Search

Highlights • New method to estimate energy use and distribution losses in hot water reheat systems. • Accurate to within 14% as assessed in case study of 11,000 m2 office building. • Gas to intentional reheat energy conversion shows 79% loss, incl. 44% distribution loss. • High boiler losses for these systems in practice due to ultra-low part load operation. • Electric reheat and PV cheaper and more efficient than current practice in many cases. Abstract We developed a new method to estimate useful versus wasted hot water reheat energy using data obtained from typically installed instrumentation that applies to all pressure independent VAV terminal units with discharge air temperature sensors. We evaluated the method using a year of 1 min interval data for a 11,000 m2 building with 98 terminal reheat units, and found a 14% upper bound for the uncertainty associated with this method. We found that just 21% of gas energy is converted to useful reheat energy in this building. The distribution losses alone were 44% of the heat output from the boiler. The results raise questions regarding the tradeoffs between hot water heating systems, which have significant distribution losses, and electric heating systems, which effectively have zero distribution losses. In this building, and likely many others, an electric reheat system supplied by a small photovoltaic panel system would have a lower operating energy cost and a lower initial cost than the hot water reheat system. Further investigations using this method will be relevant to designers and standards developers in deciding between electric and hot-water reheat, particularly for modern designs using dual-maximum controls and low minimum airflow setpoints. Graphical abstract Image, graphical abstract [ABSTRACT FROM AUTHOR]

This paper describes a new supply air temperature control strategy for multi-zone variable air volume systems. We developed the strategy with the intent that it is simple enough to implement within existing building management systems. At 5-min intervals, the strategy estimates the cost of fan, heating and cooling energy at three different supply air temperatures (current, higher, lower), and chooses the one with the lowest cost as the setpoint. We then implemented this strategy in a seven floor, 13,000 m 2 office building and compared the energy costs to the industry best practice control strategy in a randomized (daily) controlled trial over a 6-month period. We showed that the new control strategy reduced total HVAC energy costs by approximately 29%, when normalized to the typical annual climate data for this location and operating only during typical office hours. These findings indicate that the current industry best practice control strategy does not find the optimal energy cost point under most conditions. This new control strategy is a valuable opportunity to reduce energy costs, at little initial expense, while avoiding more complex approaches, such as model predictive control, that the industry has been hesitant to adopt. We describe the new control strategy in language common to the industry (see sequence of operations included as supplemental material) so that readers may easily specify and implement this immediately, in new construction or controls retrofit projects. [ABSTRACT FROM AUTHOR]

Typical variable air volume (VAV) terminals spend the majority of time at their minimum airflow setpoints. These are often higher than the minimum ventilation requirements defined by code, resulting in excess energy use and a risk of over-cooling the spaces. We developed and tested a time-averaged ventilation (TAV) control strategy in an institutional building on the UC Berkeley campus to address this issue. Whenever a zone does not require cooling, TAV alternates the VAV damper between partially open and fully closed so that the average airflow matches a predefined ventilation setpoint. Compared to the existing, base case scenario using single-max VAV logic, this strategy reduced the mean zone airflow fraction from 0.44 to 0.27 during the intervention period. The corresponding reductions in average heating, cooling, and fan power were 41%, 23%, and 15% respectively. In addition to being programmed directly in a native control system, TAV may be applied via sMAP as a low-cost retrofit strategy in any building that has a BACnet network and direct digital control (DDC) to each VAV terminal. [ABSTRACT FROM AUTHOR]

Underfloor air distribution (UFAD) systems use the underfloor plenum beneath a raised floor to provide conditioned air through floor-mounted diffusers, which typically discharge cool air with both horizontal and vertical momentum components. These systems usually create a vertical temperature stratification when in cooling mode and this has an impact on energy, indoor air quality and thermal comfort. The purpose of this study was to characterize the stratification performance of a previously unstudied type of floor diffuser that discharges air horizontally, with almost no vertical velocity component, and that aims to combine the benefits of both UFAD and displacement ventilation (DV) strategies. We performed 19 full scale laboratory experiments in which we varied the number of diffusers and the internal loads over a range of values typically found in office spaces. We quantified the amount of thermal stratification by measuring the dimensionless temperature at ankle height and found a degree of stratification that is typical of DV systems – higher than is typical in UFAD systems. We developed a model based on these results that can be used to simulate these systems in whole building energy simulation tools, such as EnergyPlus, and simplified UFAD design tools. [ABSTRACT FROM AUTHOR]

