Application of machine learning in fault diagnostics of mechanical systems

A diagnostic method based on Bayesian Networks (probabilistic graphical models) is presented. Unlike conventional diagnostic approaches, in this method instead of focusing on system residuals at one or a few operating points, diagnosis is done by analyzing system behavior patterns over a window of operation. It is shown how this approach can loosen the dependency of diagnostic methods on precise system modeling while maintaining the desired characteristics of fault detection and diagnosis (FDD) tools (fault isolation, robustness, adaptability, and scalability) at a satisfactory level. As an example, the method is applied to fault diagnosis in HVAC systems, an area with considerable modeling and sensor network constraints.

Complete and competent engineers : a coaching model to developing holistic graduates

The growing gap between engineering practice and engineering has been identified at the level of certain essential skills needed among practising engineers but not developed through the current education system. Coaching approach to learning and teaching has been proven to be an effective way to develop people in the workplace. A pilot coaching program is proposed to engineering students at Queensland University of Technology to enable holistic growth in order to better integrate them to the work force and society at large. The success measures and insights gained will be published on completion of the program. It is hoped that the outcomes of this study will better inform curriculum design and development in the engineering disciplines towards better transition between engineering education and engineering practice.

Experimental investigation of post-fire mechanical properties of cold-formed steels

Cold-formed steel members are widely used in residential, industrial and commercial buildings as primary load-bearing elements. During fire events, they will be exposed to elevated temperatures. If the general appearance of the structure is satisfactory after a fire event then the question that has to be answered is how the load bearing capacity of cold-formed steel members in these buildings has been affected. Hence after such fire events there is a need to evaluate the residual strength of these members. However, the post-fire behaviour of cold-formed steel members has not been investigated in the past. This means conservative decisions are likely to be made in relation to fire exposed cold-formed steel buildings. Therefore an experimental study was undertaken to investigate the post-fire mechanical properties of cold-formed steels. Tensile coupons taken from cold-formed steel sheets of three different steel grades and thicknesses were exposed to different elevated temperatures up to 800 oC, and were then allowed to cool down to ambient temperature before they were tested to failure. Tensile coupon tests were conducted to obtain their post-fire stress-strain curves and associated mechanical properties (yield stress, Young’s modulus, ultimate strength and ductility). It was found that the post-fire mechanical properties of cold-formed steels are reduced below the original ambient temperature mechanical properties if they had been exposed to temperatures exceeding 300 oC. Hence a new set of equations is proposed to predict the post-fire mechanical properties of cold-formed steels. Such post-fire mechanical property assessments allow structural and fire engineers to make an accurate prediction of the safety of fire exposed cold-formed steel buildings. This paper presents the details of this experimental study and the results of post-fire mechanical properties of cold-formed steels. It also includes the results of a post-fire evaluation of cold-formed steel walls.

Using DNA as a drug : challenges and opportunities for process engineers

The maturing of the biotechnology industry and a focus on productivity has seen a shift from discovery science to small-scale bench-top research to higher productivity, large scale production. Health companies are aggressively expanding their biopharmaceutical interests, an expansion which is facilitated by biochemical and bioprocess engineering. An area of continuous growth is vaccines. Vaccination will be a key intervention in the case of an influenza pandemic. The global manufacturing capacity for fast turn around vaccines is currently woefully inadequate at around 300 million shots. As the prevention of epidemics requires > 80 % vaccination, in theory the world should currently be aiming for the ability to produce around 5.3 billion vaccines. Presented is a production method for the creation of a fast turn around DNA vaccine. A DNA vaccine could have a production time scale of as little as two weeks. This process has been harnessed into a pilot scale production system for the creation of a pre-clinical grade malaria vaccine in a collaborative project with the Coppel Lab, Department of Microbiology, Monash University. In particular, improvements to the fermentation, chromatography and delivery stages will be discussed. Consideration will then be given as to how the fermentation stage affects the mid and downstream processing stages.

A new method for the in vivo identification of mechanical properties in arteries from cine MRI images: Theoretical framework and validation

Quantifying the stiffness properties of soft tissues is essential for the diagnosis of many cardiovascular diseases such as atherosclerosis. In these pathologies it is widely agreed that the arterial wall stiffness is an indicator of vulnerability. The present paper focuses on the carotid artery and proposes a new inversion methodology for deriving the stiffness properties of the wall from cine-MRI (magnetic resonance imaging) data. We address this problem by setting-up a cost function defined as the distance between the modeled pixel signals and the measured ones. Minimizing this cost function yields the unknown stiffness properties of both the arterial wall and the surrounding tissues. The sensitivity of the identified properties to various sources of uncertainty is studied. Validation of the method is performed on a rubber phantom. The elastic modulus identified using the developed methodology lies within a mean error of 9.6%. It is then applied to two young healthy subjects as a proof of practical feasibility, with identified values of 625 kPa and 587 kPa for one of the carotid of each subject.

