More than $18 billion of energy is wasted annually in the U.S. commercial building sector. Retro-Commissioning services have proven to be successful with relatively short payback times, but tools that support the commissioning effort in maintaining the optimal energy performance in a building are just not readily available. The current work in the field of fault detection and diagnostics of HVAC systems, its cost, complexity and reliance on improved sensor technology, will require years until it can become the mainstay in building energy management. In the meantime, a simplified system is needed today that can be robust and universal enough to use in most types of buildings, address the main concerns of building owners by focusing on consumption deviations that significantly affect the bottom line and provide them some assistance in the remediation of these problems. This thesis presents the results of the development and testing of an advanced prototype of the Automated Building Commissioning Analysis Tool (ABCAT), which has detected three significant energy consumption deviations through four live building implementations. The ABCAT has also demonstrated additional functional benefits of tracking the savings due to retro-commissioning efforts, verifying billed utility data in addition to its primary function of detecting significant consumption faults. Although similar attempts have been made in FDD at the whole building level, the simplification, flexibility, robustness and benefits of this new approach are expected to exhibit the characteristics that will be desired and desperately needed by industry professionals.

Abstract Taiwan, which is located in the subtropics, is in need of air-conditioning system to provide its people with a more comfortableenvironment. Since apparatus is used frequently, equipments may break down easily. Often, we¡¦ll need a great amount of budget for mending these damaged equipments. Therefore, we have come up with an idea in order to overcome the difficulties. AI (artificial intelligence) related technology has been under research for several years. Due to the fact that people demand computers to become more intelligent, there are many new innovations to be developed within the field of AI. Expert system is one of the newest technologies that increase the benefits brought to us through AI. Different from the conventional program, the expert system has plenty of advantages which include graphical interface, inferring mechanism, and knowledge-base, which can store a lot of professional knowledge. Moreover, the program is almost a real expert for it could teach people by its dialogue function. Based on this advantage, the unit can save an amount of money from using the expert system. Prolog is a program language used to write expert systems. It was invented by Colmerauer and Roussel in 1972. Because of its powerful function, Japan uses it as a tool to create the Fifth Computer Generation System (FGCS). According to the plan, the system must contain many user-friendly functions, which enables it to receive information by listening to people¡¦s voice, and it can learn everything ¡V almost as much as human beings. As for the programming of the VII language, Prolog uses a special syntax. When the predicate matches the fact, the function would be executed. This special program syntax is good for describing things. From now on, Prolog will be much more welcomed because it works in a more acceptable way for people. Finally, we would like to promote the techniques to the field of air-conditioning in order to offer people a new solution to deal with the flaws.

This thesis investigates the development of an automated fault detection system developed for a novel lightweight railway material haulage system; in particular, the study aims to detect railway track faults at the incipient stage to determine the feasibility of maintenance decision support, ultimately with the function of preventing catastrophic failure. The proposed approach is an extension of the current state of the art in fault detection of unsteady machinery. The most common railway track faults associated with train derailment were considered; namely, horizontal and transverse crack propagation, mechanical looseness, and railbed washout were the faults of interest. A series of field experiments were conducted to build a database of vibration, speed, and localization data in healthy and faulted states. These data were used to develop, investigate, and validate the effectiveness of various approaches for fault detection. A variety of feature sets and classification approaches were investigated to determine the best overall configuration for the fault detector. The feature sets were used to condense data segments and extract characteristics that were sensitive to damage, but insensitive to healthy variations due to unsteady operation. The pattern recognition classifiers were used to categorize new data members as belonging to the healthy class or faulted class. The fault detection results from the proposed approach were promising. The feasibility of an automated online fault detection system for the lightweight material haulage system examined in this study was confirmed. The conclusions of this research outline the major potential for an iv effective fault detection system and address future work for the practical implementation of this system.

