Search

Compared to other types of positive displacement compressors, linear compressors show many advantages due to low friction, low power losses and compactness. Relatively little information is available in the literature with respect to linear compressor performance comparisons between oil-lubricated and oil-free linear compressors. Currently, there are two commercial linear compressors available in the market that are used in residential refrigerator/freezers. One of them uses synthetic POE oil while the other is oil-free. Both compressors operate over a wide range of capacity modulation instead of switching on and off, which results in smaller temperature variations and smaller vibration. The oil-free compressor has a special surface coating and a single bearing lubricated by the refrigerant gas itself. This paper presents experimental comparisons between the oil-lubricated and oil-free linear compressors for a domestic refrigerator/freezer. A hot-gas bypass test stand is used for testing the two compressors. During all tests, the following compressor measurements were recorded: compressor mass flow rate, suction and discharge temperatures, suction and discharge pressures, intermediate pressure and temperature, and compressor power consumption. Using these measurements, the two compressors were characterized and compared in terms of dimensionless performance parameters.

The characterization of heating and cooling performance of HVAC&R equipment is required of all manufacturers for determining seasonal energy efficiency ratings. The current rating standards, e.g. AHRI Standard 210/240, CSA C656, determine seasonal energy efficiency (e.g., SEER) using a bin method along with data from steady-state tests at different operating conditions. These standards were originally developed for equipment with single or two-stage thermostat control, but have been incrementally updated to consider equipment with variable-speed compressors and fans. However, the equipment testing is performed using control overrides and doesn’t consider the interaction of the integrated controls with the equipment. As a result, the standard ratings don’t capture the full range of part-load operation and don’t necessarily reward manufacturers who have superior controllers. To address this issue, we have been working with the Canadian Standards Association (CSA) to develop and evaluate a load-based testing methodology for evaluating the seasonal performance of variable-capacity equipment. This new testing methodology could be applied to both variable-speed and staged equipment, enabling a more direct and fair comparison of their performance. The testing methodology involves emulating the response of a building’s sensible and latent loads to equipment controls by dynamically adjusting temperature and humidity setpoints of the psychrometric chamber reconditioning system. Convergence criteria have been developed to automate the overall testing methodology so that the equipment performance can be fully evaluated using short-term tests (e.g., 1 day). Ultimately, automated load-based testing could lead to a practical approach for capturing equipment performance models that could be used in energy simulation programs for determining more accurate and application specific performance ratings. This paper presents the overall methodology of load-based testing in addition to a discussion of some of the experimental results.

Maintenance strategies currently used for commercial building rooftop units (RTU) can be classified into two categories: reactive strategies and proactive strategies. In reactive strategies, maintenance and service is performed only when needed, e.g. when a system is unable to maintain setpoint. In proactive strategies, maintenance is scheduled at routine intervals to avoid service interruptions regardless of whether the system actually needs it. While these strategies could not be more different, it is unclear which strategy is more optimal. Moreover, whether one strategy is more optimal than the other more than likely depends on the application – contributing to much uncertainty. A third category of maintenance has been enabled by automated fault detection and diagnostics (AFDD) technologies that aims to provide building operators and service providers more detailed information about the actual state of equipment in the field. This third strategy, called condition-based maintenance, aims to optimize service and maintenance decisions throughout the life of equipment based on updated measurements of performance and service costs. In this work, these three types of maintenance strategies are compared using a commercial building simulation model utilizing a fault impact equipment model. Along with comparing different strategies under the same fault scenario, ambient conditions, and loads, optimal maintenance schedules are generated using dynamic programming. Benefits of a condition-based maintenance approach utilizing a suite of AFDD methodologies are highlighted with respect to reducing operating costs.

In the last two decades, great progress has been made in improving the efficiency of air-conditioning equipment. In addition to improved performance of new equipment, there has been an increasing interest in technologies that can maintain performance over time. This has led to research and development of Fault Detection and Diagnosis (FDD) techniques for air conditioning systems, that can support building owners in scheduling cost-effective maintenance and repairs. Among FDD techniques, thermoeconomic diagnosis is a novel method for the identification of faults occurring in air conditioning systems. A very limited number of papers have focused on this topic, and the methodology is still at a very early stage of development. Thermoeconomic diagnosis is an exergy-based method to quantify the additional energy consumption (or the EER penalty) associated with individual or combinations of faults. It has been initially tested for very simple vapor compression systems through simulation, but has never been evaluated using experimental data. This work aims to assess the performance of thermoeconomic diagnosis using experimental data obtained from a five-ton variable-speed packaged rooftop air conditioning unit (RTU). The RTU was tested in psychrometric chambers under a wide range of operating conditions and fault levels. Three faults that are commonly found in rooftop systems were investigated: (i) evaporator fouling, (ii) condenser fouling and (iii) refrigerant undercharge. The experimental results were used as inputs in an equipment model, to characterize the exergy behavior of each component in presence of faults and apply the approach of Symbolic Exergoeconomics. Experimental results show the technique had difficulty in detecting some faults and its performance is quite sensitive to operating conditions. Based on these results, improvements to the FDD methodology based on empirical models of plant components are proposed. These improvements act to isolate the effects of operating conditions from the thermoeconomic effects of different faults, improving overall performance.