We assess the impact that furniture and contents (i.e. internal mass) have on zone peak cooling loads using a perimeter zone model in EnergyPlus across 5400 parametric simulation runs. The zone parameters were HVAC system type (overhead, underfloor, and thermally activated building system (TABS)), orientation, window to wall ratio, and building envelope mass. The internal mass parameters were the amount, area, and the material type used. We also evaluated a new internal mass modeling method, which models direct solar radiation on the internal mass surface, an effect that is missing in current methods. We show how each of these parameters affect peak cooling load, highlighting previously unpublished effects. Overall, adding internal mass changed peak cooling load by a median value of −2.28% (−5.45% and −0.67% lower and upper quartiles respectively) across the studied parameter space. Though the median is quite low, this study highlights the range of effects that internal mass can have on peak cooling loads depending on the parameters used, and the discussion highlights the lack of guidance on selecting reasonable values for internal mass parameters. Based on this we recommend conducting an experimental study to answer outstanding questions regarding improved specification of internal mass parameters. [ABSTRACT FROM AUTHOR]

Whole building energy simulation (BES) models play a significant role in the design and optimisation of buildings. Simulation models may be used to compare the cost-effectiveness of energy-conservation measures (ECMs) in the design stage as well as assessing various performance optimisation measures during the operational stage. However, due to the complexity of the built environment and prevalence of large numbers of independent interacting variables, it is difficult to achieve an accurate representation of real-world building operation. Therefore, by reconciling model outputs with measured data, we can achieve more accurate and reliable results. This reconciliation of model outputs with measured data is known as calibration. This paper presents a detailed review of current approaches to model development and calibration, highlighting the importance of uncertainty in the calibration process. This is accompanied by a detailed assessment of the various analytical and mathematical/statistical tools employed by practitioners to date, as well as a discussion on both the problems and the merits of the presented approaches. [ABSTRACT FROM AUTHOR]

Studies have indicated that 20-30 % HVAC system energy savings are achievable by recommissioning air handling units (AHU) to rectify faulty operation. Studies have also demonstrated that on-going commissioning of building systems for optimum efficiency can yield savings of an average of over 20 % of total energy cost. Automated fault detection and diagnosis (AFDD) is a process concerned with automating the detection of faults and their causes in physical systems. AFDD can be used to assist the commissioning process at multiple stages. This article presents a review of the research work that has been carried out on the use of AFDD tools in improving the efficiency of AHUs. This updates and expands upon the most recent literature review in this area, published in 2005. The article offers a comparative analysis of the FDD techniques currently in use and offers an opinion as to which show most potential for widespread adoption as part of the on-going commissioning process. It then details the issues which have impeded the adoption of existing AFDD tools for AHUs to date before concluding with an appraisal of current and recommended areas for future research to overcome the barriers to the widespread adoption of AFDD tools in AHUs. [ABSTRACT FROM AUTHOR]

Abstract: Heating Ventilation and Air Conditioning (HVAC) system energy consumption on average accounts for 40% of an industrial sites total energy consumption. Studies have indicated that 20 – 30% energy savings are achievable by recommissioning Air Handling Units (AHUs) in HVAC systems to rectify faulty operation. Studies have also demonstrated that on-going commissioning of building systems for optimum efficiency can yield savings of an average of over 20% of total energy cost. Automated Fault Detection and Diagnosis (AFDD) is a process concerned with automating the detection of faults and their causes in physical systems. AFDD can be used to assist the commissioning process at multiple stages. This paper outlines the development of an AFDD tool for AHUs using expert rules. It outlines the results of the alpha testing phase of the tool on 18 AHUs across four commercial & industrial sites with over €104,000 annual energy savings detected by the AFDD tool. [Copyright &y& Elsevier]