This thesis is an investigation of possible applications for a low cost non-contact magnetic rotational position sensor. A single stage gearbox operating spur gears was instrumented with these sensors along with typical optical encoders. These rotational position devices were used independently to measure gearbox Transmission Error (TE) during operation. Basic filtering techniques were used to condition the TE so that localized faults were observable. Characteristic feature extraction on the TE using RMS, Kurtosis and Crest Factor was used to quantify gearbox dynamics. These features were able to measure dynamic changes in gearbox health, such as wearing in the gears or the progression of a fault resulting in full tooth failure. These sensor attributes are ideal for machine condition monitoring applications where catastrophic failure can be forewarned by incipient fault detection. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2010-10-12 17:21:13.125

Production downtime caused by machine faults presents a major area of concern for the manufacturing industry and can especially impact the productivity of assembly systems. Traditional fault detection systems use a variety of conventional sensors that measure operating variables such as pressure, force, speed, current and temperature. Faults are detected when a reading from one of these sensors exceeds a preset threshold or does not match the predicted value provided by a mathematical model of the system. The primary disadvantage of these methods is that the relationship between sensor reading and fault is often indirect (if one exists at all). This can lead to time delays between fault occurrence and ‘fault reading’ from a sensor, during which additional machine damage could accumulate. This thesis describes progress with a project whose goal is to examine the effectiveness and feasibility of using machine vision to detect ‘visually cued’ machine faults in automated assembly equipment. It is proposed that machine vision technology could complement traditional methods and improve existing detection systems. Two different vision-based fault detection methods were developed and tests were conducted using a laboratory-scale assembly machine that assembles a simple 3-part component Typical faults that occurred with this machine were targeted for inspection. The first method was developed using Automated Visual Inspection (AVI) techniques that have been used extensively for quality inspection of manufactured products. The LabVIEW 2010 software was used to develop the system. Test results showed that the Colour Inspection tool performed the best with 0% false negative and false positive fault detection rates. Despite some success, this approach was found to be limited as it was unable to detect faults that varied in physical appearance or those that had not been identified prior to testing. The second method was developed using a video event detection method (spatiotemporal volumes) that has previously been used for traffic and pedestrian monitoring. This system was developed with MATLAB software and demonstrated strong false negative and false positive fault detection rates. It also showed the ability to detect faults that had not previously been identified as well as those that varied in appearance. Recommendations were made for future work to further explore these methods. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2012-07-13 16:04:57.829

Artificial neural networks (ANNs) have been used in many fault detection and identification (FDI) applications due to their pattern recognition abilities. In this study, two ANNs, a supervised network based on Backpropagation (BP) learning and an unsupervised network based on Adaptive Resonance Theory (ART-2A), were tested for FDI on an automated assembly machine and compared to a conventional rule-based method. Three greyscale sensors and two redundant limit switches were used as cost-effective sensors to monitor the machine's operating condition. To test each method, sensor data were collected while the machine operated under normal conditions, as well as 10 fault conditions. Features were selected from the raw sensor data to create data sets for training and testing. The performance of the methods was evaluated with respect to their ability to detect and identify known, unknown and multiple faults. Their modelling and computational requirements were also considered as performance measures. Results showed that all three methods were able to achieve perfect classification with the test data sets; however, the BP method could not classify unknown or multiple faults. In all cases, the performance depended on careful tuning of each method’s parameters. The BP method required an ideal number of neurons in the hidden layer and good initialization. The ART-2A method required tuning of its classification parameter. The rule-based method required tuning of its thresholds. Although it was found that the rule-based system required more effort to set up, it was judged to be more useful when unknown or multiple faults were present. The ART-2A network created new outputs for these conditions, but it could not give any more information as to what the new fault was. By contrast, the rule-based method was able to generate symptoms that clearly identified the unknown and multiple fault conditions. Thus, the rule-based method was judged to be the best overall method for this type of application. It is recommended that future work examine the application of computer vision-based techniques to FDI with the assembly machine. The results from this study, using cost-effective sensors, could then be used as a performance benchmark for image-based sensors. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2014-05-16 17:21:13.676