To better characterize the performance of variable-speed DX (Direct Expansion) equipment in a laboratory environment, a load-based psychrometric chamber testing methodology has been developed as an alternative to existing steady-state testing approaches. The methodology allows equipment to respond dynamically to a virtual building model using its integrated controls. To mimic an actual building, a virtual building model incorporates sensible and latent loads along with simple lumped capacitance building dynamics that interact with the variable-speed equipment. The rated capacity of the test equipment is used along with a specified sizing factor and target sensible heat ratio (SHR) to specify the building sensible and latent load models. In addition, heuristic approaches are used to specify and scale sensible and latent capacitances of the virtual building model. Two companion papers present the overall methodology and results for different variable-speed heat pumps using default building parameters. This paper studies the impact of the virtual building load parameters on overall performance and dynamic behavior of the equipment for load-based testing. It is shown that equipment seasonal performance can increase significantly with increasing sizing factor. In addition, performance increases with decreasing building SHR results. In addition to simple lumped capacitance models, more detailed two-node models are investigated to evaluate more realistic dynamics and their impacts on seasonal efficiency ratings. In addition, the impact of the thermostat location on equipment dynamics and performance ratings is considered.

Passenger cars roughly represent 12% of the CO2 emissions in EU. Passenger car gasoline engines present large losses in thermal energy (roughly 60% in engine coolant and exhaust gases). This paper presents a new architecture of organic Rankine cycle using both sources of heat to decrease the emissions of the car with the cooling engine fluid as working fluid. It allows to decrease the cost, the complexity, environmental and safety issues. In this paper, the main losses of the system are taken into account: under and over-expansion of the expander, bypass flow, thermal power at the condenser limitation and exhaust gas exchanger pinch-point. Two working fluids are considered (mixture of water and ethanol and mixture of glycol and water). New improvements of the cooling engine loop of passenger cars allow to reach 20 bar and 200°C, which are used as constraints in the simulation. First, a parametric study emphasizes the main losses of the proposed architecture and how to minimize them by adjusting control variables (fraction of bypass flow and pressure). Following this, the performance is presented in function of the speed of the vehicle. The simulations show a high benefit in terms of fuel consumption for speeds higher than 75 km/h. Following this, the consumption improvements are analysed on two driving cycles and are compared with other classical architectures. The maximum consumption decrease on a WLTP cycle reaches 6% with an optimal control and a steady-state model. This theoretical investigation shows the potential of this new architecture and would require experimental investigations to validate the promising results.

It has been previously demonstrated that the most significant computational requirements for vapor compression system models are associated with evaluation of thermodynamic properties. This is particularly important for transient models because properties are evaluated at each time step and the overall model can often run less than real time. The typical approach for evaluating thermodynamic properties involves the use of complicated equations of state (EOS), such as utilized in standard software tools such as RefProp and CoolProp. Overall computation speed can be significantly enhanced using interpolation methods that are based on pre-calculated thermodynamic properties. This paper presents an improved interpolation method to quickly and accurately retrieve refrigerant properties based on the neural networks. Since the approach has an explicit functional form, it is able to avoid the computation time to find nearest points in the thermodynamic database. Comparisons between the proposed method and Refprop are provided for a wide range of pressure and enthalpy to show its accuracy. Then, performance comparisons between the proposed and conventional interpolation methods for a transient vapor compression cycle simulation are provided.

Electrochemical cells have been widely explored for their use in high efficiency energy systems. In this paper a novel heat pump cycle is proposed which utilizes chemical looping driven by electrochemical cells. Chemical looping is a method that has been applied to various applications such as combustion and air separation. It consists of the cycling of a substance between different chemical compositions in order to produce a desired effect. When the chemical composition of a fluid changes, various properties such as its saturation pressure will also change. In the proposed concept, the chemical looping of electrochemically active fluids was leveraged in order to generate a heat pumping effect. A number of electrochemically active liquid organic hydrogen carriers including alcohols such as cyclohexanol and isopropanol have been investigated for use in conjunction with electrochemical cells. These organic fluids were integrated into a thermodynamic model of the proposed cycle. When operating as an air conditioner the model indicated that an increase in cooling COP of over 20% could be achieved in comparison to a conventional vapor compression system using R410A.

The Wade power plant at Purdue University produces chilled water, steam and electricity using CCHP (Combined Cooling, Heating and Power) systems to meet the campus cooling, heating and electricity demands. Steam generated from utility boilers is not only distributed through a steam tunnel system for campus heating but also used for power generation, chilled water production and in-plant auxiliary usage. Chilled water is generated using both steam driven chillers and electric chillers and is delivered through a closed water circulation loop to campus to meet the time-varying cooling demand. The electricity generated using two steam turbine driven generators provide 30-50% of the electricity required to meet campus needs. The remainder of the electricity is purchased from the local electric utility and includes a real-time pricing component that varies with time. Plant primary energy use and costs depend upon decisions regarding generation and purchasing of electricity in response to time varying factors so as to keep the operating cost minimum and meet campus electricity, heating, and cooling demands subject to time-varying prices, loads, and environmental conditions. This paper presents an approach for optimizing the operation of the CCHP system using a multimodal genetic algorithm based on hourly load forecasts and fuel pricing with a joint characteristic for the energy components to minimize the total operating cost of the plant. The tool is used to evaluate the benefits of optimal control for the Purdue CCHP plant as a function of different (possibly future) utility rate incentives.

