A fault detection and diagnosis strategy of VAV air-conditioning systems for improved energy and control performances

This paper presents the results of a site survey study on the faults in variable air volume (VAV) terminals and an automatic fault detection and diagnosis (FDD) strategy for VAV air-conditioning systems using a hybrid approach. The site survey study was conducted in a commercial building. 20.9% VAV terminals were ineffective and 10 main faults were identified in the VAV air-conditioning systems. The FDD strategy adopts a hybrid approach utilizing expert rules, performance indexes and statistical process control models to address these faults. Supported by a pattern recognition method, expert rules and performance indexes based on system physical characteristics are adopted to detect 9 of the 10 faults. Two pattern recognition indexes are introduced for fault isolation to overcome the difficulty in differentiating damper sticking and hysteresis from improper controller tuning. A principal component analysis (PCA)-based method is developed to detect VAV terminal flow sensor biases and to reconstruct the faulty sensors. The FDD strategy is tested and validated on typical VAV air-conditioning systems involving multiple faults both in simulation and in situ tests.     

A robust fault detection and diagnosis strategy for pressure-independent VAV terminals of real office buildings

A robust fault detection and diagnosis (FDD) strategy using a hybrid approach is presented for pressure-independent variable air volume (VAV) terminals in this paper. The residual-based cumulative sum (CUSUM) control charts are utilized to detect faults in VAV terminals. The residuals between the temperature error and its predication are generated using autoregressive time-series models. The standard CUSUM control charts are used to monitor the residuals which are statistically independent. If the CUSUM value exceeds the chart limits, it means the occurrence of fault or abnormity in the corresponding VAV terminal. The residual-based CUSUM control chart can improve the accuracy of fault detection through eliminating the effects of serial correlation on the performance of control charts. Also, the residual-based CUSUM control chart can enhance the robustness and reliability of fault detection through reducing the impacts of normal transient changes. A rule-based fault classifier consisting of expert rules and fault isolation algorithms is developed to isolate 15 fault sources. The FDD strategy was online tested and validated using in real time data collected from real VAV air-conditioning systems.     

Online fault detection and diagnosis in VAV air handling unit by RARX modeling

Assimilation of cost-effective fault detection and diagnosis (FDD) technique in building management system can save enormous amount of energy and material. In this paper, recursive autoregressive exogenous algorithm is used to develop dynamic FDD model for variable air volume (VAV) air handling units. A methodology, based upon frequency response of the model is evolved for automatic fault detection and diagnosis. Results are validated with data obtained from a real building after introducing artificial faults. Robustness of the method is further established against sensor errors arising out of faulty bias during long term use or lack of proper commissioning. It is concluded that the method is quite robust and can detect and diagnose several types of faults. A short and simple method is also included in this paper to detect the faults of VAV units operating in the same zone by comparing their behavior. The new method, which requires very small amount of computation time, was tested with the aforementioned database and shows satisfactory results.     

A study on the energy penalty of various air-side system faults in buildings

Automatic fault detection and diagnosis (FDD) can help enhance building energy efficiency by facilitating early detection of occurrence of system faults, especially those of air-conditioning systems, thus enabling rectification of the faults before much energy is wasted due to such faults. However, building owners may not invest in FDD unless they are convinced of the energy cost savings that can be achieved. This paper presents the results of a study on the energy cost impacts of a range of common system faults in variable air volume (VAV) air-conditioning systems, which are widely adopted for their good part-load energy efficiency. The faults studied include room air temperature sensor offset, stuck VAV box damper, supply air temperature sensor offset, stuck outdoor air damper and stuck/leaking cooling coil valve. The simulation results indicate that some faults may significantly increase energy use in buildings, for example, negative room air temperature sensor offset, stuck open VAV box damper, negative supply air temperature sensor offset, stuck open outdoor air damper and stuck open and leaking cooling coil valve. Since building occupants may adapt to the symptoms of these faults, such as reduced room air temperature, and thus may not complain about them, the occurrence of such faults are not immediately apparent unless a FDD system is available. Some other faults, e.g. positive supply air temperature sensor offset, positive room air temperature sensor offset, stuck closed cooling coil valve and stuck closed VAV box damper, may allow less energy to be used but will lead to unbearable indoor environmental conditions, such as high indoor temperature. Such faults, therefore, can easily be detected even without a FDD system, as there will be feedback from the building occupants.     

A development of easy-to-use tool for fault detection and diagnosis in building air-conditioning systems

There are many reports about faulty status in building air-conditioning systems recently. It becomes difficult to keep indoor air temperature appropriately as faults occur, and the faults cause waste of building energy consumption. The model-based fault detection and diagnosis (FDD) methods have been researched for specific parts of air-conditioning system such as chillers, coils, variable air volume units (VAV units), etc. It needs, however, much time and labor to monitor and check every single part because we cannot predict where and when the faults occur. The purpose of this study is to examine indoor air temperature changes and energy consumption increase when faults occur and to develop an easy-to-use FDD tool that helps to find out the faulty place through the whole building air-conditioning system. And then, we treat the reliability of the proposed FDD tool and effectiveness to control of indoor environment deterioration and energy consumption increase by the tool is evaluated based on building air-conditioning system simulation in this paper.     

