Buildings account for about 48 percent of the energy consumed in the United States. Of this energy, heating and cooling systems use about 55 percent, while lights and appliances use the other 35 percent of energy use of existing buildings. If energy-use trends continue, buildings will become the largest consumer of global energy by 2025. The development of building energy savings methods and models becomes apparently more necessary for a sustainable future. Most new buildings are equipped with modern building automation system BAS and direct digital control that have the ability to collect large amounts of data. However, even with modern technologies, those buildings are unfortunately still not operating optimally due to the absence of computational means and centralized solutions embedded into the BAS. Therefore, there is a significant need to investigate how modern computational techniques can help generate the analysis needed to gain full benefit from real-time data and at the same time perform many potential intelligent applications such as modelling, optimization, energy efficiency and energy assessment, and fault detection and diagnosis. This paper discusses the modeling methodologies for building energy system using time series auto regression artificial neural networks. The model can be integrated into energy solution tools for building energy assessment, optimization, fault detection and diagnosis, and many other applications. The model predicts whole building energy consumptions as function of four input variables, dry bulb and wet bulb outdoor air temperatures, hour of day and type of day. To train and test the models, data from twenty existing buildings and from simulations are used for the analysis. Advanced computational methods are used for data analysis and preprocessing. The wavelet basis is used to remove the noise and anomalies. Different neural network structures are tested along with various input delays to determine the one yielding the best results in term of mean square error. The results show that the developed model can provide results sufficiently accurate for its use in various energy efficiency and saving estimation applications.

Due to energy becoming a prominent topic in the sciences, creating baseline energy models for buildings has become a major area of research. Baseline models help determine dependency on ambient temperature or other parameters, while also helping show if energy reduction is due to building retrofits, occupancy, or ambient temperature. Several different methods of creating a baseline models for commercial and residential buildings, however few attempts have been made to create baseline energy models in industrial facilities, with only simple methods being analyzed. Since industrial facilities account for 33% of the annual energy usage within the United States (EIA), energy usage and finding energy saving opportunities in industrial facilities needs to be analyzed. While most industrial facilities energy usage is strongly dependent on production, some can be very dependent on weather especially if the facility is being conditioned. Even in an industrial facilities major energy consumption is used for production the energy usage might still be temperature dependent due to equipment being temperature dependent (i.e. chillers, furnaces, boilers, etc.). To understand an industrial facility’s energy usage is to create a baseline energy model. Currently only change-point regression models have been commonly used for analysis in industrial facilities. An analysis of the effectiveness of using Gaussian process regression (GPR) to develop a baseline energy usage models in industrial facilities from utility bill data and hourly data is presented in this paper.  Gaussian process regression uses a covariance matrix of the input variables to construct the model compared to using a pre-defined relationship between the input-output variables. By using a covariance matrix Gaussian process regression is more flexible than traditional parametric regression models. Two case studies are presented: using utility bill data to create a Gaussian process regression model and a using hourly data to create a Gaussian process regression model. In all cases the baseline regression models gave a CV-RMSE of 20% or lower showing that hourly data or utility bill data is capable of producing accurate baseline models defined by ASHRAE Guideline 14.