Virtual sensors have previously been developed and demonstrated that can provide a low cost and relatively accurate estimation of the amount of refrigerant charge contained in packaged (rooftop) air conditioners.  One particular virtual refrigerant charge sensor approach uses four surface-mounted temperature measurements to determine suction superheat, liquid-line subcooling and evaporator inlet quality that are inputs to an empirical model for charge. The empirical parameters of the model are determined using linear regression applied to laboratory data collected from the system. In previous studies, extensive psychrometric chamber testing was required at different refrigerant charge levels and ambient conditions to obtain sufficient data for the regression. This testing is expensive for equipment manufacturers and it can be difficult to find available test facilities. The current work describes the development of an automated open lab training kit for calibrating the virtual refrigerant charge level sensor in an open laboratory space. The developed automated training kit algorithm has the ability to modulate the condenser and evaporator fans to simulate the effects of different ambient conditions and automatically add different amounts of refrigerant. The charge level is automatically adjusted and monitored using solenoid valves and a digital weighing scale. This approach reduces the human involvement to a great extent and eliminates the need for psychrometric chambers. An optimal set of test conditions has been determined using optimal experimental design techniques and implemented as a Python application. An Arduino microcontroller is used to continuously send data from the sensors to a personal computer which is used to supervise the process, including determining when the system has reached steady-state. The training kit has been applied to several different rooftop units in an open lab space. A comparison of the virtual refrigerant charge sensor accuracy and time/cost for calibration determined using the automated system and using psychrometric chamber test facilities will be presented in the paper.

In this paper, the results of lab and field evaluation of several RTU FDD technologies will be presented and discussed. The focus the study was on RTUs that provide cooling of small and medium commercial buildings. This work was conducted by UTRC in close collaboration Purdue University within Consortium for Building Energy Innovation (CBEI). The goal of the lab activity was to engineer and assess low-cost, embeddable fault detection and diagnostics (FDD) for a new RTU product that exceeds DOE's High Performance RTU performance specification. Primary focus of evaluation was on overall unit performance degradation (COP and capacity) resulted from single and multiple operational faults. High confidence and low false alarm rate of COP degradation were demonstrated. The cost effectiveness of FDD will be discussed. Due to high interest and close engagement of Original Equipment Manufacturer (OEM) and National Account Customer field evaluation of FDD became possible. Several commercial sited with state-of-the-art RTUs were instrumented for continues monitoring of RTU performance and assessment of frequency of typical operational faults. Also some operational faults are intentionally injected in order evaluate FDD effectiveness. Several levels of FDD applicable in the field will be discussed.

Previous research has been presented that suggests using instantaneous system power measurements can be used to perform fault detection for rooftop units. The methodology generates a normal system power model using measurements of system power and outdoor-air temperature, which can be then used to determine if system performance has deviated due to the presence of faults. This work presents data collected from several rooftop units subjected to different faults and ambient conditions using psychrometric chamber test facilities. The results of the testing show that total system power is not very sensitive to many common faults affecting direct-expansion air-conditioning equipment. In fact, only condenser fouling faults increase instantaneous power at significant levels. Other faults, such as improper refrigerant charge level, liquid-line restriction, or compressor valve leakage, may lead to total power reduction and also, in general, have relatively weak impacts on total system power.  Virtual sensor measurements of total cooling capacity or coefficient of performance (COP) are more sensitive fault detection indicators than system power.

Previous research on automated fault detection and diagnostics (FDD) for HVAC systems has shown promising benefits like earlier detection and more accurate isolation of different faults. While most researchers, equipment manufacturers, and policymakers agree that HVAC system FDD is important and has the potential to reduce significant energy waste due to faulty system operation, widespread adoption of these tools has been slow. An automated fault detection and diagnosis system has been developed for packaged (rooftop) air conditioners based on the VOLTTRONTM monitoring and controls framework developed by the Department of Energy. The system implements a virtual-sensor-based FDD methodology capable of isolating common rooftop unit faults such as improper refrigerant charge level, heat exchanger fouling, liquid-line restrictions, and compressor valve leakage. A fault impact evaluation component has also been implemented in order to determine the relative impact that faults have on system performance. This is accomplished using virtual sensor outputs and manufacturers’ performance map reference models for performance indices such as cooling capacity and COP. This system has been implemented using low-cost electronics components and was be tested using a 5-ton RTU in a laboratory environment. In this work, a high-level overview of the automated rooftop unit (RTU) FDD system structure will be presented detailing how individual software agents interact along with a description of the computational and network requirements of the system. Alternative system architectures will also be discussed in comparison to the hybrid system presented. A review of the FDD algorithms is also presented that details the virtual sensors implementations along with the methodology to detect, diagnose, and evaluate different faults.  Finally, the performance of the FDD system will be demonstrated using laboratory test data collected for a 4-ton RTU with micro-channel condenser. The goal of this research is to produce a field ready FDD tool for RTUs that can be used to show the benefits of FDD in real systems. Ultimately, the software implementation (using Python) and hardware designs of all the systems components will be released under an open source license in an effort to reduce the engineering effort required by equipment manufacturers interested in a complete AFDD solution.

