No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Model-Based Fault Detection and Diagnosis for Cooling Towers
Abstract
A model-based method for the detection and diagnosis of faults in the cooling tower circuit of a central chilled water facility is presented. Faults that occur in the cooling water temperature sensor, the cooling tower pump, and the cooling tower fan are considered. The faults are detected through deviations in the values of characteristic quantities from the corresponding values for fault-free operation. The characteristic quantities chosen to represent the performance are the conductance-area product of the tower, the approach, the effectiveness, and fan power. With faults present, the deviations are significant, which allows the faults to be detected. The pattern of the deviations is different for each fault, allowing rules to be developed that allow diagnosis of the source of the fault.
... The proposed FDD scheme was established based on five important characteristic features that allow six process faults to be identified. Ahn and Mitchell [11] proposed a model-based method for detecting and diagnosing some faults that could occur in cooling towers. In their study, the faults were detected through monitoring the deviations in the values of the three selected characteristics quantities (conductance-area product, approach temperature and temperature effectiveness) from the corresponding values obtained from fault-free operation. ...
... In the HQS method, the near optimal setting generated by the fixed approach control method is used as the search center to define a relatively narrow search range, as shown in Eq. (11). Based on this narrow search range defined, the exhaustive search method is then used to search for globally optimal settings with an increment of 0.1 K. ...
... The chiller model is validated in terms of the power consumption and coefficient of performance (COP). These five performance indices have been proved to be effective in evaluating the performance of cooling towers and chillers [4,11]. Fig. 7 summarizes the validation results of different PIs in terms of the coefficient of determination (R 2 ) by using the data collected in the second-five summer days. ...
This paper presents a robust strategy for online fault detection and optimal control of condenser cooling water systems. The optimal control strategy is formulated using a model-based approach, in which simplified models and a hybrid quick search (HQS) method are used to optimize the performance of the overall system by changing the settings of the local process controllers. A system level online fault detection scheme is embedded into the control system and used to monitor whether the system operates in a healthy condition. The faults considered are mainly the component performance degradations. When a fault is detected, the control system will be reconstructed to regain the control through using robust schemes. The performance of the proposed strategy is tested and evaluated against on a simulated virtual system representing the actual condenser cooling water system in a super high-rise building. The results show that the fault detection scheme is effective in identifying system performance degradations and the fault-tolerant control strategy with online fault detection and optimal control can enhance the overall system performance significantly when the operation of condenser cooling water systems suffers from performance degradations, as compared to that using optimal control only.
... The papers on chillers FDD are not included because they are summarized in the next section. For cooling towers, a temperature sensor fault, a condenser water pump fault and a cooling tower fan fault were simulated on TRNSYS platform, and diagnosed using the proposed performance indices based on Ahn et al. (2001). A fouling model was developed by Khan and Zubair (2004) to study the effect of fill fouling on performance characteristics of cooling towers. ...
... The operating performance of other sub-systems may be affected indirectly by the deterioration of cooling tower system performance. The temperature effectiveness, approach temperature and power consumption are chosen to model PIs based onAhn et al. (2001). ...
... Some researchers discussed the FDD application in the other HVAC systems, such as cooling towers, heat exchangers, air handling units (AHUs) and pumps. A temperature sensor fault, a condenser water pump fault and a cooling tower fan fault were simulated on TRNSYS platform, and diagnosed using the proposed PIs based on Ahn et al. [12]. A fouling model was developed by Khan and Zubair [13] to study the effect of fill fouling on performance characteristics of cooling towers. ...
... Although the parameter changing cannot affect the operating performance of the other sub-systems directly, the performance may be affected indirectly by the deterioration of cooling tower system performance. The temperature effectiveness, approach temperature (difference between the discharge water temperature and the inlet air wet-bulb temperature of cooling towers) and power consumption are chosen based on Ahn et al. [12]. These three PIs were proved to be valid in diagnosing the fan fault due to blockage and fan motor degradation in their study. ...