Simulations were conducted in order to make assessments regarding life cycle costs of hybrid ground source heat pump systems with different pumping configurations on an elementary school design.  The long term simulations evaluated the systems’ performance over a 20 year period. The basecase in this study is an all-ground heat exchanger (all-GHX) configuration.  A closed circuit cooling tower (CCCT) and dry fluid cooler (DFC) were separately modeled to provide heat rejection in the hybrid systems.  Life cycle costs of the all hybrid systems presented herein for an example building are estimated between 35-40% less than an all-GHX configuration.  The basecase life cycle cost is estimated at $894,000 while the hybrid options ranged from $565,000 to $578,000. These likely do not represent the lowest life cycle cost designs which would balance the sizing of the system components including the ground source heat exchanger and supplement heat rejection device with the associated energy costs’ present value.  In many instances, there are space constraints on sizing individual components, most notably the ground heat exchanger.  As such, the alternate hybrid designs arbitrarily utilize 140 bores which is one-half the size of the base design at 280 bores, with the addition of a supplement heat rejection device which is sized to provide acceptable borefield temperatures. The control scheme utilized across all hybrid systems considered common in industry (Thornton 2014),  inherently allows loop temperatures to elevate and forms a solid basis upon which to make valid conclusions regarding life cycle comparisons between an all-GHX design and Hybrid systems.  The heat pump entering water loop temperature (EWT) target was designated as 95⁰F, but iterations to reduce the sizing of the supplemental heat rejection device and allow higher temperatures of no more than 100⁰F were acceptable if resultant life cycle costs were favorable.  DFC physical size and acoustical concerns were peripheral considerations. Hybrid systems are designated as Case One thru Case Four. Additional simulations with further reduced borefields are briefly explored which indicated possible life cycle costs savings of over 40% with total life cycle costs of around $500,000, and presented as Case 1A and 4A.  Pumping configurations included the use of dual individual circulator pumps, single circulator pumps w/central variable speed pump, and a single central variable speed pump.  The basecase which utilized dual circulator pumps used the most pump energy, while a central variable speed pump achieved 60% pump energy savings.

Ground Source Heat Pumps (GSHP) provide with the opportunity to be coupled with refrigeration units. In principle, the heat rejected by refrigerators can be harnessed to raise the efficiency of the heat pumps.
This paper analyses the operational and economic performance of this innovative system deployed in Sainsbury’s supermarkets. First, the efficiency of the GSHP is evaluated, throughout the stores and over the period under consideration. Then, an economic analysis comparing the efficiency of investing in GSHP rather than in gas boiler systems is conducted. Recommendations on cost reductions are finally developed. Results show the Coefficient of Performance (COP) of GSHP systems to be highly dependent on the period of the year. During the summer, efficiency is roughly 40% less than during the winter. Overall, the efficiency of all the GSHP systems appear to be above the eligibility threshold for the Renewable Heat Incentive (RHI), with the average Seasonal COP (SCOP) of the stores being 3.0 in 2014. From an economic perspective, this average performance leads to roughly £120,000 of operational savings per year compared to gas boiler systems, with significant contribution stemming from the improvement in the refrigeration systems. Calculations show an investment payback time (PBT) of less than 8 years, a figure projected to rise slightly in the upcoming years as electricity becomes more expensive than gas.
Finally, this research project highlights cost reductions, achievable through two different approaches. First, by turning off heat pumps only when most economically convenient, up to 5.5% of the electricity costs can be saved among the stores and nearly 15% in stores boasting high thermal efficiency. Second, the profitability of the system deprived of the boreholes is evaluated. Despite the ineligibility for the RHI, the small CAPEX of this configuration could lower the PBT to 6 years.

Steady-state heat transfer inside boreholes is usually assumed when sizing geothermal boreholes and a constant borehole thermal resistance is used to calculate the temperature difference from the fluid to the borehole wall. Thus, heat rejected into the fluid is assumed to be transferred immediately at the borehole wall. In reality, steady-state borehole heat transfer is rarely present. Rejected heat will heat the fluid and the grout before reaching the borehole wall and be transferred to the ground. These transient effects, caused by the fluid and grout thermal capacities, are beneficial as they reduce the peak ground loads and, consequently, the required borehole length. This paper proposes improvements to the ASHRAE vertical borehole sizing equation to account for borehole thermal capacity. In the first part of this study, annual TRNSYS energy simulations are performed on a residential ground-source heat pump system. Borehole models that account for thermal capacity are used to quantify the borehole transient effects for a range of operating conditions. In the second part of the paper, modifications to the current ASHRAE sizing equation are proposed to consider borehole thermal capacity. Results show that neglecting borehole transient effects leads to oversized boreholes and overestimated heat pump energy consumption. By considering the fluid and grout thermal capacity, it appears that borehole length can be reduced by about 10% and heat pump energy consumption by 5%. The largest reductions occur when heat pumps operate intermittently.