[Display omitted] • Largest known field study of ceiling fans sequenced with air conditioning for cooling. • Mean zone indoor temperature increased 1.9 °C in period after fan installation. • 36% total measured cooling season compressor energy savings (from 2.8 to 1.8 W/m2). • Mean ceiling fan power only 8 W, at speeds <60% maximum, fans ran 76% occupied hours. • All interviewees felt fans provided adequate cooling; most preferred automated fans. Controlled air movement is an effective strategy for maintaining occupant comfort while reducing energy consumption, since comfort at moderately warmer temperatures requires less space cooling. Modern ceiling fans provide a 2–4 °C cooling effect at power consumption comparable to LED lightbulbs (2–30 W) with gentle air speeds (0.5–1 m/s). However, very limited design guidance and performance data are available for using ceiling fans and air conditioning together, especially in commercial buildings. We present results from a 29-month field study of 99 automated ceiling fans and 12 thermostats installed in ten air-conditioned buildings in a hot/dry climate in California. Staging ceiling fans to automatically cool before, and then operate together with air conditioning enabled raising air conditioning cooling temperature setpoints in most zones, with overall positive occupant interview and survey responses. Overall measured cooling season (April– October) compressor energy savings were 36%, normalized by floor area served (41% during summer peak billing hours). Weather-normalized changes in zone energy use varied from 24% increase to 73% decrease across 13 compressors, reflecting variation in occupant schedules and other uncontrolled factors in occupied buildings. Median weather-normalized energy savings per compressor were 21%. Staging ceiling fans and air conditioning provided comfort across a wider temperature range, using less energy, than air conditioning alone. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, an EnergyPlus model was used to simulate the operation of a novel integrated HVAC system. This system combines an underfloor air distribution system with a cooled radiant ceiling slab. A cooling tower supplies water to pre-cool the structural slabs during the night and early morning period. The paper compares the performance of this system to both a typical overhead system and a typical UFAD system in the cooling season for the Sacramento, California climate. When compared to the overhead system, the integrated UFAD/Radiant system shows a 22–23% reduction in total energy consumption during the peak cooling months (June to August) and a 31% reduction in peak hourly electricity demand. When compared to the UFAD system, these reductions are 21–22% and 24% respectively. An investigation of the simulation results showed that the integrated UFAD/Radiant system also improves occupant thermal comfort and reduces thermal decay issues in the underfloor plenum. [Copyright &y& Elsevier]

Abstract: This paper demonstrates a systematic, evidence-based methodology for calibrating whole building energy models. The methodology uses version control software to store a complete history of the calibration process, including the evidence on which decisions were made. This paper details the calibration of a whole building energy model to hourly energy consumption data using the methodology. The case study building was a 30,000m2, four-floor office building located on Intel''s campus in Ireland. The final calibrated model represents the building to a high level of detail using a large number of zones and uses measured lighting and plug load data in the simulation at hourly intervals. The results show excellent correlation with the measured HVAC consumption data for the analysed year (2007), demonstrating the effectiveness of the methodology. Mean Bias Error (MBE) and Cumulative Variation of Root Mean Squared Error (CVRMSE(hourly)) for HVAC electrical consumption were −4.16% and 7.8%, respectively for the final model. This model was then used to investigate Energy Conservation Measures (ECMs) for feasibility. The paper concludes with a discussion of discrepancies remaining in the model, the issues encountered related to the criteria used for determining when a model is calibrated, and recommendations for future calibration case studies. [Copyright &y& Elsevier]

Abstract: This paper reviews existing case studies and methods for calibrating whole building energy models to measured data. This research describes a systematic, evidence-based methodology for the calibration of these models. Under this methodology, parameter values in the final calibrated model reference the source of information used to make changes to the initial model. Thus, the final model is based solely on evidence. Version control software stores a complete record of the calibration process, and the evidence on which the final model is based. Future users can review the changes made throughout the calibration process along with the supporting evidence. In addition to the evidence-based methodology, this paper also describes a new zoning process that represents the real building more closely than the typical core and four perimeter zone approach. Though the methodology is intended to apply to detailed calibration studies with high resolution measured data, the primary aspects of the methodology (evidence-based approach, version control, and zone-typing) are independent of the available measured data. [Copyright &y& Elsevier]