A strategy of fault detection and diagnosis (FDD) for HVAC sub-systems at the system level is presented in this paper. In the strategy, performance indices (PIs) are proposed to indicate the health condition of different sub-systems including cooling tower system, chiller system, secondary pump systems before heat exchangers, heat exchanger system and secondary pump system after heat exchangers. The regression models are used to estimate the PIs as benchmarks for comparison with monitored PIs. The online adaptive threshold determined by training data and monitored data is used to determine whether the PI residuals between the estimation and calculation or monitoring are in the normal working range. A dynamic simulation platform is used to simulate the faults of different sub-systems and generate data for training and validation. The proposed FDD strategy is validated using the simulation data and proven to be effective in the FDD of heating, ventilating and air-conditioning (HVAC) sub-systems. Copyright © 2009 John Wiley & Sons, Ltd.
... Following a systematic literature study on predictive maintenance 4.0 applications on cooling towers, this component was not given that much attention in research point of view though it is too important (Almobarek et al. 2022). A research study developed an industrial engineering approach via simulation model to predict three faults (Ahn et al. 2001). Two more studies used simulation model to assess multiple faults of cooling towers . ...
... Compared to the studies on chillers, the studies on cooling towers were limited and were either part of chillers' ones or were discussed separately. Ahn and others developed a simulation model to detect three faults of cooling towers [126]. Their model was built based on the deviation of different operational parameters such as the difference between the water temperatures that are leaving the tower and the temperatures that are entering the same. ...
Predictive maintenance plays an important role in managing commercial buildings. This article provides a systematic review of the literature on predictive maintenance applications of chilled water systems that are in line with Industry 4.0/Quality 4.0. The review is based on answering two research questions about understanding the mechanism of identifying the system’s faults during its operation and exploring the methods that were used to predict these faults. The research gaps are explained in this article and are related to three parts, which are faults description and handling, data collection and frequency, and the coverage of the proposed maintenance programs. This article suggests performing a mixed method study to try to fill in the aforementioned gaps.
... Carling (2002) compared three fault detection methods based on field data of an air-handling unit. Ahn et al. (2001) studied fault detection and diagnosis of cooling towers using a model-based method. ...
Fault diagnosis is of crucial importance to mechanical systems. To find an efficient way of conducting fault diagnosis for refrigeration systems, probabilistic neural network, and back-propagation network were employed to diagnose seven types of typical faults for a 90-ton centrifugal chiller. These include system-level faults such as refrigerant leak/undercharge, refrigerant overcharge, and excess oil and also component-level faults such as condenser fouling, reduced condenser water flow, noncondensables in the refrigerant, and reduced evaporator water flow. Eight major features of the chiller system were selected as indicative parameters for diagnosing. The establishment of the fault diagnosis models based on the probabilistic neural network and back-propagation network and the optimization processes of the networks were elaborated. Comparison in terms of diagnostic performance was performed based on the optimized networks. The results showed that the overall diagnostic performance of the probabilistic neural network was better than that of the back-propagation network. The probabilistic neural network has a correct rate that was 3.48% higher than that of back-propagation, and its diagnosis time was lower by more than 400 times. Moreover, the diagnosis of a single probabilistic neural network training was more reliable than that of the back-propagation network. It was also demonstrated that system-level faults were more difficult to be recognized by the model than component-level faults because of their widespread influence on the system operation. The probabilistic neural network model has a better performance on system-level faults with an improvement of correct rates about 4%∼8.7% from those of back-propagation model.
... Unresolved Issues One of the many unresolved issues that are currently being investigated in the HVAC industry is FDD (fault detection and diagnosis) (Ahn, Mitchell, & McIntosh 2001;Chen & Braun 2001;Dexter & Ngo 2001;House, Vaezi-Nejad, & Whitcomb 2001). Although only briefly alluded to in this paper, it is a topic that appears to be of recent growing interest to facility managers, HVAC vendors, and researchers. ...