Geothermal heat pumps are a greener alternative to the traditional heating and cooling systems for buildings. Instead of using as much fuel or electricity to heat and cool a building as a conventional system does the geothermal heat pump saves energy by using the ground or nearby water source as a heat sink to displace the thermal energy. They tend to be rather large and have an expensive initial cost but in the long run they save money and fossil fuels. Most heat pump’s pipe configurations are buried under the ground where they will not be easily accessed at a later date. Because of this, the design of the configuration must be right the first time. This can be very difficult because the thermal properties of the ground varies from location to location. The ground in one place might be mostly sand and a mile away the ground might be mostly clay, so it becomes very difficult to design configurations and they can be easily oversized or undersized for the building depending on the thermal properties of the ground. This paper will explain a test process that can be done to test for the thermal properties of the ground before designing the configuration for a geothermal heat pump. This experiment is performed at the site of which the heat pump would be installed to gain the thermal properties of that particular location. A low cost and simple to use system consisting of a pump, tank, thermocouples, flow meter and data collector is used. By using this information a proper pipe configuration can be designed to best fit the needs of the building and configured to fit the available land on the property.

The session investigates the system-wide effects of heating system cost in high bay spaces. Heating system type, space temperature set points, and infiltration rates were considered while weighing the first cost of system components against the energy cost of the system operation. The three system types analyzed were forced-air unit heaters, radiant heating slab, and overhead infrared heating. The impact of combustion efficiency, heating effectiveness, parasitic losses, and occupant comfort were analyzed. Parametric energy simulations compare system selection, energy cost, and initial costs for various climate zones.

The majority of the buildings that will exist by the year 2050 in the developed countries have already been built. Therefore, in order to achieve significant reduction in the energy use in the buildings sector, vast changes have to be implemented in the existing buildings. Installing radiant panels with Phase Change Material (PCM) is a solution that could contribute in achieving this goal. This presentation summarizes the fundamentals of PCMs, the advantages and drawbacks of implementing them in radiant systems and an experimental setup in which radiant ceiling panels with embedded PCM are installed and tested in a climate chamber.

Radiant heating and cooling systems present several advantages over other heating and cooling systems in terms of thermal indoor environment and energy. One of the benefits of radiant systems is that they enable integration of renewable energy resources (ground, night-time radiative cooling, etc.) into the heating and cooling systems in buildings. These advantages of radiant systems make them particularly attractive to be used in plus-energy houses. This presentation summarizes the design, simulation and year-round measured performance of two plus-energy houses equipped with radiant systems. Different performance aspects of the houses will be covered and improvement suggestions will be provided.

The management of electricity grids requires the supply and demand of electricity to be in balance at any point in time. Electricity suppliers tend to minimize their procurement costs by offering consumers time-of-use variable electricity tariffs as an incentive to shift their demand from peak to off-peak hours. Typical new residential buildings are considered, equipped with air-to-water heat pumps that supply either radiators or floor heating system. The energy market is represented through time-varying electricity price profile. Different heating control strategies are compared in terms of thermal comfort, energy use, cost and flexibility, ranging from rule-based to predictive optimized control.

Stable operation of electric power systems requires power supply and demand to be matched on multiple time scales. For short time intervals of seconds to minutes, balance is generally achieved by actively controlling grid resources based on frequency deviations – hence the term frequency regulation. This presentation explores the practicality of using HVAC chillers as a demand side resource to provide frequency regulation ancillary service. Experiments were conducted on two commercial buildings in Boston. Results are discussed in the context of PJM Interconnection’s performance requirements and market structure.

