Haorong Li

A Methodology for Diagnosing Multiple Simultaneous Faults in Vapor-Compression Air Conditioners

Existing methods addressing automated fault detection and diagnosis (FDD) for vapor-compression air-conditioning equipment have good performance for faults that occur individually but have difficulty handling multiple simultaneous faults. In addition, these methods either require high-cost measurements or measurements over a wide range of conditions for training reference models, the development of which can be time consuming and cost prohibitive. This paper formulates model-based FDD in a generic way and demonstrates that decoupling is the key to handling multiple simultaneous faults. To eliminate a cost-prohibitive overall system model, an alternative physical decoupling methodology to mathematical decoupling is developed. During the mathematical development, a previously developed FDD method termed the statistical rule-based method is reexamined and cast within the general mathematical framework. The paper also includes an evaluation of the FDD method in terms of both sensitivity and robustness.

A review of virtual sensing technology and application in building systems

A virtual sensor uses low-cost measurements and mathematical models to estimate a difficult to measure or expensive quantity. Virtual sensors have been sucessfully developed and applied in other fields within the past two decades. This article reviews developments of virtual sensors in other fields and early applications for buildings. It is believed that widespread application of virtual sensors for buildings would enable a level of building optimization and improvement not previously considered to be economical. It is hoped that this article can provide a resource for these future developments and applications.

Adaptive photovoltaic system

This paper discussed a new photovoltaic (PV) system topology that uses PV energy in efficient way in order to improve system power output during different operating conditions. The proposed topology provides flexible connection between PV modules to achieve different configurations of PV array. A new switching matrix has been developed to achieve the required configurations. Preliminary simulations provide promising results for an adaptive PV array. Comparison between traditional PV system configurations and adaptive configuration is considered. A significant improvement in power curves is achieved by the proposed topology.

Virtual Refrigerant Pressure Sensors for Use in Monitoring and Fault Diagnosis of Vapor-Compression Equipment

Refrigerant pressures are critical measurements for monitoring, control, diagnostics, and optimization of vapor compression cycle equipment. Direct refrigerant pressure measuring practices are expensive and more than often problematic. This paper describes a method, termed virtual pressure sensing, wherein refrigerant pressure values are indirectly derived from low-cost temperature sensors that can be surface-mounted. In this manner, physical pressure sensors are eliminated and pressure sensing can be achieved at a much lower cost and in a non-invasive way. Five virtual pressure sensors are developed to obtain the five most important pressures in vapor compression cycle equipment: compressor discharge line pressure, condensing pressure, liquid line pressure, evaporating pressure, and suction line pressure. The performance of the proposed virtual pressure sensors, in terms of accuracy in estimating pressures and inferring liquid line subcooling, suction superheat, compressor power consumption, and refrigerant flow rate, is evaluated extensively using laboratory data collected from four systems. These systems include air conditioners and heat pumps, split and packaged systems, refrigerants R-22 and R-410a, fixed-orifices and thermal expansion valves, and reciprocating compressors and scroll compressors. Ultimately, the virtual sensors are used as a part of a decoupling-based fault detection and diagnosis (FDD) technique to diagnose multiple simultaneous faults. The impact of the virtual pressure sensors on the FDD performance is evaluated extensively using the laboratory data collected from the four various systems.

Decoupling features for fault detection and diagnosis on centrifugal chillers (1486-RP)

With growing realization of the benefits brought by fault detection and diagnosis(FDD), a large body of research pertinent to FDD methods for centrifugal chillers has been conducted during the past two decades. However, at the present time, relatively few FDD methods can handle multiple simultaneous faults on centrifugal chillers. This article describes the development of some decoupling features, which are the basis of a decoupling-based FDD methodology that can handle multiple simultaneous faults on centrifugal chillers. The performance of the decoupling features were validated and evaluated by the laboratory data from a 90-ton centrifugal chiller tested in a laboratory environment.

Virtual calibration of a supply air temperature sensor in rooftop air conditioning units

Supply air temperature (SAT) measurement is an important element in sequencing control and automated fault detection and diagnosis (AFDD) in HVAC systems to ensure the comfort of building occupants, decrease energy consumption, and lower maintenance cost. But in rooftop air conditioning units (RTUs) with gas-fired heating, the accuracy and reliability of manufacturer-installed supply air temperature (MSAT) sensors are notoriously difficult to attain. Experimental evaluations in this study, covering both the cooling and heating modes and using both direct measurements of a MSAT sensor and a multi-sensor measuring grid, demonstrate that direct measurements cannot obtain the true value of SAT in RTUs in the heating mode. Erratic measurement errors exist due to nonuniform temperature distribution and intensive thermal radiation in a compact chamber. An innovative indirect virtual calibration method for an MSAT sensor is proposed in this article to solve this issue. It demonstrates that a virtual calibrated MSAT sensor can provide accurate results when combined with a linear correlation for offset error that depends on heating stage and outside air damper signals. The linear correlation could be determined using the calculated temperature difference between the predicted theoretical true value of SAT and the direct MSAT measurement. This virtual calibration method is generic for all RTUs with similar construction of gas furnaces and can be implemented for long-term use. Further experimental evaluation and uncertainty analysis prove that the virtual calibration method can accurately predict the true value of SAT in RTUs within ±1.2°F (0.7°C) uncertainty. This economical technology will not only improve energy management of packaged units in sequencing control but also better facilitate real-time automated control and fault detection and diagnosis.