The operation and maintenance of commercial building HVAC (heating, ventilation, and air-conditioning) systems is illustrative of an industry that can benefit from the insight- ful use of all available information sources. Modern HVAC systems using direct digital control can potentially provide useful performance data. Occupant feedback complaint data and HVAC system trend data are stored within modern main- tenance management databases. This paper will address the specific issue of integration and application of these funda- mental sources of information, using some modern and novel techniques. Examples are found in, but are not limited to the following areas: discrete-event, continuous, and hybrid control system theory, artificial intelligence, statistics, sys- tem identification, databases, etc. Methods specific to these areas can be used to synthesize a supervisory controller. The objective of this controller is to reduce energy costs, improve building occupant comfort, fault detection and diagnostic ca- pability. At the same time, the entire process needs to be stable, and the influence of occupant behavior needs to be taken full advantage of. The controller will achieve these ob- jectives by performing and/or prioritizing the appropriate ac- tions to be taken either automatically or by facility operators. Furthermore, the cost and scalability of the entire effort de- scribed can be positively influenced by recent technological advances in computing power, sensors, and databases.
... Computer-based control systems have the capability to collect and store sensor and control signals that could be analyzed to detect and diagnose faults. A considerable amount of research work has been carried out to develop FDD techniques for HVAC systems and, recently, to test these techniques in realistic laboratory settings and in real buildings (Ahn et al. 2001;Chen and Braun 2001;Dexter and Benouarets 1996;Dexter and Ngo 2001;Haves et al. 1996a;House et al. 2001;Hyvarinen 1996;Lee et al. 1996aLee et al. , 1996bLi et al. 1996;Peitsman and Bakker 1996;Salsbury 1996;Stylianou and Nikanpour 1996;Tsutsui and Kamimura 1996;Yoshida et al. 1996). ...
Detection and diagnosis of faults (FDD) in HVAC equipment have typically relied on measurements of variables available to a control system, including temperatures, flows, pressures, and actuator control signals. Electrical power at the level of a fan, pump, or chiller has been generally ignored because power meters are rarely installed at individual loads. This paper presents two techniques for using electrical power data for detecting and diagnosing a number of faults in air-handling units. The results from the two techniques are compared and the situation for which each is applicable is assessed.One technique relies on gray-box correlations of electrical power with such exogenous variables as airflow or motor speed. This technique has been implemented with short-term average electrical power measured by dedicated submeters. With somewhat reduced resolution, it has also been implemented with a high-speed, centralized power meter that provides component-specific power information via analysis of the step changes in power that occur when a given device turns on or off. This technique was developed to detect and diagnose a limited number of air handler faults and is shown to work well with data taken from a test building. A detailed evaluation of the method is presented in the companion paper, which documents the results of a series of semiblind tests.The second technique relies on physical models of the electromechanical dynamics that occur immediately after a motor is turned on. This technique has been demonstrated with submetered data for a pump and for a fan. Tests showed that several faults could be successfully detected from motor startup data alone. While the method relies solely on generally stable and accurate voltage and current sensors, thereby avoiding problems with flow and temperature sensors used in other fault detection methods, it requires electrical data taken directly at the motor, downstream of variable-speed drives, where current sensors would not be installed for control or load-monitoring purposes.
... The compressor power consumption is selected as the PI of the chillers. These two PIs have been proven to be valid in detecting the faults in cooling towers and chillers (Ahn et al. 2001;Zhou et al. 2009). It is worth noticing that, at the component level, the FDD scheme is only used to identify which component suffers from the faults. ...