Valves are widely used in HVAC systems to regulate liquid flow rate, such as hot- and chilled-water valves utilized in cooling and heating coils. These valves are typically controlled with motorized actuators where significant backlash hysteresis might exist and the backlash magnitude mostly depends on the clearance of the manufactured gearbox. Due to the hysteresis effect, unsatisfactory tracking performance results when using a conventional PI controller. This paper proposes a self-learning backlash inverse control approach to mitigate the backlash effect with moderate modifications to a conventional PI control. In the proposed approach, a self-learning procedure is carried out at the beginning of the control implementation period to estimate the backlash magnitude for a specific valve. Then a backlash inverse block is added to an existing PI controller to compensate for the hysteresis effect residing in the valve. The validity of the proposed method was verified with both simulation and experimental tests and significant improvement was observed in the control performance.

This paper presents an application of a multi-agent control approach for supervisory control of large central cooling plants. The starting point for this work was a multi-agent control simulation framework developed by Cai (2015).  To adapt the framework to the problem at hand several tasks were accomplished: agents representing the performance of the different devices of the plant were developed and inserted in the framework and generalized heuristics were incorporated to make the approach less computationally intensive. A case study of an existing cooling plant with significant complexity was utilized to conduct an extensive evaluation of the approach in terms of optimality and computational resources. Simulations were carried out using one year of historical data to predict the performance of the plant under three different control strategies: 1) multi-agent control, 2) centralized optimization based on mathematical programming techniques and 3) a heuristic control strategy. The results showed that significant savings can be achieved through the implementation of multi-agent control. It is expected that, if each hardware component of the plant comes with an integrated agent that represents its behavior, then the proposed multi-agent framework could automatically generate the multi-agent structure and control algorithm after some relatively simple pre-configuration steps. This will reduce the site-specific engineering and will provide a more economic and easy to configure solution for central cooling systems.

This study compares the performance of a system employing passive chilled beams with a conventional air system in a typical open plan office setting. Both energy efficiency and thermal comfort are taken into account as performance indicators. A typical single duct variable air volume system, widely used in large office buildings, was considered as a conventional system; a combination of the air system and passive chilled beams was considered as a typical passive chilled beam system. Measurements from full-scale experiments conducted in open plan offices (Living Labs of the Center for High Performance Buildings at Purdue) are used to develop a data-driven passive chilled beam model that can predict the total cooling capacity as well as surface temperatures under different operating conditions. In simulating the performance of the two systems, the same operative temperature was used in order to have a fair comparison. Additionally, the effect of radiation cooling of the passive chilled beam was investigated. The overall results showed about maximum 10-21% total electrical energy savings (5-23% reduction in chiller and 55% reduction in supply fan electrical energy) and thermal comfort improvement of 0.3-0.4 on a PMVscale (12-15% on a PPD scale) associated with the passive chilled beam system, depending on the system configuration. The radiation cooling of the passive chilled beam is not significant, since the effective surface area for radiant exchange with the room is much less than the surface area for convective heat transfer and temperature differences are relatively low.Â

Model Predictive Control (MPC) based approaches have recently seen a significant increase in applications to the supervisory control of building heating, ventilation and air-conditioning (HVAC) systems, thanks to their ability to incorporate weather, occupancy, and utility price information in the optimization of heating/cooling strategy while satisfying the physical constraints of HVAC equipment. Many of the proposed MPC solution approaches are centralized ones that often have difficulty in dealing with large-scale building clusters or even a single building consisting of multiple thermal zones due to the high computational cost caused by the large number of decision variables and the overhead in information gathering and distribution. This paper investigates a decomposition technique based on a variant of the Alternating Direction Method of Multipliers (ADMM) that can significantly reduce the computational and communication costs of centralized solutions, and more importantly, facilitates a plug-and-play implementation. The proposed Distributed Model Predictive Control (DMPC) framework takes advantage of parallel computation and collaborations among multiple agents, each of which is assigned to address a smaller dimensional optimization problem. The proposed method is general in that it can accommodate some fairly general types of couplings in agent dynamics as well as in the cost function; therefore it can be used for a broad class of HVAC optimal operation problems. A case study on one of the Purdue Living Labs is carried out to demonstrate the effectiveness of the proposed method. The Living Lab considered is an open office space served by one central air handling unit (AHU) with multiple diffusers whose openings can be controlled individually. The office space is partitioned into multiple thermal zones with individual thermostat controls. In view of significant thermal couplings due to direct air exchange and noticeable load gradient between zones, a multiple thermal zone coordination problem is formulated with the objective of optimally scheduling the different thermostat setpoints for energy minimization and comfort delivery while satisfying actuation constraints. Preliminary results show that the proposed method successfully converges to the optimal solution for the problem concerned in this case study. The optimal solutions demonstrate significant opportunities of inter-zonal coordination and energy savings potentials.