• Assessed the difference between t r ¯ and t a from over 200000 pairs of measurements. • Median absolute difference between t r ¯ and t a is 0.4 °C. • t a is an appropriate estimate of t r ¯ in mechanically-conditioned offices. • Spatial and temporal variations in air temperature can be greater than t a − t r ¯. • Operative temperature sensors would not likely benefit thermal comfort. We assessed the difference between mean radiant temperature ( t r ¯) and air temperature (t a) in conditioned office buildings to provide guidance on whether practitioners should separately measure t r ¯ or operative temperature to control heating and cooling systems. We used measurements from 48 office buildings in the ASHRAE Global Thermal Comfort Database, five field studies in radiant and all-air buildings, and five test conditions from a laboratory experiment that compared radiant and all-air cooling. The ASHRAE Global Thermal Comfort Database is the largest of these three datasets and most representative of typical thermal conditions in an office; in this dataset the median absolute difference between t r ¯ and t a was 0.4 ∘C (with 5th, 25th, 75th, and 95th percentiles = 0.2, 0.2, 0.6, and 1.6 °C). More specifically, the median difference shows that t r ¯ was 0.4 ∘ C warmer than t a (with 5th, 25th, 75th, and 95th percentiles = −0.4 °C, 0.2 °C, 0.6 °C, and 1.6 °C). The laboratory experiments revealed that in a radiant cooled space t r ¯ was significantly (p < 0.05) cooler than t a (average difference −0.1 ∘C), while in the all-air cooled space t r ¯ was significantly (p < 0.05) warmer than t a (average difference +0.3 ∘C). These observations indicate that t r ¯ and t a are typically closer in radiant cooled spaces than in all-air cooled spaces. Although the differences are significant, the effect sizes are negligible to small based on Cohen's d and Spearman's rho. Therefore, we conclude that measurement of t a is sufficient to estimate t r ¯ under typical office conditions, and that separate measurement of t r ¯ or operative temperature is not likely to have practical benefits to thermal comfort in most cases – this is especially true for buildings with radiant systems. Furthermore, spatial and temporal variations in t a can be greater than or equal to the difference between t r ¯ and t a at any one location in a thermal zone, thus we expect that such variations typically have a greater impact on occupant thermal comfort than the differences between t r ¯ and t a. Image, graphical abstract [ABSTRACT FROM AUTHOR]

Thermally activated building systems have the potential to achieve significant energy savings, yet, the exposed concrete may also create acoustical challenges due to the high reflectivity of the hard surface. Free-hanging acoustical clouds reduce the acoustical issues, but also the cooling capacity of a radiant chilled ceiling system. Fan-induced air movement can be used to compensate for the cooling capacity reduction. We experimentally assess the combined effect of acoustical clouds and fans on the cooling capacity for an office room. We installed a ceiling fan between the clouds (blowing in the upward or downward direction) and small fans above the clouds (blowing horizontally) at the ceiling level to increase the convective heat transfer along the cooled ceiling. We tested the different fan configurations against a reference case with no elevated air movement. The tests conducted without fans showed that cooling capacity decreased, but only by 11%, when acoustical cloud coverage was increased to 47%, representing acceptable sound absorption. The ceiling fan increased cooling capacity by up to 22% when blowing upward and up to 12% when blowing downward compared to the reference case over the different cloud coverage ratios. For the variants with small fans, cooling capacity increases with coverage, up to a maximum increase of 26%. This experiment proves that combining fans with acoustical absorbents close to the radiant surface increases cooling capacity while simultaneously providing improved acoustical quality, and quantifies the impact. [ABSTRACT FROM AUTHOR]