In the context of day-ahead electricity prices, the method of Economic Model Predictive Control (EMPC) has been shown to provide expenditure reduction in building HVAC systems with thermal energy storage. However, these reductions can only be achieved if the EMPC prediction horizon is sufficiently large. This work develops an alternate controller, constrained economic linear optimal control (CELOC), and shows that CELOC will yield performance similar to large-horizon EMPC but with a virtual insensitivity to horizon size. Thus, application of CELOC will require a fraction of the computational effort while yielding nearly identical economic performance.

Generally, modelers must accept either long wait times or devise clever shortcuts or simplifications. Parallel computation allows simulations to run many times faster, which often means that less involvement is required from the modeler, and graphics processing hardware is increasingly putting parallel computation in the hands of individuals. Case studies involving radiant heat exchange will be presented showing the speedup of complex simulations on highly parallel graphics processors that reduce both human and computer hours spent on simulation. The results show how parallel simulation hardware and software lead to time and cost savings in design and to more efficient buildings.

Until recently the infrastructure and knowledge required to employ massively parallel computing techniques for building energy simulation was restricted to national labs and the largest engineering firms.  Now that cloud computing resources, and the tools to use them without needing a doctorate in computer science, are available and cheap running one simulation is just as easy as running a hundred. This seminar explores how these tools are affecting and enhancing the building energy modeler's workflow and our ability to understand ourselves and communicate to our customers how buildings behave and where the best opportunities for energy efficiency are positioned.

Foundation heat transfer calculations for annual energy simulation is a complicated three-dimensional problem that can require days or weeks to solve using traditional numerical approaches. This presentation demonstrates how high performance computing enables the exploration of the parameters that impact both computation time and accuracy. By applying new calculation approaches, the computation time can be reduced to a matter of seconds while still maintaining greater than 97% accuracy.

The objectives of this presentation are to quantify the influence of several single-path duct disturbances on volumetric air flow rate measurements using traversing techniques, and to develop guidelines for field technicians to assist them in making more accurate volumetric airflow measurements in rectangular ducts during test and balance operations. Data are presented that attempt to quantify the error caused by the distance from single-path disturbances (straight ducts, elbows, 60º and 90º transitions) to a given airflow measurement (traverse) location, using both thermal anemometer and Pitot-static probes. The traversing algorithms used were the Log-Tchebycheff (LT) method and Equal Area (EA) method.

The objectives of this presentation are to quantify the influence of several multiple-path (tee) duct disturbances on volumetric air flow rate measurements using traversing techniques, and to develop guidelines for field technicians to assist them in making more accurate volumetric airflow measurements downstream of diverging tee fittings during test and balance operations. Data are presented to quantify the error caused by the distance from multiple-path disturbances (diverging tees) to a given airflow measurement (traverse) taken in the branch downstream, using both thermal anemometer and Pitot-static probes. The traversing algorithms used were the Log-Tchebycheff (LT) method and Equal Area (EA) method.

Nanolubricants are nanoparticles finely dispersed in a lubricant – are a potential cost-neutral technology that is able to increase the energy efficiency of HVAC&R systems. This talk focuses on the nanoparticles interaction with the working fluid during phase change processes. The speaker will provide a summary of the research conducted on smart nanolubricants with the support of the ASHRAE IRG. Nanolubricants provided augmented heat transfer rate in the heat exchangers with very small pressure drop penalization. The oil in circulation with the refrigerant in air conditioning system components was transformed from an unwanted contaminant to an effective energy efficiency promoter.

This talk focuses on how a new innovative technology - Biofiltration - can improve air quality in residential buildings by explaining how the clean air delivery rate is computed for a botanical filter and then will explain how a biowall was implemented in a research home.

The ASHRAE Innovative Research Grant (IRG) program was established in 2011, but the first two, and only grants issued to date, were awarded in 2012. After 2012, RAC chose to suspend the program and allow first two IRGs to run their full three course and then reevaluate the program before soliciting new grant proposals.  RAC has now completed their review of the IRG program and they will annouce through this talk what are the future plans for this program.