Hysteresis and nonlinearity exist in most valves that are utilized in building HVAC systems, such as in cooling or heating coil valves. The hysteresis mainly comes from the manufacturing clearance in an actuator gearbox and nonlinearity in the variation of the regulated flow with respect to control signal originates from valve design. The combination of valve hysteresis and nonlinearity poses significant difficulties in obtaining a satisfactory control performance when utilizing a conventional PI controller. This paper presents experimental testing results of a set of valves in an operating HVAC system serving an office building. The test valves include VAV reheat coil and AHU cooling coil valves provided by two main-stream manufacturers. Open-loop experiments were carried out that can be used to analyze the valve performance in terms of hysteresis and nonlinearity. Valve closed-loop performance was also investigated and the results showed that valves with significant hysteresis and nonlinear characteristics could lead to unsatisfactory supply air or zone air temperature controls with severe temperature oscillation and control chattering.Â

Organic Rankine Cycles (ORC) are thermodynamic power cycles designed to generate work from low temperature sources, typically between 80 °C to 270 °C.  The low temperature heat input makes this technology attractive for applications in waste heat recovery from industrial processes, exhaust gas from diesel engines, solar systems, geothermal systems, and others.  The expander has the greatest effect on increasing the efficiency of an ORC. The operating  conditions that the expander is subjected to are directed related to its efficiency.  The performance of a 5 kW scroll expander with a displacement of 73.6 cm3 per revolution, operating at speeds from 500 to 3600 rpm, and using R245fa as the working fluid in a traditional Organic Ranking cycle is experimentally investigated In this paper. Tests were conducted varying the mass flow rate of the working fluid and varying source temperatures, while measuring the effective expander power production and the isentropic efficiency.  The experimental data was then used to develop a model of the scroll expander. Using this model, thermodynamic simulations were carried out for applications similar to exhaust gas waste heat recovery from internal combustion engines.

In heat engine design, the usual objective is to maximize thermal efficiency. However, for heat engines applied to waste heat recovery, an appropriate objective is to maximize power production by converting as much of the waste heat as possible into work. An organic Rankine cycle (ORC) is particularly well-suited to waste heat recovery because of its compactness relative to a steam Rankine cycle at typical waste heat temperatures. For a single-phase (sensible) waste heat stream with a finite capacity, maximization of thermal efficiency does not result in maximum power production. Therefore, traditional approaches aimed at increasing cycle thermal efficiency are not helpful. Instead, it is necessary to find designs that properly balance heat extraction from the source and thermal efficiency of the heat engine. In this regard two alternative ORC configurations are studied and compared using a uniform modeling strategy. These configurations are the ORC with two-phase flash expansion and the ORC with zeotropic working fluid mixture (ZRC). There are two key elements of the modeling strategy. Pinch point temperature differences are used to characterize the heat exchangers, and the air-side condenser fan power requirements are estimated. Each cycle configuration is modeled and compared to a baseline ORC for a range of potential working fluids and source fluid temperatures. Based on the model, the ORC with flash expansion shows the most consistent improvement over the baseline ORC. The highest increase in net power of 84% over the baseline is seen at the low source temperature of 80 °C with water as the working fluid. However, this cycle and working fluid present more costly challenges to implementation, particularly in the expander design. This is due to the high volume ratios needed to expand low-pressure, two-phase water. The ZRC gives some of the highest relative improvements, but only when condenser fan power consumption is high. For a 100 °C source temperature and a mixture of R134a and R245fa as the working fluid, an improvement of 92% over the baseline is seen if the required condenser fan power is 846 W/(m3/s). In addition, the ZRC has the benefit that it can utilize existing ORC expanders, giving it a potentially lower cost than the ORC with flash expansion. These results are valid in terms of their comparison of the thermodynamic potential of the different cycles and working fluids. However, a more detailed analysis incorporating the geometry and cost of each component is necessary to arrive at a final recommendation for a given application.

Stuck outdoor-air dampers can lead to significant energy waste when undiagnosed for extended periods of time. This is especially true for packaged (rooftop) air conditioners where preventative maintenance may not be frequent or only reserved for emergencies. Automated fault detection and diagnosis (AFDD) tools for outdoor-air dampers and economizers have been proposed in the past to reduce the effort and cost for this kind of maintenance and are even required by some new building standards (CA - Title 24 2013). While qualitatively, the effects of stuck damper faults are understood, much less has been written about these faults’ impacts on cooling cycle performance and actual operating costs. An investigation of incorrect outdoor-air fraction on cooling capacity, efficiency, sensible heat ratio (SHR), and run-time is presented. An evaluation of the commanded damper position based on economizer controller logic is used to capture impacts of stuck damper faults at the full range of position and under different ambient conditions. Using experimental data, these models are validated for a 4-ton rooftop air-conditioner (RTU) with integrated economizer. The combined effect of these impacts are analyzed based on air-side virtual sensors outputs and modified version of an economic performance degradation index (EPDI) first proposed by Li and Braun (2006). This performance index estimates fault impacts on operation costs as well as the added equipment costs due to the need to operate the air-conditioner longer. These economic performance impact outputs can be used in an optimal maintenance scheduling tool in future work.

