess the performance of whole-building M&V methods. We then illustrate the test procedure by evaluating the accuracy of ten baseline energy use models, against measured data from a large dataset of 537 buildings. The results of this study show that the already available advanced interval data baseline models hold great promise for scaling the adoption of building measured savings calculations using Advanced Metering Infrastructure (AMI) data. Median coefficient of variation of the root mean squared error (CV(RMSE)) was less than 25% for every model tested when twelve months of training data were used. With even six months of training data, median CV(RMSE) for daily energy total was under 25% for all models tested. These findings can be used to build confidence in model robustness, and the readiness of these approaches for industry uptake and adoption.

No abstract

Building energy information systems (EIS) are a powerful customer-facing monitoring and analytical technology that can enable up to 20% site energy savings for buildings. Few technologies are as heavily marketed, but in spite of their potential, EIS remain an under-adopted emerging technology. One reason is the lack of information on purchase costs and associated energy savings. While insightful, the growing body of individual case studies has not provided industry the information needed to establish the business case for investment. Vastly different energy and economic metrics prevent generalizable conclusions. This paper addresses three common questions concerning EIS use: what are the costs, what have users saved, and which best practices drive deeper savings? We present a large-scale assessment of the value proposition for EIS use based on data from over two-dozen organizations. Participants achieved year-over-year median site and portfolio savings of 17% and 8%, respectively; they reported that this performance would not have been possible without the EIS. The median five-year cost of EIS software ownership (up-front and ongoing costs) was calculated to be $1,800 per monitoring point (kilowatt meter points were most common), with a median portfolio-wide implementation size of approximately 200 points.In this paper, we present an analysis of the relationship between key implementation factors and achieved energy reductions. Extent of efficiency projects, building energy performance prior to EIS installation, depth of metering, and duration of EIS were strongly correlated with greater savings. We also identify the best practices use of EIS associated with greater energy savings. IntroductionBuilding energy information systems (EIS) are broadly defined as the web-based analysis software, data acquisition hardware, and communication systems used to store, analyze, and display whole-building, system-level, or equipment-level energy use (Granderson et al. 2009;Motegi et al. 2003). Fig. 1 shows the schematic diagram of an EIS. At a minimum, an EIS provides hourly or sub-hourly interval meter data with graphical and analytical capabilities. The data in an EIS comes primarily from electric and gas meters, but can also include other data, such as those from building automation systems (BAS); the data integrated into the system depends on the level of monitoring that is present at the site. A data acquisition system in the building gathers the data and transmits it to a server that is on-site or on the cloud. The server storages and analyzes the data. External data sources such as weather data, or utility price and demand response information may in some cases be integrated into the EIS to support its analytical capabilities. EIS users can view the data and analysis results in graphical or report format through the user interface. A key set of EIS analytical capabilities (Granderson, Piette, and Rosenblum 2011;Kramer et al. 2013) include:

No abstract

Measured energy performance data are essential to national efforts to improve building efficiency, as evidenced in recent benchmarking mandates, and in a growing body of work that indicates the value of permanent monitoring and energy information feedback. This paper presents case studies of energy information systems (EIS) at four enterprises and university campuses, focusing on the attained energy savings, and successes and challenges in technology use and integration. EIS are broadly defined as performance monitoring software, data acquisition hardware, and communication systems to store, analyze, and display building energy information. Case investigations showed that the most common energy savings and instances of waste concerned scheduling errors, measurement and verification, and inefficient operations. Data quality is critical to effective EIS use, and is most challenging at the subsystem or component level, and with nonelectric energy sources. Sophisticated prediction algorithms may not be well understood but can be applied quite effectively, and sites with custom benchmark models or metrics are more likely to perform analyses external to the EIS. Finally, resources and staffing were identified as a universal challenge, indicating a need to identify additional models of EIS use that extend beyond exclusive in-house use, to analysis services.

No abstract

(PNNL) to identify monitoring and control needs for small-and medium-sized commercial buildings, and to recommend possible solutions. The scope of this study is to characterize the monitoring and controls needs for the various end uses (for both efficiency and demand response), determine requirements to develop control packages, and calculate the target cost of doing so. Section 1.0 introduces the study scope and analysis approaches used. Discussions regarding the number of buildings in the U.S that comprise "small-size" and "medium-size" buildings, their lack of building automation systems (BAS) and potential energy improvements, as well as challenges, are detailed in this section. Section 2.0 covers the characterization of both small-and medium-sized buildings. Drawing upon Energy Information Administration's Commercial Building Energy Consumption Survey data from various surveys, detailed discussions of energy end-use and electrical end-use consumption values are provided. This section spring boards into further discussions for the various end-use loads and the present penetration of "intelligent" controls in the existing market. Discussions of existing and possible future control methods, strategies and concepts that are applicable (including heating, ventilation and air conditioning (HVAC); lighting and miscellaneous end-use loads) complete this section. Section 3.0 discusses the different communication architectures that might be found in a small-or medium-sized building BAS, as it relates to the communication networks needed to support them. This discussion covers the different technologies that have been in place (older) or are becoming more prevalent (newer), and how they work. This includes wired solutions, wireless solutions or a combination of both (hybrid wired-wireless) networks and industry standards, open and proprietary protocols. For each solution, the limitations of each technology are detailed (speed, bandwidth, reliability, etc.). Cost factors are also discussed because this relates to how these systems are being pushed to the market, and their acceptance (or lack of). Section 4.0 describes the BAS, as has historically been seen and known in large building applications and the small-or medium-sized building applications. This section describes the history of BASs and how they have evolved and improved over time, and summarizes their core functions. This description proceeds to discuss the major architectural requirements needed by new BASs to allow for greater penetration in the existing building stock in the U.S. This section concludes by providing three different options of what a future BAS configuration might look like for either a small-sized building (two different options) or for a medium-sized building (one option). Section 5.0 presents the requirements and capabilities of various devices used to monitor and control different end-use loads found in small-and medium-sized buildings. This includes a robust presentation of the different requirements for the gateway, master controller, co...