Radiant slab systems have the potential to achieve significant energy savings, yet, when applied in the ceiling (e.g., thermally activated building system) the exposed concrete may also create acoustical challenges due to the high reflectivity of the hard surface. Balancing all of the building indoor environmental quality factors is important in the design of an effective workspace for the occupants, and so we need to consider the interactions between thermal and acoustic comfort. We assessed the cooling capacity in a hydronic test chamber and the sound absorption in a reverberation chamber to study the effects, for an office room, of different coverage areas of free-hanging acoustical clouds below a radiant chilled ceiling. The cooling experiments showed that for 47% cloud coverage of the ceiling area, we measured only an 11% reduction in cooling capacity caused by the blockage of radiant exchange between the ceiling and the room. The acoustical results showed that if the cloud covered 30% of the ceiling in a private office or 50% in an open plan office, acceptable sound absorption at the ceiling was achieved. We showed that good acoustic quality can be achieved with only a minor reduction of cooling capacity. [ABSTRACT FROM AUTHOR]

The contribution of buildings towards total worldwide energy consumption in developed countries is between 20 and 40 %. This is expected to rise by an average rate of 1.5 % per annum over the next 20 years. Heating ventilation and air conditioning (HVAC) and more specifically air handling units (AHUs) energy consumption accounts on average for 40 % of an industrial site's energy consumption. Building systems rarely perform as well in practice as anticipated during design due to improper equipment selection or installation, lack of commissioning, or improper maintenance to cite but a few reasons. Studies have indicated that 20-30 % energy savings are achievable by recommissioning HVAC systems, and more specifically AHU operations, to rectify faulty operation. Automated Fault Detection and Diagnosis (AFDD) is a process concerned with potentially automating the commissioning process through the detection of faults. This paper seeks to illustrate the effectiveness of a new rule-based expert system developed for AHUs known as 'AHU InFO' when compared to the Air Handling Unit Performance Assessment Rules (APAR). AHU InFO has proven to be more effective than APAR in tests against both derived and field test data on a variety of AHU types. In tests against 52 derived faults, AHU InFO identified all 52 issues whereas the APAR identified just ten primarily due to a lack of instrumentation negating the use of many of its constituent rules. In comparisons against field test data, the comparison sought to highlight the developments implemented in AHU InFO with these tests showing that the AHU InFO outperformed APAR in each category tested. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the key factors methodology that supports energy managers in determining the optimal building operation strategy in relation to both energy consumption and thermal comfort. The methodology is supported by the utilization of calibrated building energy simulation models that match measured data gathered by an extensive measurement framework. Building management systems, do not allow energy managers to test the impact of changes in control settings of energy systems on energy consumption and thermal comfort. The proposed methodology and prototype tool chain enables energy managers to virtually develop and test the impact of proposed changes to the control settings in the building energy systems prior to their implementation in the physical building. The paper outlines the proposed methodology defining the underpinning concepts and illustrating the performance metrics required to capture the effect of different building operation strategies. A case study is discussed to demonstrate the application of the methodology. [Copyright &y& Elsevier]

Ceiling-fan-integrated air-conditioning (CFIAC) is a concept in which terminal supply ducts and diffusers are replaced by vents/nozzles that jet supply air into the vicinity of ceiling fans to be mixed and distributed within the room. CFIAC distributes the supply air within the room and convectively cools the occupants. This could allow raised thermostat setpoints and reduced energy for the air-conditioning system's supply fan and compressor. Previous work on CFIAC shows that the air temperature in the occupied zone is spatially uniform, but the air speed is not. This paper evaluates the thermal comfort performance of a CFIAC system under various ambient temperatures and fan operation modes. Human subject experiments and thermal manikin tests were conducted to characterize how subjects evaluate the thermal comfort performance across the room's floorplate, and how CFIAC affects human body heat transfer. Despite the spatial variation in air speed across the floorplate, CFIAC created uniform thermal comfort perceptions. Comfort at 28°C was similar to that of 26°C for the overhead supply neutral reference condition. Human subjects preferred having the increased air movement over that of the reference condition. The paper evaluates thermal comfort indexes appropriate for evaluating and designing CFIAC. [ABSTRACT FROM AUTHOR]