The purpose of Research Project RP1627 is to determine the effectiveness of 30% Advanced Energy Design Guidelines for K-12 schools and small office buildings, determine the factors common to well and poorly performing buildings, and to provide recommendations for how future AEDGs could be made more effective. Group14 collected utility data and developed weather-normalized Energy Utilization Indices (EUIs; site energy use per unit area per year) for a sample of small office and K-12 school buildings designed in accordance with the first (30%) ASHRAE AEDGs. The results were compared to the modeled ASHRAE Standard 90.1-1999 Baseline and 30% Savings EUIs from the Technical Support Documents used to develop the two AEDGs. The sample included 30 schools and 23 office buildings; most designed to AEDG requirements. A total of 14 of the 53 building sample were designed to meet the ASHRAE Standard 90.1-1999 code; these code buildings provide a means for comparison to the AEDG buildings. Most buildings meeting 30%AEDG requirements achieved energy savings exceeding 50% of the ASHRAE 90.1 Baselines. However, non-AEDG schools, constructed to code, performed nearly as well. Most school districts are committed to energy-efficient buildings. The non-AEDG small office average EUI was near the Baseline EUI, but AEDG small offices EUIs were about 50% of Baseline EUIs, indicating significantly better performance for AEDG small office buildings. On-site energy and indoor environmental quality (IEQ) audits were performed on 5 schools and 5 small office buildings with different designs and in different climate zones to verify AEDG required strategies were included in the designs and that strategies are operational and effective. In general, most, but not all, AEDG strategies were in place and operational. The research project is still underway and is scheduled for completion in January 2016. The next steps are: to determine the factors common to relatively well-performing buildings, as well as the factors common to relatively poorly-performing buildings, based on building surveys.  The next step is to provide recommendations for how future AEDGs for small office and K-12 school buildings could be made more effective in achieving better energy performance than required by ASHRAE Standard 90.1 while providing acceptable indoor environmental quality. This project supports goals of ASHRAE’s Research Strategic Plan 2010 – 2015 and will help ASHRAE maintain its leadership position in the effort to help engineers, designers, and contractors build progressively more energy-efficient buildings that deliver acceptable indoor environmental quality at a reasonable cost.

The Energy Policy Act of 2005 (EPACT 2005) required energy use in federal buildings to be metered. Since then, the US Army has installed electricity and natural gas meters on new and many existing facilities. To manage vast amounts of metered data from installations across the country, a centralized Metering Data Management System (MDMS) was put into place. MDMS continuously collects metered data, and makes it available at a central location for easy access. However, it quickly became clear that for the building energy data to be useful, it had to be put into context. CBECS and Energy Star’s portfolio manager provide context for commercial buildings in the private sector. A similar reference was needed for Army buildings that are often quite different in design and function to commercial buildings. The US Army Corps of Engineers (USACE) in collaboration with Pacific Northwest National Laboratory (PNNL) developed baseline models for five common Army building types. The focus of this initial study was on newly constructed buildings (post 2008) that were designed using standardized design documents created in response to EPACT 2005. Using metered data from these newly constructed buildings, EnergyPlus whole building energy models of the standard designs were calibrated. These models were then reverted to a Standard 90.1-2007 baseline using rules established by Appendix G of Standard 90.1-2007. The Standard 90.1-2007 baseline is important because it allows comparison of the performance of buildings to the EPACT 2005 requirement of building performance being 30% better than Standard 90.1-2007. This paper describes the methodology behind the creation of the baseline models and also describes how these models will be used to output results for MDMS. The methodology starts with processing of raw data from MDMS, choosing data for calibration, performing model calibration and reverting calibrated models to the Standard 90.1-2007 baseline. By using calibrated models to generate the baseline, the actual operating of buildings for a given location is captured in the models. The paper will describe the advantages and disadvantages of the approach, and will summarize ways in which baseline models will be used and how they will benefit MDMS and its users.