A system-level fault detection and diagnosis strategy for HVAC systems involving sensor faults

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.     

冷凝器结垢故障对传感器故障诊断方法的影响分析

传感器测量质量是实施系统监控，优化控制及部件故障诊断的重要保证。制冷机系统的部件故障和传感器故障很有可能同时发生。因而在研究传感器故障诊断的方法及应用这一方法进行传感器故障诊断时必须考虑到制冷机的部件故障的影响。根据离心式制冷机的试验数据研究了冷凝器结垢故障对基于主元分析法的制冷机传感器故障诊断方法的影响，研究结果表明基于主元分析的制冷机传感器方法对冷凝器结垢故障不敏感，该方法对有无结垢故障条件下的传感器故障都能成功地进行检测诊断及数据重构。     

A critical review of fault modeling of HVAC systems in buildings

Buildings consumed about 40% of primary energy and 70% of the electricity in the U.S. It is well known that most buildings lose a portion of their desired and designed energy efficiency in the years after they are commissioned or recommissioned. Majority of the Heating, Ventilation, and Air-Conditioning (HVAC) systems have multiple faults residing in the systems causing either energy, thermal comfort, or indoor air quality penalties. There are hundreds of fault detection and diagnostics (FDD) algorithms available, but there is lacking a common framework to assess and validate those FDD algorithms. Fault modeling is one of the key components of such a framework. In general, fault modeling has two purposes: testing and assessment of FDD algorithms, and fault impacts analysis in terms of building energy consumption and occupants’ thermal comfort. It is expected that fault ranking from the fault impact analysis can facilitate building facility managers to make decisions. This paper provides a detailed review of current state-of-the-art for the fault modeling of HVAC systems in buildings, including fault model, fault occurrence probability, and fault simulation platform. Fault simulations considering fault occurrence probability can generate realistic faulty data across a variety of faulty operating conditions, and facilitate testing and assessment of different FDD algorithms. They can also help the fault impact study. Three research gaps are identified through this critical literature review: (1) The number of available fault models of HVAC systems is still limited. A fault model library could be developed to cover all common HVAC faults for both traditional and non-traditional HVAC systems. (2) It is imperative to include the fault occurrence probability in fault simulations for a realistic fault impacts analysis such as fault ranking. (3) Fault simulation platforms need further improvements to better facilitate the fault impact analysis.     

基于主成分法的VAV系统传感器故障检测与诊断

众所周知，VAV系统对控制的要求很高，作为控制系统中非常关键的元件——传感器，一旦出现故障将直接影响控制系统的决策，从而使VAV系统的运行偏离设计要求。因此，VAV系统传感器的故障检测与诊断研究是很有必要的。本文采用主成分分析法(PCA，Principal Component Analysis)对传感器故障的检测、确认与重构进行分析，以期获得一种可行的方案。     

A diagnostic tool for online sensor health monitoring in air-conditioning systems

Healthy sensors are essential for the reliable monitoring and control of building automation systems (BAS). This paper presents a diagnostic tool to be used to assist building automation systems for online sensor heath monitoring and fault diagnosis of air-handling units. The tool employs a robust sensor fault detection and diagnosis (FDD) strategy based on the Principal Component Analysis (PCA) method. Two PCA models are built corresponding to the heat balance and pressure-flow balance of an air-handling process. Sensor faults are detected using the Q-statistic and diagnosed using an isolation-enhanced PCA method that combines the Q-contribution plot and knowledge-based analysis. The PCA models are updated using a condition-based adaptive scheme to follow the normal shifts in the process due to changing operating conditions. The sensor FDD strategy, the implementation of the diagnostic tool and experimental results in an existing building are presented in this paper.     

基于BAS的空调系统过程监测与故障诊断

提出了基于BAS的空调系统过程监测与故障诊断模式，阐述用主成分分析法进行故障诊断的过程，并用其进行空调系统的传感器故障诊断。结果表明，主成分分析法具有很好的故障检测、故障识别和故障重构能力，还表明，故障诊断与BAS相结合是可行的。     

Heating system performance estimation using optimization tool and BEMS data

Causes and effects of a few real faults in a hydronic heating system are explained in this paper. Since building energy management system (BEMS) has to be utilized in fault detection and diagnosis (FDD), practical explanations of faults and their related effects are important to building caretakers. A simple heat balance model is used in this study. The model is calibrated using the optimization tool. Site data from the BEMS of a real building are calibrated against the model. Desired and real data are compared, so that the effects of the following faults are analyzed: faults in an outdoor air temperature sensor, fault in the time schedule, and a water flow imbalance problem. This paper presents an overview of the real causes of the faults and their effects both on the energy consumption and on the indoor air temperature. In addition, simple instructions for the building caretakers for fault detection in the hydronic heating systems are given.     