• A new approach for building environmental and energy performance assessment is proposed. • The approach integrates scenario modelling and Fuzzy Analytic Network Process. • Scenario modelling maps performance indicators to multiple granular building objects. • Fuzzy Analytic Network Process enables objective analysis on relative weights of indicators. • A case study is conducted to demonstrate the engineering value of the approach. A well-recognised gap exists between measured and predicted building energy performance. Some practical assessment approaches offer the potential to reduce this gap using multiple indicators that evaluate building performance. Such approaches rely on subjective analysis of indicators' relative weights but are typically limited to a fixed assessment structure. Scenario modelling is one method that enables flexible and multi-granular environmental and energy performance assessment by coupling building function with other pivotal aspects of building operation. However, this method weighs all performance criteria equally. The objective of this paper is to empower building managers with enhanced environmental and energy performance assessment by integrating scenario modelling with a Fuzzy Analytic Network Process. Scenario modelling decomposes environmental and energy performance assessment into a set of flexible mappings between performance indicators and multi-granular building objects while Fuzzy Analytic Network Process enables calculation of relative weights by encapsulating ambiguity in domain expertise and complex interactions among often conflicting criteria. A case study demonstrated the engineering value of this approach. The sports centre obtained an operational score of 56.9 out of 100, or level 4 of 6 (i.e. very good) in terms of operational performance classification using calculated relative weights and intermediate results for eight carefully-identified indicators. When compared to an equivalent assessment using equally weighted criteria, the proposed approach enables more informative and targeted evaluations. With these results, building managers can quickly identify inefficient areas of building operation and improve energy consumption while maintaining building function. The approach is applicable for a wide range of buildings. [ABSTRACT FROM AUTHOR]

• Radiant cooling must remove more thermal energy from a space than all-air cooling. • Buildings with all-air cooling reject more heat by passive means than buildings with radiant cooling. • The peak space heat extraction rate for radiant cooling is larger than for all-air cooling. • These differences are larger where heat gains are highly radiant. • These differences are larger where there is large opportunity for passive cooling. For radiant cooling to maintain equivalent comfort conditions as all-air cooling it must remove more heat from a space, the peak space heat extraction rate must be larger, and the peak must occur earlier. In this article, we assess how the magnitudes of these differences are influenced by heat gain characteristics and by the use of natural ventilation night precooling. We present measurements from a series of multi-day side-by-side comparisons of radiant cooling and all-air cooling in a pair of experimental testbed buildings, with equal heat gains, and maintained at equivalent comfort conditions. In a five-day experiment with mixed internal heat gains, solar gains, and natural ventilation night precooling, radiant cooling had to remove 35% more heat than the all-air system in equivalent circumstances; and the peak heat extraction rate was 20% larger (median difference on multiple days). In a similar experiment with highly convective internal gains the differences were smaller (26% more thermal energy, 12% larger peak), while in an experiment with highly radiant gains the differences were larger (40% more thermal energy, and 21% larger peak). The differences were much smaller in an experiment without natural ventilation night precooling (7% more thermal energy, 5% larger peak). These findings have consequences for the choice, design, and control of mechanical cooling systems, especially in buildings that also use passive cooling strategies such as natural ventilation night precooling. Image, graphical abstract [ABSTRACT FROM AUTHOR]

Highlights • Interior surface temperatures are lower in buildings with radiant cooling. • Buildings with radiant cooling store less heat in non-active masses. • Buildings with all-air cooling systems reject more heat by passive means. • Radiant cooling must remove more heat to maintain equivalent comfort conditions. • The peak sensible space heat extraction rate for radiant surfaces must be larger than for all-air system. Abstract Radiant cooling systems extract heat from buildings differently than all-air cooling systems. These differences impact the time and rate at which heat is removed from a space, as well as the total amount of thermal energy that a mechanical system must process each day. In this article we present measurements from a series of multi-day side-by-side comparisons of radiant cooling and all-air cooling in a pair of experimental testbed buildings, with equal heat gains, and maintained at equivalent comfort conditions (operative temperature). The results show that radiant cooling must remove more heat than all-air cooling – 2% more in an experiment with constant internal heat gains, and 7% more with periodic scheduled internal heat gains. Moreover, the peak sensible space heat extraction rate for radiant cooling (heat transfer at the cooled surface, not the cooling plant) must be larger than the peak sensible space heat extraction rate for all-air systems, and it must occur earlier. The daily peak sensible space heat extraction rate for the radiant system was 1–10% larger than for the all air system, and it occurred 1–2 hours earlier. These findings have consequences for the design of radiant systems. In particular, this study confirms that cooling load estimates for all-air systems will not represent the space heat extraction rates required for radiant systems. Graphical abstract Image, graphical abstract [ABSTRACT FROM AUTHOR]