This paper presents a fault-tolerant supervisory control strategy for building condenser cooling water systems. The proposed strategy mainly consists of a model-based predictive control (MPC) scheme, a fault detection and diagnosis (FDD) scheme and a fault accommodation and tolerant (FAT) scheme. The MPC scheme using systematic optimization is employed to identify optimal control settings for the local process controllers. The FDD scheme is utilized to detect and diagnose major possible faults that may happen in the routine operation of condenser cooling water systems. The faults considered mainly include critical sensor faults, physical component performance degradations and malfunctions of control logics. According to the types of faults happened, the FAT scheme is then used to handle the faults in order to regain the control as far as possible. The performance of this strategy is tested and evaluated in a simulated virtual system representing the actual condenser cooling water system in a super high-rise building. The results show that the proposed strategy is capable of maintaining acceptable control performance and can help save about 0.18%∼5.23% total energy of the chillers and cooling towers when the operation of condenser cooling water systems suffers from some faults, as compared to that using the same control strategy but without using the FAT scheme.
... Carling (2002) compared three fault detection methods based on field data of an air-handling unit. Ahn et al. (2001) studied fault detection and diagnosis of cooling towers using a model-based method. ...
Fault detection and diagnosis (FDD) is the basis for timely maintenance to keep chiller systems operate at a normal and efficient condition. This study investigates a hybrid model that combines support vector machine (SVM) with genetic algorithm (GA) and parameter tuning technique for chiller FDD applications, where GA is responsible for searching potential feature subsets and SVM behaves both as an FDD tool and an evaluation method for feature selection. Subsets of 6, 7, 8, 9, and 10 features were studied, respectively, and compared with the original 64-feature set in terms of overall performance – correct rate (CR), and individual performance – hit rate (HR) and false alarm rate (FAR). The results showed that the eight-feature subset (Feat8) singled out by the hybrid SVM model behaves the best with its testing CR about 2% higher than that of the simple SVM model (64-features) while consuming less computational time. Further validation and comparison analysis with C4.5 FDD model also convalidate the outstanding performance of Feat8. Fewer features also mean fewer sensors required for data acquisition and accordingly less sensor cost. Moreover, a drastic improvement in individual performance (HR, FAR) was observed for the two most-difficult-to-be-identified faults – refrigerant leak (RefLeak) and refrigerant overcharge (RefOver).
... For the other HVAC systems, the study on FDD can be also found in literatures, such as Ahn et al. [9] and Khan and Zubair [10] for cooling towers, Weyer et al. [11] and Upadhyayaa and Eryurekb [12] for heat exchangers, Norford et al. [13], Carling [14], and Qin and Wang [15] for air handling units (AHUs) and variable air volume (VAV) systems. ...
This paper presents a strategy for fault detection and diagnosis (FDD) of HVAC systems involving sensor faults at the system level. Two schemes are involved in the system-level FDD strategy, i.e. system FDD scheme and sensor fault detection, diagnosis and estimation (FDD&E) scheme. In the system FDD scheme, one or more performance indices (PIs) are introduced to indicate the performance status (normal or faulty) of each system. Regression models are used as the benchmarks to validate the PIs computed from the actual measurements. The reliability of the system FDD is affected by the health of sensor measurements. A method based on principal component analysis (PCA) is used to detect and diagnose the sensor bias and to correct the sensor bias prior to the use of the system FDD scheme. Two interaction analyses are conducted. One is the impact of system faults on sensor FDD&E. The other is the impact of corrected sensor faults on the system FDD. It is found that the sensor FDD&E method can work well in identifying biased sensors and recovering biases even if system faults coexist, and the system FDD method is effective in diagnosing the system-level faults using processed measurements by the sensor FDD&E.
... The premise of this method is that the relative accurate mathematical model can be obtained for sure. Some researchers paid attention to this method, such as Ann [2], Dexter [3], Stylianou [4], etc. ...
This paper presents a fault tolerant control method to control the outdoor air ventilation and AHU supply air temperature, which concerned indoor air quality and humidity, respectively to satisfy ASHRAE Standard in VAV systems. The principal component analysis method, joint angle method, and compensatory reconstruction are used to detect, isolate, and reconstruct the fault, respectively for fault tolerant control. They are tested and evaluated in a simulation environment under the condition of temperature and flow sensors with fix bias faults.