Rolling element bearings are one of the most significant elements and frequently-used components in mechanical systems. Bearing fault detection and diagnosis is important for preventing productivity loss and averting catastrophic failures of mechanical systems. In industrial applications, bearing life is often difficult to predict due to different application conditions, load and speed variations, as well as maintenance practices. Therefore, reliable fault detection is necessary to ensure productive and safe operations. Vibration analysis is the most widely used method for detection and diagnosis of bearing malfunctions. A measured vibration signal from a sensor is often contaminated by noise and vibration interference components. Over the years, many methods have been developed to reveal fault signatures, and remove noise and vibration interference components. Though many vibration based methods have been proposed in the literature, the high frequency resonance (HFR) technique is one of a very few methods have received certain industrial acceptance. However, the effectiveness of the HFR methods depends, to a great extent, on some parameters such as bandwidth and centre frequency of the fault excited resonance, and window length. Proper selection these parameters is often a knowledge-demanding and time-consuming process. In particular, the filter designed based on the improperly selected bandwidth and center frequency of the fault excited resonance can filter out the true fault information and mislead the detection/diagnosis decisions. In addition, even if these parameters can be selected properly at beginning of each process, they may become invalid in a time-varying environment after a certain period of time. Hence, they may have to be re-calculated and updated, which is again a time-consuming and error-prone process. This undermines the practical significance of the above methods for online monitoring of bearing conditions. To overcome the shortcomings of existing methods, the following four non-parametric and non-filtering methods are proposed: 1. An amplitude demodulation differentiation (ADD) method, 2. A calculus enhanced energy operator (CEEO) method, 3. A higher order analytic energy operator (HO_AEO) approach, and 4. A higher order energy operator fusion (HOEO_F) technique. The proposed methods have been evaluated using both simulated and experimental data.

A complete method for calculating the largest Lyapunov exponent is developed in this thesis. For phase space reconstruction, a time delay estimator based on the average mutual information is discussed first. Then, embedding dimension is evaluated according to the False Nearest Neighbors algorithm. To obtain the parameters of all of the sub-functions and their derivatives, a multilayer feedforward neural network is applied to the time series data, after the time delay and embedding dimension are fixed. The Lyapunov exponents can be estimated using the Jacobian matrix and the QR decomposition. The possible applications of this method are then explored for various chaotic systems. Finally, the method is applied to some real world data to demonstrate the general relationship between the onset and progression of faults and changes in the largest Lyapunov exponent of a nonlinear system.

Rolling element bearings are one of the most significant elements and frequently-used components in mechanical systems. Bearing fault detection and diagnosis is important for preventing productivity loss and averting catastrophic failures of mechanical systems. In industrial applications, bearing life is often difficult to predict due to different application conditions, load and speed variations, as well as maintenance practices. Therefore, reliable fault detection is necessary to ensure productive and safe operations. Vibration analysis is the most widely used method for detection and diagnosis of bearing malfunctions. A measured vibration signal from a sensor is often contaminated by noise and vibration interference components. Over the years, many methods have been developed to reveal fault signatures, and remove noise and vibration interference components. Though many vibration based methods have been proposed in the literature, the high frequency resonance (HFR) technique is one of a very few methods have received certain industrial acceptance. However, the effectiveness of the HFR methods depends, to a great extent, on some parameters such as bandwidth and centre frequency of the fault excited resonance, and window length. Proper selection these parameters is often a knowledge-demanding and time-consuming process. In particular, the filter designed based on the improperly selected bandwidth and center frequency of the fault excited resonance can filter out the true fault information and mislead the detection/diagnosis decisions. In addition, even if these parameters can be selected properly at beginning of each process, they may become invalid in a time-varying environment after a certain period of time. Hence, they may have to be re-calculated and updated, which is again a time-consuming and error-prone process. This undermines the practical significance of the above methods for online monitoring of bearing conditions. To overcome the shortcomings of existing methods, the following four non-parametric and non-filtering methods are proposed: 1. An amplitude demodulation differentiation (ADD) method, 2. A calculus enhanced energy operator (CEEO) method, 3. A higher order analytic energy operator (HO_AEO) approach, and 4. A higher order energy operator fusion (HOEO_F) technique. The proposed methods have been evaluated using both simulated and experimental data.

Underground power cables encounter various problems caused by manufacturing defects and/or environmental contact. In keeping with the Smart Grid vision, researchers must develop diagnostic techniques that can be utilized to facilitate the decision making processes regarding replacement prior to failure can occur, thereby minimizing impact to customers. Due to the impact of the aging infrastructure and in particular underground polymeric cables, various offline and online methods have been developed for the detection of the remaining life of underground cables. The offline methods require power outage, which can lead to further difficulty in their implementation. Signal processing techniques hold promise to provide real time or near real time diagnostics. In this thesis, three different signal processing techniques; fast Fourier transform, short-time Fourier transform, and wavelet transform; are investigated for identifying and classifying various fault types encountered in underground power cables based on cable current and voltage measurements.