Development, evaluation, and validation of a robust virtual sensing method for determining water flow rate in chillers

Monitoring the water flow rate through chillers is necessary and important for safety and optimal operation of a chiller system. Conventional measuring methods are reluctantly accepted by end users due to high implementation cost and continuous maintenance requirements. In this study, a cost-effective virtual sensing method was developed to determine the water flow rate in chillers using generally available chiller onboard measurements. The method was implemented, evaluated, and demonstrated in both a laboratory test environment and a field test environment. The test results show that the method is capable of accurately monitoring the water flow rate in the condenser loop and evaporator loop using low-cost and noninvasive measurements obtained while the system is operating. In terms of application, the proposed method is promising for embedment within a chiller onboard controller or a monitor system to monitor actual water flow rate in the chiller system.

Real Time Power Monitoring & integration with BIM

A Real Time Power Monitoring (RTPM) System is proposed in which end-use detailed energy consumption data is provided for each load level. The data will be integrated with a BIM (Building Information Modeling) Model to create a Realtime on-line electronic BIM Model. This paper describes the RTPM system and the process leading to its creation. The basic components of the proposed system were designed and tested using a prototype board.

A Gray-Box Based Virtual SCFM Meter in Rooftop Air-Conditioning Units

Knowledge of supply airflow rate (SCFM) measurement in packaged rooftop air-conditioning units (RTUs) is vital for improving energy management and indoor air quality and facilitating real-time automated control and fault detection and diagnosis. Despite the importance of SCFM measurement in RTUs, the conventional SCFM metering devices are very vulnerable. The credibility of SCFM measurement would be compromised dramatically after a long-term use in adverse duct work surroundings. Moreover, application of conventional SCFM meters in RTUs is very costly in regard to procurement, installation, and periodic maintenance. A cost-effective and accurate nonconventional first principles based SCFM meter in RTUs was proposed previously to virtually monitor SCFM measurement. In order to overcome the deficiencies of the first principles based virtual SCFM meter in model implementation and fault diagnostics, experiments with a wider combination and coverage are investigated in this study. It is found that a gray-box based virtual SCFM meter can be obtained with available system information (outside air damper status) and low-cost temperature measurements (direct measurement of a manufacturer-installed supply air temperature sensor (SAT mfr,meas ) and outside air temperature). Further experiment evaluations demonstrate that the gray-box based virtual SCFM meter could predict the true value of SCFM very accurately (the uncertainty is ±5.9%) with significantly enhanced applicability in model implementation and capability in fault diagnostics. Additionally, the gray-box based virtual SCFM meter also inherits good characteristics of the first principles based virtual SCFM meter, such as high cost-effectiveness, good robustness against variations in multivariable operating conditions, and applicability to similar RTUs. This innovative virtual meter could serve as a permanent monitoring tool to indicate real-time SCFM measurement and/or to automatically detect and diagnose an improper quantity of SCFM for RTUs.

An improved virtual calibration of a supply air temperature sensor in rooftop air conditioning units

Accurate supply air temperature (SAT) measurements are vital for improving the energy management of packaged rooftop air-conditioning units (RTUs) through better sequencing control and real-time automated fault detection and diagnosis (FDD). However, the accuracy and reliability of single measured manufacturer-installed supply air temperature (MSAT) are greatly compromised in the heating mode due to severe temperature stratification and high thermal radiation. There exists unacceptable erratic measurement errors and the traditional calibration method can hardly overcome the defect. An easy-to-use and very cost-effective nontraditional calibration method was proposed previously to calibrate virtually an MSAT sensor in RTUs. In order to overcome the deficiencies of the virtual calibration (VCal) method in fault tolerance and fault diagnostics, experiments with wider combination and coverage are investigated in this study. It is found that an improved virtual calibration (IVCal) method can be obtained by correlating the offset errors with available system information (the outside air damper status[OADst]) and low-cost temperature measurements (the MSAT and outside air temperature [OAT]). Further experimental evaluations demonstrate that the IVCal method could predict the true value of SAT with greater accuracy (at a relative error of only ±1.4°F [0.8°C]) with excellent fault tolerance and significantly enhanced performance of a virtual supply airflow rate meter. Additionally, the IVCal method also inherits the good characteristics of the VCal method, such as its high cost-effectiveness, ease of use, and applicability to all RTUs with similar constructed gas furnaces.