... Other equipment in chilling systems was studied as well as chillers. Ahn et al. (2001) applied a model-based method for the detection and diagnosis of faults in the cooling tower circuit of a central chilling system. Faults that occur in the cooling water temperature sensor, the cooling tower pump, and the cooling tower fan are considered. ...
xvii, 234 leaves : ill. ; 30 cm. PolyU Library Call No.: [THS] LG51 .H577P BSE 2006 Qin Variable air volume (VAV) systems and their control strategies become more and more complex in order to meet the increasing demands on indoor environment quality and energy conservation. Automatic monitoring and control of VAV systems are inevitable in modern buildings. Many supervisory VAV control strategies, such as supply air temperature reset, static pressure reset and advanced fresh air flow rate control, have been put into operation as well. Both components and sensors are playing essential roles in operation and control. Components and sensors in VAV air distribution systems often suffer from complete failure (hard fault) and partial failure (soft fault) easily, which result in energy waste, performance degradation or totally out of control. Therefore, fault detection and diagnosis (FDD) for VAV systems, especially for large-scale systems in which dozens of VAV boxes are involved, provides great benefits in improving system control and indoor environment quality, enhancing building energy efficiency, and prolonging components' life. However, studies on FDD for VAV systems are not sufficient and there is no applicable automatic commissioning tool for the whole VAV air distribution systems. An automatic FDD strategy for VAV air distribution systems is developed in this study. A software package is developed on the basis of the FDD strategy for automatic commissioning. Prior to developing the FDD strategy, a site survey on the faults in practical VAV terminals was conducted. It was about a commercial building with 1251 VAV terminals in total. 20.9% VAV terminals were found ineffective and eleven root faults were identified in pressure-independent VAV systems. The FDD strategy therefore chooses these eleven root faults as the objects to be handled. The FDD strategy is built up based on system knowledge, qualitative states and object-oriented SPC (statistical process control) models. Eight FDD schemes, organized at two steps, are set up to detect the eleven VAV root faults within the qualitative/quantitative FDD strategy. Ten faults, which would affect the system operation, are handled at Step 1 in parallel using the first seven schemes. The eleventh fault, which would not affect the basic system operation but would lead to imperfection under advanced supervisory control, is analyzed at Step 2 using the eighth scheme. The FDD strategy is tested and validated on typical VAV air-conditioning systems involving multiple faults both in simulation and in-situ tests. Integrating quantitative models with qualitative knowledge helps to solve decision making problems more effectively and efficiently. Three schemes are developed simply from characteristic equations or based on qualitative states for simple fault detection like controller hard failure or damper stuck. However, other five schemes need further quantitative SPC models for fault detection or identification after the faulty patterns are recognized by characteristic equations or qualitative states. The eighth scheme is developed for VAV terminal flow sensor bias detection and sensor reconstruction. PCA (Principal Component Analysis) models, at both system level and terminal level, are built and employed in the scheme. Sensor biases are detected using both T2 statistic and SPE (Square Prediction Error) and isolated using SPE contribution plot. As the reliability and sensitivity of fault isolation may be affected by the multiple sensor faults at the system level, terminal level PCA model is designed to further examine the suspicious terminals. The faulty sensor is reconstructed after it is isolated by the scheme and the fault detection process repeats using the latest reconstructed measurements until no further fault can be detected. Thus, the sensitivity and robustness of the scheme are enhanced significantly. A software package is developed to implement the FDD strategy for automatic commissioning. With the data downloaded from the BMS, the pre-defined root faults could be detected and faulty sensor(s) could be reconstructed by the software. The main FDD report presents a list of major information of commissioning and a few other reports give other detailed information related to the characteristic parameters of the system concerned. Ph.D., Dept. of Building Services Engineering, The Hong Kong Polytechnic University, 2006.