Direct solar radiation on a cooled radiant floor increases its cooling capacity. There is limited measured evidence of this phenomenon reported in the literature. We assessed the effect of solar radiation, increased air movement, and carpet on the cooling capacity of the radiant floor in a laboratory exposed to the outside environment. We performed experiments for different chilled water supply temperature. The cooling capacity of the chilled radiant floor was measured to increase from 32 up to 110 W/m 2 under direct solar radiation. The surface temperature region exposed to solar radiation reached a peak temperature of 26   °C while the unexposed areas were between 20 and 21   °C. Increasing the chilled water supply temperature from 12 to 18   °C caused a decrease in cooling capacity from ∼110 to ∼95 W/m 2 . Higher air speeds along the floor created by ceiling fans increased the radiant slab cooling capacity by ∼12% (from 32 to 36 W/m 2 ) when the operative temperature was 24   °C and, up to ∼19% (40 W/m 2 ) when it is increased to 26   °C. The presence of thin carpet tiles reduced the radiant floor cooling capacity for ∼5% compared to an exposed floor slab. [ABSTRACT FROM AUTHOR]

This paper presents a laboratory study of Offline-trained Reinforcement Learning (RL) control of a Heating Ventilation and Air-Conditioning (HVAC) system. We conducted the experiments on a radiant floor heating system consisting of two temperature zones located in Denmark. The buildings are subjected to real-world weather. A previous paper describes the algorithm we tested, which we summarize in this paper. First, we present a benchmarking test which we conducted during spring 2021 and winter 2021/2022. This data is used in the Offline RL framework to train and deploy the RL policy, which we then tested during winter 2021/2022 and spring 2022. An analysis of the data shows that the RL policy showed predictive control-like behavior, and reduced the oscillations of the system by a minimum of 40%. Additionally, we show that the RL policy is minimum 14% more cost-effective than the traditional control policy used in the benchmarking test. • Real-world data from laboratory test with underfloor heating. • Offline Multi-Agent Reinforcement Learning can eliminate poor behavior during training while converging. • Long Short Term Memory layers are an effective method for obtaining training models for Reinforcement Learning. • Heating costs are reduced by approximately 15% and comfort is maintained. [ABSTRACT FROM AUTHOR]

The ability to inexpensively monitor indoor air speed and direction on a continuous basis would transform the control of environmental quality and energy use in buildings. Air motion transports energy, ventilation air, and pollutants around building interiors and their occupants, and measured feedback about it could be used in numerous ways to improve building operation. However indoor air movement is rarely monitored because of the expense and fragility of sensors. This paper describes a unique anemometer developed by the authors, that measures 3-dimensional air velocity for indoor environmental applications, leveraging new microelectromechanical systems (MEMS) technology for ultrasonic range-finding. The anemometer uses a tetrahedral arrangement of four transceivers, the smallest number able to capture a 3-dimensional flow, that provides greater measurement redundancy than in existing anemometry. We describe the theory, hardware, and software of the anemometer, including algorithms that detect and eliminate shielding errors caused by the wakes from anemometer support struts. The anemometer has a resolution and starting threshold of 0.01 m/s, an absolute air speed error of 0.05 m/s at a given orientation with minimal filtering, 3.1° angle- and 0.11 m/s velocity errors over 360° azimuthal rotation, and 3.5° angle- and 0.07 m/s velocity errors over 135° vertical declination. It includes radio connection to internet and is able to operate standalone for multiple years on a standard battery. The anemometer also measures temperature and has a compass and tilt sensor so that flow direction is globally referenced regardless of anemometer orientation. The retail cost of parts is $100 USD, and all parts snap together for ease of assembly. Image, graphical abstract [ABSTRACT FROM AUTHOR]