An organic Rankine cycle (ORC) is a thermodynamic cycle that is particularly well-suited for waste heat recovery. It is generally employed for waste heat with temperatures in the range of 80 °C – 300 °C. When the application is strictly to convert waste heat into work with no restrictions on heat source exit temperature, thermal efficiency is not as relevant as other aspects of the cycle performance. In such an application, maximization of net power may be the objective rather than maximization of thermal efficiency. An air-cooled ORC for waste-heat conversion is studied in the present work. Two alternative cycle configurations which could increase the net power produced from a heat source with a given temperature and flow rate are proposed and analyzed. These cycle configurations are: • An ORC with two-phase flash expansion • An ORC with a zeotropic working fluid mixture (ZRC) A simplified ORC model is introduced which calculates the pinch point in the heat exchangers based on a specified minimum temperature difference. This model is used to assess the merits of each cycle configuration with respect to a baseline ORC when the finite capacity of the heat source and heat sink fluids is considered. The finite capacity of the heat sink fluid is incorporated into the model in terms of a condenser fan power requirement. Of all working fluids studied for the baseline ORC, R134a and R245fa result in the highest net power. The ORC with two-phase flash expansion offers the most improvement over the baseline cycle provided the expander can handle two-phase flow at the same isentropic efficiency as in the baseline case. Relative improvements are highest at low source temperatures. The maximum increase in net power is 84% over the baseline ORC when water is the working fluid at a source temperature of 80 °C. At low source temperatures, the improvements decrease with increasing condenser fan power requirements. The improvements of the ZRC are also higher for low heat source temperatures. The ZRC shows improvement between 20% and 40% over the baseline as long as the condenser fan power is not negligible. At the highest estimated condenser fan power, the ZRC shows up to 92% improvement at a source temperature of 100 °C, while the ORC with flash expansion is no longer beneficial. This work represents a first step toward identifying a more optimal ORC configuration for waste heat recovery. Other data, including experimental validation, operating experience, and economic analysis particular to an application will be required to support a final recommendation.

Implementation of advanced controls and diagnostic features in small commercial buildings typically requires real-time monitoring of the energy flows, such as the power consumption and cooling capacity of rooftop units. However, these measurements are expensive and therefore limit the application of these advanced features. In order to lower the measurement cost, virtual sensing technology for rooftop unit power consumption and cooling capacity are being developed. Power transmitters and thermocouples are installed on the rooftop units to train the virtual sensors. The idea is to recycle the power transmitters after training to save cost, and then the virtual sensors estimate hourly electrical consumption and cooling capacity using only low cost, non-invasive temperature measurements. In this paper, development and validation of the virtual sensors are presented. The virtual power consumption sensors are validated for 4 different rooftop units installed in the field, whereas the cooling capacity sensors are validated using 3 laboratory-tested rooftop units. The reliability of the sensors is also investigated by studying the uncertainty of the virtual sensor outputs under different operating conditions. A cost comparison between the virtual sensors and direct measurement methods is also conducted to evaluate the potential for widespread application of the virtual sensing technology.

Model Predictive Controllers (MPC) in supervisory level building control systems may employ solar irradiance data to make decisions, particularly for building zones with a high fraction of window area. Solar irradiance on tilted surfaces can be calculated from Global Horizontal Irradiance (GHI) by a common algorithm that includes an empirical correlation to decompose the beam and diffuse components, followed by the application of an anisotropic sky model. Control decisions based on the modeled solar data can be incorrect, particularly if offsite or satellite derived GHI data is employed. This work investigates the impact of uncertain solar estimation on MPC performance with respect to energy consumption and comfort by comparing MPC responses for a multi-zone building based on: measured local Plane of Array (POA) solar data and satellite GHI based modeled POA solar data. The case study results inform the value of information, revealing that MPC decisions based on satellite derived solar data cause comfort violations that depend strongly on the building system characteristics, such as HVAC and internal gains.