This review aims to provide an up-to-date, comprehensive, and systematic summary of fault detection and diagnosis (FDD) in building systems. The latter was performed through a defined systematic methodology with the final selection of 221 studies. This review provides insights into four topics: (1) glossary framework of the FDD processes; (2) a classification scheme using energy system terminologies as the starting point; (3) the data, code, and performance evaluation metrics used in the reviewed literature; and (4) future research outlooks. FDD is a known and well-developed field in the aerospace, energy, and automotive sector. Nevertheless, this study found that FDD for building systems is still at an early stage worldwide. This was evident through the ongoing development of algorithms for detecting and diagnosing faults in building systems and the inconsistent use of the terminologies and definitions. In addition, there was an apparent lack of data statements in the reviewed articles, which compromised the reproducibility, and thus the practical development in this field. Furthermore, as data drove the research activity, the found dataset repositories and open code are also presented in this review. Finally, all data and documentation presented in this review are open and available in a GitHub repository.
A localized miniature AHU is a simple device consisting of a heating and cooling heat exchanger or 'coil' and fan. It will form a part of an HVAC system and can be used in residential, commercial, and industrial buildings. It can be used with a ductwork, and is useful to control the temperature in the space where it is installed, or serve multiple spaces. Due to their simplicity and flexibility they seem to be more economical to install than AHUs and air conditioners. Working Principle The localized miniature AHU will work on the basic refrigeration cycle a Design and Operation A miniature AHU will be installed within a service zone. The coil receives hot or cold water from a central plant, and removes heat from or adds heat to the air through heat transfer. It can contain its own internal thermostat or can be wired to operate with a remote thermostat or it can be used with a Building Energy Management System (BEMS) where the control of the AHU unit will be by a local digital controller. Cold and hot water is generated at a chiller or a cooling tower and at boiler or a commercial water heater respectively. Advantages of AHUs AHU's are highly energy efficient compared to electric AIR Conditioners AHU's provide comfortable indoor conditions thus increasing productivity. Clean air supply thus maintaining better health. It provides centralized control It has no risks of leakage of refrigeration gases. Problem with AHU's Though AHU is energy efficient but it has extremely high installation cost. It does not provide a localized control It is economical only when the space is completely occupied. It is not comfortable in situations where there are large number of small zones. It causes mixing of air. Advantages of Split Air Conditioners Ductless and Small size Flexibility: great for zoning or the heating and cooling of individual rooms Better localized control. Capable of saving you energy and thus saving you money Avoid energy loss associated with ductwork, which can account for up to 30% of energy consumption Problems with Split Air Conditioners Extremely high energy consumption High risk of leakage of refrigeration gases. Probable Advantages of Miniature Localized AHU Energy Efficient : Since it uses the same installation as of AHU , thus it is energy efficient compared to the electric air conditioners. Providing Localized Control : It is Economical even when space is not completely occupied and provides localized control for each zone. Most suitable for small zones. No mixing of Air NO risk of refrigerant leaks in the building.
s
Sensors play essential roles in industrial automatic control systems. The faulty or inaccurate sensors may cause uncomfortable thermal environments, shortened component lifetime and energy consumption loss. Considering the condition-adaptive issue of principal component analysis (PCA) models in fault diagnosis, a data-driven optimized statistical model applied for sensor fault detection and diagnosis (FDD) is proposed in the paper: the subtraction clustering and k-means clustering are combined to identify and classify modeling measurements of unsteady operating conditions. Sensor measurements from a real water source heat pump air-conditioning system is tested and the result shows that the clustering-based statistical model can enhance the ability of dealing with data of multiple operation conditions compared with the traditional PCA model; Different statistical indexes show sensitivity difference in detecting faults in the case of same sensors and same faults.
Principal component analysis (PCA) and joint angle method are presented to detect and diagnose fixed bias of temperature, flow rate and pressure sensors on the basis of variable air volume (VAV) simulator developed. According to the normal history data of the system, PCA model is set up to detect the sensor faults by comparing the projection on the residual subspace to which the real measurement vector and normal vector are projected. In addition, joint angle method is used to isolate the fault source on line through comparing the angles of the new fault vector and the known ones in both subspaces. The simulation tests show that the method can detect and diagnose the sensor faults in VAV systems accurately and rapidly.