Both the current and future buildings will have very well-insulated building envelopes heated mostly by internal heat gains from occupants and household electricity. The occupant related energy uses, household electricity and domestic hot water heating, will have a large impact on the energy performance. In low-energy buildings the heating of domestic hot water is in the same order of magnitude as the energy for space heating. Knowledge of occupancy levels is critical for prediction and verification as well as optimization of various demand controlled systems. This article presents how household electricity and domestic hot water use varies due to occupancy level. Occupancy level, household electricity and domestic hot water were measured in 79 apartments during 12 days per apartment. Occupancy level was measured by electronic diaries in which those living in and visiting the apartment marked their attendance by pressing buttons when entering and leaving the apartment. Household electricity and domestic hot water were measured hourly while the diaries recorded data every second. The result shows that there are relatively weak correlations between occupancy level and the studied energy uses demonstrated by a large variation in both household electricity and hot water use at the same occupancy level. Household electricity has a stronger correlation to the occupancy level during the day compared to the occupancy level during the night, where the level during the night is assumed to describe how many people are living in the apartment. The hot water has a stronger correlation to the occupancy level during the night compared to the occupancy level during the day. One possible explanation is that larger quantities of domestic hot water could be due to showering in the morning, which would depend on how many people spent the night in the apartment while household electricity could depend more on how much of the day there are people in the apartment. The result shows that the average occupancy level during the day or the number of people living in an apartment describes only a small part of the use of household electricity and domestic hot water. Probably there is considerable individual differences in how people use electricity and hot water, which leads to a need to describe the variation by statistical distributions.

The importance of follow ups of buildings energy use during operation to optimize the performance and identify deviations and errors has attracted attention during recent years. As buildings are becoming more energy efficient with a better exterior envelope, household electricity is becoming an increasingly important part of the total heating energy and, in turn, the optimization of the thermal performance of the exterior envelope. Variations between different years can be thought to have a significant impact. Unfortunately, it seems that enough knowledge, studies and data that describe how the household electricity varies between different years are missing. A low household electricity use during one year might imply a certain hvac and building design while a high household electricity use might imply another optimal design, and this may vary over time. This might also in part explain reported gaps between calculated and measured energy use. To supply the industry with reference data on typical variations during time and between different users, there is a need for several consecutive years of measurements of household electricity in the same apartments in a large enough sample of apartments to obtain good statistics. Household electricity has been measured in 539 apartments in 25 multi-family buildings in Sweden during six years. This paper presents statistics and characteristics of the variation between years and between different apartments and buildings. The results show large differences between years, regarding both apartments and buildings, and a conclusion is that this should be considered booth during design and operation.

Historically, building simulation users have used ‘typical’ year weather data to represent climatic conditions for a location or region. With advent of increasingly powerful computers, using a single year of data is no longer necessary. Prior studies have shown that a single year of data often does not well represent the range of climate conditions over a period. We demonstrate how several sets of international typical meteorological data sets compare to the actual period of record that they represent, and demonstrate the inter-annual variability of energy use due to real weather data in comparison to TMY-type data.

Typical year weather files give a convenient snapshot of the likely weather conditions in a location. However, they provide no information of the year-to-year variability of the weather, which can have a dramatic impact on a building's heating and cooling energy use. This presentation shows a procedure using the variable-base degree day method to determine from the period of record which years would produce the highest heating or cooling loads and calculate the standard deviation in loads from the typical year. These results are building-specific, depending on how sensitive is the building to conduction, convection, or radiation heat flows.

Typical Meteorological Year (TMY) data sets provide industry standard resource information for building designers and the solar industry. Historically, TMY data sets were only available for certain locations, but current TMY data sets are available on the same grid as the new 4-km by 4-km gridded National Solar Radiation Database (NSRDB) data and are referred to as the gridded TMY. In this presentation, we analyze the temporal and spatial variability of the typical year data sets, thereby providing insight into the representativeness of a particular TMY data set for use in building performance modeling.