A methodology for evaluating the performance of fault detection and diagnostics (FDD) tools applied to unitary air conditioners has been developed by Yuill and Braun (2013). Data from faulted and unfaulted systems operating over a range of driving conditions are fed to the FDD tools, and the FDD responses are compared to the known operating conditions. The methodology originally relied upon experimental measurement data, but the amount of available data is limited, and evaluations can be far more meaningful if the operating conditions of the inputs can be controlled. Furthermore, a finite input set can be learned by an FDD algorithm, and the evaluation can be thereby gamed. To solve these problems, a large library of data from multiple systems under a wide range of conditions, with and without faults of varying magnitude, was generated with simulations from a novel gray-box modeling approach (Cheung and Braun 2013a, 2013b). The simulation outputs are used to train a neural network model, which is coupled to software that executes the evaluation method. The neural network model is much simpler than the semi-empirical approach, so it can produce evaluation inputs very quickly. This facilitates the evaluator generating semi-random conditions to provide a unique set of evaluation data that are sufficiently accurate and numerous to provide repeatable results. Some evaluation results from several FDD protocols are used to illustrate the success of this approach.

A companion paper (Bach et al. 2014) experimentally investigated the effects of vapor injected compression and hybrid evaporator flow control on capacity and COP. The goal of this paper is to provide insight into the effects of these technologies on heating seasonal performance (HSPF). HSPF was calculated using a modified version of the ANSI/AHRI 210/240 method, and parametric studies were performed to better understand the seasonal performance with a focus on comparing vapor injected and single stage system configuration. It was found that part load degradation and reduced capacity at low ambient temperature are factors that can degrade the seasonal performance. The increase in heat pump COP for the vapor injected configuration leads to only a small benefit - the main contributor to the increased HSPF of the vapor injected system configuration is its increased capacity towards low ambient temperatures. References: Bach C. K., Vetsch, B., Groll, E. A., Braun, J. E., and Horton, W. T, (2014), Experimental Investigation of Vapor Injected Compression for Cold Climate Heat Pumps and its Effects on their Performance, 15th International Refrigeration and Air Conditioning Conference at Purdue, West Lafayette.

Direct expansion (DX) cooling systems are commonly used in small to medium sized commercial buildings due to low capital investment and maintenance costs. Nowadays more DX units are equipped with variable-speed fans and capacity modulating techniques to enable higher efficiency and better comfort control. This complexity provides greater energy savings potential for control optimization compared to single stage DX units with constant air flow. This paper presents a generalized control heuristic for resetting supply air temperature setpoints for multi-stage DX units configured with variable-speed supply fans. The heuristic is shown to work well for a wide range of different DX systems with different compressor types and fan-duct combinations. Digital scroll, variable-speed and multi-cylinder reciprocating compressors are considered and modeled using curve fits to manufacturer’s data. Different evaporator and condenser characteristics are also considered using models constructed from catalog data of a representative split DX cooling system. The component models are coupled to provide an overall performance model for each DX system. Fan power is another critical portion of the total HVAC power consumption. In this study, a simple method was used to characterize the performance of a duct system based on typical relationships between pressure drop and air flow rate. Different duct characteristics were coupled to different types of supply fans to provide a set of fan power maps. Optimization was then performed for each combination to find the optimal supply air temperature that minimizes total fan and compressor power. This provided a baseline for developing and validating results of the heuristic control strategy. To assess the integrated energy savings potential of the heuristic strategy, a simulation platform was developed for a medium sized commercial building to consider different DX cooling systems and climates. Simulations were carried out for a 100-day cooling season indifferent locations of US and under different control strategies, including conventional control with a constant supply air temperature setpoint, the heuristic reset strategy and optimal supply air temperature setpoint control. It is shown that the heuristic control has performance almost as good as optimal control with significant cost savings relative to conventional control that depend on the system type and location.

Building heating requirements increase with decreasing ambient temperature, while the coefficient of performance of air-source heat pumps (ASHPs) shows the opposite trend. Additionally, heating capacity decreases with ambient temperature, which leads to the utilization of inefficient electric reheat below the design point. Increasing the capacity and COP at lower ambient temperatures is important for improving the market penetration of heat pumps in climates having significant operating time at low ambient temperature. Simulation studies previously showed that compressor vapor injection leads to an increase of COP under exactly those conditions. Furthermore, reduced capacity degradation towards smaller ambient temperatures was predicted. The work presented in this paper shows experimental results obtained from a commercially available 5-ton heat pump that was retrofitted with a two-port vapor injected scroll compressor. The injection ports within the two compression pathways are located in the fixed scroll with different distance from the suction chamber. The vapor for the two injection pressure levels was generated using two vapor separators in a cascade configuration. This configuration made it necessary to not only control the superheat but also the liquid levels in the separators and subcooling of the refrigerant leaving the condenser. Baseline performance data of the heat pump without vapor injection was obtained and compared with that for the vapor injection and other system configurations. For the baseline, the injection lines to the compression pockets were plugged within the fixed scroll to reduce dead volume and re-expansion losses. Also, the vapor-separator section was shut off and bypassed. In the second step, the plugs were removed and a staged expansion process was performed using the separator section. The generated vapor from each separator was injected into the respective compressor port causing an intercooling effect on the compression process. With identical compressor speed, a 28% improvement in capacity was achieved at the 8.33°C (47 F) design point, when compared to the baseline without vapor injection. When the baseline and vapor injected system capacity were matched by adjusting compressor speed, the COP increased by up to 6% at -8.33°C (17 F). Preliminary results of a bin-type analysis of the experimental data predicts an improvement in HSPF of 6% for Minneapolis and nearly 7% for AHRI climate region 5. The benefit is mainly caused by a reduction of the auxiliary electric heater’s runtime.