Supply air temperature optimal and the corresponding fault tolerant control methods for VAV systems were presented. This method can satisfy the higher demand of indoor air quality and humidity strictly required in ASHRAE Standard 62-2001. Moreover, the principal component analysis, joint angle analysis and compensatory reconstruction methods were used to detect, identify and reconstruct the fault respectively so as to achieve the target of fault tolerant and optimal control of the VAV systems. The proposed methods were tested and evaluated in the VAV simulator.
In present paper, an active fault tolerant control (FTC) system is proposed for three phase induction motor (IM) drives subjected to the mechanical faults caused by both stator and rotor failures. FTC design consists of two main parts. The first part is design of a nominal controller in the fault-free case to achieve control objectives. In this work, a feedback linearizing controller has been employed. The second one is the design of a sliding mode observer (SMO) in order to estimate additive faults which model mechanical faults and production of extra control inputs for compensating their undesirable effects on IM performance. SMO is also used for online estimation of axial fluxes in the both conditions. The simulations are performed using Matlab/Simulink software to illustrate the effectiveness of the proposed FTC scheme.
Failures can lead to a series of problems in the complex heating, ventilation and air-conditioning (HVAC) systems in buildings. Fault detection and diagnosis (FDD) is an important technology to solve these problems. Models can represent the behaviors of the HVAC systems, and FDD can be realized with models. Using the model as intermediary, a link between system simulation and FDD can be built. Simulation has provided a convenient platform of operation for FDD, the overall simulation methodology in FDD of HVAC systems is briefly introduced. Various reference models, faulty behaviors, modeling environments, and algorithms for FDD are discussed or compared. Finally, the model-based FDD schemes for HVAC systems proposed by many researchers in various ways have been reviewed.
An overview of commonly used methodologies based on the artificial intelligence approach is provided with a special emphasis on neural networks, fuzzy logic, and genetic algorithms. A description of selected applications to building energy systems of AI approaches is outlined. In particular, methods using the artificial intelligence approach for the following applications are discussed: Prediction energy use for one building or a set of buildings (served by one utility), Modeling of building envelope heat transfer, Controlling central plants in buildings, and Fault detection and diagnostics for building energy systems.
As building systems become more integrated, one inherently introduces coupling between previously independent designs. Model-based controls and analysis tools are keys to identify problems and solution-paths early on. A model-based approach facilitates the developments of monitoring equipment, fault-tolerance, diagnosis and controls algorithms, that are critical in the operation of integrated system. This paper introduces a rich set of problems in controls and optimization related to commercial chilled water cooling plants and related systems. The problems range from opportunities in multivariable optimal control, dynamic resource allocation and control over networks. As an example, dither-based extremum seeking based controls is proposed as an online optimization algorithm for chilled water plant control
A method of fault detection, diagnosis (FDD) and data recovery is proposed for building heating/cooling billing system in this paper. Principal component analysis (PCA) approach is used to extract the correlation of measured variables in heating/cooling billing system and reduce the dimension of measured data. The measured data of billing system under normal operating condition are used to build PCA model. Sensor faults of bias, drifting and complete failure are introduced to building heating/cooling billing system for detection and identification. Square prediction error (SPE) statistic is used to detect sensor faults in the system. Then, sensor validity index (SVI) was employed to identify faulty sensors. Finally, a reconstruction algorithm is presented to recover the correct data of faulty sensor in accordance with the correlations among system variables. A program for the FDD and data recovery method is developed and employed in the heating/cooling billing system of a real small-scale laboratory building to test its applicability and effectiveness. Validation results show that the proposed FDD and data recovery method is correct and effective for most faults in building heating/cooling billing system.