用小波分析法分离变风量系统流量传感器故障

阐述了主成分分析法对变风量（VAV）系统流量传感器故障识别失效的原因,给出了小波分析法的特点及其原理,着重描述了如何运用小波分析分离VAV系统中的共线性流量传感器故障。实例研究结果表明：小波分析能够很好地识别共线性的流量传感器故障。     

A hybrid FDD strategy for local system of AHU based on artificial neural network and wavelet analysis

This paper presents a self-adaptive sensor fault detection and diagnosis (FDD) strategy for local system of air handing unit (AHU). This hybrid strategy consists of two stages. In the first stage, a fault detection model for the AHU control loop including two back-propagation neural network (BPNN) models is developed. BPNN models are trained by the normal operating data of system. Based on sensitive analysis for the first BPNN model, the second BPNN model is constructed in the same control loop. In the second stage, a fault diagnosis model is developed which combines wavelet analysis method with Elman neural network. The wavelet analysis is employed to process the measurement data by extracting the approximation coefficients of sensor measurement data. The Elman neural network is used to identify sensor faults. A new approach for increasing adaptability of sensor fault diagnosis is presented. This approach gains clustering information of the approximations coefficients by fuzzy c-means (FCM) algorithm. Based on cluster information of the approximation coefficients, the unknown sensor fault can be identified in the control loop. Simulation results in this paper show that this strategy can successfully detect and diagnose fixed biases and drifting fault of sensors for the local system of AHU.     

A robot fault diagnostic tool for flow rate sensors in air dampers and VAV terminals

A robot fault diagnostic tool for flow rate sensors in air dampers and VAV terminals is presented to ensure well capacity of energy conservation in building air conditioning systems. Principal component analysis (PCA) is used to detect the sensor faults including fixed bias, drifting bias and complete failure. To improve the detection efficiency, several PCA models are built through employing the conservation equations and control relations of the system. With the historical data, PCA models are trained to capture most useful information of normal operation. As a result, the training models can identify whether the present condition is abnormal through comparing the residues with the thresholds. Since the principal component subspace and residue subspace of the operation data space are obtained using PCA decomposition, these two subspaces are used to develop the fault isolation scheme. The new fault detected and the known ones in the library are all projected into the principal component subspace and residue subspace decomposed by PCA. The joint angle plot, illustrating the direction relations of the projections in both subspaces between the new fault and the known ones, is used to diagnose the fault source.     

基于IPSO-DBN 的管道故障诊断方法

针对城市燃气管道故障诊断效果不佳的问题，提出了一种基于改进粒子群算法优化深度信念网络（IPSO-DBN）的管道故障诊断方法。该方法首先对粒子群算法（PSO）中的惯性权重ω、加速因子C1 和C2 进行修正，得到改进粒子群优化算法（IPSO），并采用两种基准函数对比测试PSO 与IPSO 的网络性能，证明所选改进方法的优越性。其次利用IPSO 优化深度信念网络（DBN）的初始权重，建立合适的DBN 网络，将4 种不同燃气管道工况下的实验数据用于IPSO- DBN 网络训练及预测。最后将实验所得的故障诊断准确率与BP、DBN、PSO-DBN 方法进行对比分析。实验结果表明，对于燃气管道不同工况下的故障分类识别，IPSO- DBN 方法的平均测试集诊断准确率高达94.5%，诊断效果优于传统的BP、DBN 以及PSO-DBN 方法。     

Fault detection and diagnosis based on improved PCA with JAA method in VAV systems

In this paper, improved principal component analysis (PCA) with joint angle analysis (JAA) is presented to detect and diagnose both fixed and drifting biases of sensors in variable air volume (VAV) systems. Fault characteristic concerned in PID controller in the VAV systems is analyzed and discussed. The squared prediction error (SPE) plot based on PCA is used to detect the sensor fixed and drifting biases. Then the JAA plot instead of conventional contribution plot is used to diagnose the faults. And they are tested and evaluated online in a simulated centralized VAV air-conditioning systems.     

Detection and diagnosis for multiple faults in VAV systems

Detection and diagnosis of multiple faults in variable air volume (VAV) systems using principal component analysis (PCA) and joint angle analysis (JAA) are presented. Multi-level principal component analysis models including system- and local-level are built to detect the multiple faults occurred in VAV systems simultaneously. As the initial detection, system-level principal component analysis model is used to discover the abnormities in view of the whole systems. And two local-level principal component analysis models are used to further confirm the occurrence of the faults. With the multiple faults separated into different locations, joint angle analysis is used to isolate the faults one by one according to corresponding local models.     

基于HVAC信息流的故障检测与诊断系统研究

介绍了一种新的HVAC故障检测与诊断系统，该系统利用人工神经网络技术处理信息流，在信息流分析的基础上，采用面向对象的编程思想，和基于模型的FDD方法相结合，最后描述了该故障检测与诊断系统的实现过程。