Flow maldistribution in evaporators can lead to significant degradation of capacity and efficiency of vapor compression equipment. A significant amount of work has previously been done to mend these issues. For variable air flow maldistribution, refrigerant compensation was proposed to reduce the performance degradation. For fixed air side maldistribution, refrigerant circuitry modifications were proposed to significantly reduce the effects of the maldistribution. However, no work has been found on modifying the refrigerant circuitry to make it less vulnerable to varying air side maldistribution. The purpose of this paper is to fill this gap in open literature. The performance of the new, interleaved circuitry approach and an active refrigerant flow control is compared to the standard circuitry for different cases of maldistribution. The results show that the interleaved circuitry recovers less of the performance losses than equalization of the exit superheats. However, the implementation cost in an actual system is expected to be significantly lower and the long term reliability is expected to be much better than for an active control approach.

Previous research indicates that a scroll compressor with multiple vapor injection ports has the potential to significantly improve the energy efficiency and maintain high heating capacity of vapor compression heat pumps, particularly at low outdoor temperatures. Based on a tradeoff between cost and benefit, a dual-port vapor injected compressor prototype was chosen for testing to provide the experimental results. The overall performance of the heat pump system employing the compressor prototype is analyzed. Coupling the compressor testing results into a three-stage expansion vapor injection flash tank cycle model, the operating injection pressure can be determined under a certain working condition. The results showed that both lower and higher injection pressures and the corresponding injected mass flow rates decreased with an increase of pressure ratio of discharge pressure to suction pressure. Comparing the COP of the vapor injection cycle with the one of a baseline cycle, the improvement in COP increased with an increase of the pressure ratio and goes up to 19% at a pressure ratio of 8 and a condensing temperature of 43.3°C. In addition, the proposed technology leads to higher energy efficiency and less degradation in heating capacity at low ambient temperatures. For Minneapolis, the seasonal efficiency (heating) of the vapor injection cycle was 13% higher than that of the standard vapor compression cycle.

A methodology for evaluating the performance of fault detection and diagnostics (FDD) tools for unitary air-conditioners has been developed (Yuill and Braun 2013). The methodology uses laboratory measurements of systems with and without faults to test FDD tools’ effectiveness. A gray box modeling method capable of modeling systems with faults was developed by Cheung and Braun (2013a and 2013b) to provide input data, as an alternative to using laboratory data that had been collected. The simulation method was validated by direct comparison with experimental data, but a comparison of FDD evaluation results provides a more direct and useful validation of the model for its intended purpose. Eight different systems have been modeled using Cheung and Braun’s method. Six FDD tools were evaluated using both experimental and modeled inputs under the same environmental and fault conditions. The fault conditions include non-standard charging, heat exchanger fouling, loss of compressor volumetric efficiency, liquid line restriction, and the presence of non-condensable gas in the refrigerant. The model’s performance is characterized by comparing its outputs from the evaluation – false alarm rates, misdiagnosis rates, missed detection rates, and rates of undiagnosed faults – with the results based upon experimental data. The model is found to be highly suitable for its purpose.

This paper presents an application of tower fan control for optimization of the performance of chiller plants combined with free cooling on the condenser water side. Mathematical models including all the main components of an existing cooling plant were developed and implemented in MATLAB. Simulation results include a mapping of the performance of the plant working in free cooling mode which was used to select control parameters for free cooling operation. Then a mapping of the plant operating with chillers was developed to find the correlation between load and near-optimal air flow, which is the basis of the near-optimal tower control (NOTC) strategy. Finally, simulations were carried out using three consecutive years of historical data to predict the performance of the plant under three different control strategies: 1) tower fan control aiming to keep the temperature of the water supplied to chiller condensers at a constant set point (current control strategy), 2) NOTC and 3) NOTC and free cooling combined. Comparison of the performance of the plant with the baseline (constant condenser water temperature) shows that significant savings can be achieved through the implementation of NOTC along with free cooling. It is expected that the methodology and results of this study provide a useful framework for optimization of cooling plants.

Model based predictive control (MPC) in building HVAC systems incorporate predictions of weather and occupancy to determine the optimal operating setpoints. However, application of MPC strategies to large buildings might not be real time feasible due to the large number of degrees of freedom in the underlying optimization problem. Decomposing the problem into several smaller sub-problems to be solved in parallel is one way to circumvent the high computational requirements. Such an approach, termed Distributed MPC, requires certain approximations about the underlying sub-problems to converge to a consistent solution thus leading to a trade off between computational load and optimality. In this paper, we present a simulation based evaluation for a Distributed MPC formulation for a case study based on a medium sized commercial building. Results indicate that distributed MPC can offer near optimal control at a fraction of the computational time that centralized optimization based MPC requires while maintaining occupant comfort. Comparison with a few other viable control algorithms will be performed and merits and drawbacks of each approach pointed out.