This paper presents theoretical background and results of experimental research into fault detection, isolation and restoration (FDIR) system for rotating machineries created in Department of Robotics and Mechatronics, AGH University of Science and Technology in Cracow. The presented FDIR system solution concerns the problem of automatic balancing of rotating systems while in operation. Simulations and experimental results proved that application of the presented solution leads to a reduction in the vibrations caused by imbalance, without hindering service. A mechatronic approach to the design process makes this process more effective.
In this paper, a new type of self-adaptation mechanism for survivability is proposed. It incorporates intrusion deception and uses it for unknown attack data collection. Attack information is extracted from the collected data through behavior analysis. Different measures are taken to deal with different types of attacks to form new rules. For example, signature generation is used against vulnerability utilization and slammer attacks. Then, the new rules are embedded into the network system to make it self-adaptive against unknown attacks. The concept system of this type of mechanism, its automata model and related formal description and the implementation of the key components are illustrated in this paper.
A methodology is developed for fault detection in heating, ventilating, and air-conditioning (HVAC) systems based on optimal control. The optimal control strategy is determined using information that could come from an energy management and control system (EMCS). This information is also used to detect faults in system operation. Deviations from optimal performance are sensed by comparing the measured system power with the power predicted with the optimal strategy. Various statistical approaches to using the difference between actual and optimal power are described. Both individual measurements and sequences of data are employed to determine the existence and location of faults. The trade-off between the level at which faults can be detected and the speed of detecting faults is discussed. The methodology is tested using simulations of an HVAC system operated both with and without faults present.
Faults in heating, ventilating and air-conditioning (HVAC) systems comprise a range of problems that can increase energy consumption or equipment wear and can adversely affect building occupants. Faults can be detected by measuring equipment performance relative to a model-based prediction or relative to more common-sense expectations of what constitutes acceptable performance. This paper first classifies faults for ventilating systems, consisting of fans, ducts, dampers, heat exchangers, and controls. It then reviews two forms of steady-state, parametric models for the electric power used by ventilation system fans and proposes a third, that of correlating power with variable-speed-drive control signal. The models are compared on the basis of prediction accuracy, sensor requirements, and ability to detect faults. Power-control signal correlations, in preference to other models, can be established with power measurements taken at the HVAC electrical distribution panel when other loads are constant, thereby eliminating the need for submetering. Error analysis associated with model fits, so crucial to determining the minimum detectable fault size, is discussed in detail. Finally, the use of parametric models to distinguish multiple fans from a single electric power measurement is described.
This article describes how the BLAST program was used to simulate the effects of errors in the sensors of automatic controls for HVAC systems. The building selected for this study was an office building of light weight construction, located in the Washington, DC area. The sensors studied were the dry-bulb and dew point sensors used to measure return and outdoor temperatures and humidity levels for enthalpy economizer control, the temperature sensors located in the mixed air stream to control the positions of the outside air and return air dampers, and the discharge air sensors to control the valves of the heating and cooling coils. The computer calculations indicate that sensor errors can increase the annual energy requirements attributable to an air handling system by as much as 30 to 50%. Yearly energy consumption was found to be most sensitive to errors in the mixed air temperature sensor and in the cooling coil discharge sensor.
Dynamic programming was used to obtain optimal service schedules and costs for cleaning the condensers and evaporators of air-conditioning equipment. Results were obtained for a range of service and energy costs, characteristic fouling times, and equipment sizes for a single building and location. Minimum operating costs were compared with regular service intervals (representative of current practice) and a strategy where service is only performed when a constraint is violated (e.g., a comfort violation). It was found that optimal service scheduling reduced lifetime operating costs by as much as a factor of two over regular service intervals, and by 50% when compared to constrained only service. For practical implementation, a simple near-optimal algorithm for estimating optimal service scheduling was developed that does not require on-line forecasting or numerical optimization and is easily implemented within a micro-controller. Over the wide range of cases tested, the near-optimal algorithm gave operating costs that were within 1% of optimal. This technique could also be applied to other systems where performance degradations are important such as large chillers and power plants.