Sensitivity of polyvinylidene fluoride films to mechanical vibration modes and impact after optimizing stretching conditions

The β-phase of polyvinylidene fluoride (PVDF) is well known for its piezoelectric properties. PVDF films have been developed using solvent cast method. The films thus produced are in α-phase. The α-phase is transformed to piezoelectric β-phase when the film is hot-stretched with various different stretching factors at various different temperatures. The films are then characterized in terms of their mechanical properties and surface morphological changes during the transformation from α- to β-phases by using X-ray diffraction, differential scanning calorimeter, Raman spectra, Infrared spectra, tensile testing, and scanning electron microscopy. The films showed increased crystallinity with stretching at temperature up to 80°C. The optimum conditions to achieve β-phase have been discussed in detail. The fabricated PVDF sensors have been tested for free vibration and impact on plate structure, and its response is compared with conventional piezoelectric wafer type sensor. The resonant and antiresonant peaks in the frequency response of PVDF sensor match well with that of lead zirconate titanate wafer sensors. Effective piezoelectric properties and the variations in the frequency response spectra due to free vibration and impact loading conditions are reported. POLYM. ENG. SCI., 2012. © 2012 Society of Plastics Engineers.

Quantitative estimation of mechanical and optical properties from ultrasound assisted optical tomography data

We demonstrate quantitative optical property and elastic property imaging from ultrasound assisted optical tomography data. The measurements, which are modulation depth M and phase phi of the speckle pattern, are shown to be sensitively dependent on these properties of the object in the insonified focal region of the ultrasound (US) transducer. We demonstrate that Young's modulus (E) can be recovered from the resonance observed in M versus omega (the US frequency) plots and optical absorption (mu(a)) and scattering (mu(s)) coefficients from the measured differential phase changes. All experimental observations are verified also using Monte Carlo simulations. (c) 2012 Society of Photo-Optical Instrumentation Engineers (SPIE). DOI: 10.1117/1.JBO.17.10.101507]

Effect of silane functionalized alumina on poly(vinyl butyral) nanocomposite films: Thermal, mechanical, and moisture barrier studies

In this work, a hybrid-polymer nanocomposite film, based on polyvinyl butyral/amino-silane functionalized nano alumina, was fabricated by melt processing. The calcium degradation measurements suggest the functionalized nanocomposite films exhibit higher resistance towards moisture penetration as compared to the neat alumina loaded films. Thermal stability, mechanical strength, and contact angle studies of the composites were also conducted to evaluate the performance of the functionalized alumina loaded films. These nanocomposite films were encapsulated over Al/P3HT/ITO Schottky structured device. The changes observed in the current density of the devices to the applied voltage before and after accelerated aging conditions are presented. The nanocomposite with functionalized alumina films exhibits 50% change in current density, which is superior to that attained with neat and non-functionalized films. POLYM. COMPOS., 35:1426-1435, 2014. (c) 2013 Society of Plastics Engineers

Improving the Sensitivity and Bandwidth of In-Plane Capacitive Microaccelerometers Using Compliant Mechanical Amplifiers

This paper presents a method to enhance both the sensitivity and bandwidth of in-plane capacitive micromachined accelerometers by using compliant mechanical amplifiers, and thus obviating the compromise between the sensitivity and bandwidth. Here, we compare one of the most sensitive single-axis capacitive accelerometers and another with large resonant frequency reported in the literature with the modified designs that include displacement-amplifying compliant mechanisms (DaCMs) occupying the same footprint and under identical conditions. We show that 62% improvement in sensitivity and 34% improvement in bandwidth in the former, and 27% and 25% in the latter can be achieved. Also presented here is a dual-axis accelerometer that uses a suspension that decouples and amplifies the displacements along the two in-plane orthogonal axes. The new design was microfabricated, packaged, and tested. The device is 25-mu m thick with the interfinger gap as large as 4 m. Despite the simplicity of the microfabrication process, the measured axial sensitivity (static) of about 0.58 V/g for both the axes was achieved with a cross-axis sensitivity of less than +/- 2%. The measured natural frequency along the two in-plane axes was 920 Hz. Displacement amplification of 6.2 was obtained using the DaCMs in the dual-axis accelerometer. 2013-0083]

Study of polymer thermal and mechanical properties using a fiber Bragg grating thermal sensor

A simple and practical method for the study of polymer thermal and mechanical properties using a fiber Bragg grating (FBG) sensor is presented for the first time, in which the FBG is embedded in a typical epoxy polymer. By measuring the sensitivity change of the FBG sensor, changes of the thermal-mechanical properties of the polymer with temperature and pressure can be measured. The experimental results show that this technique is capable of providing continuous in-line monitoring such properties with high sensitivity during transformation between the glassy state and the rubbery state of a polymer within the temperature and pressure range of 20 to 180 C and 0 to 15 MPa. (c) 2007 Society of Photo-Optical Instrumentation Engineers.

Study of polymer thermal and mechanical properties using a fiber Bragg grating thermal sensor

A simple and practical method for the study of polymer thermal and mechanical properties using a fiber Bragg grating (FBG) sensor is presented for the first time, in which the FBG is embedded in a typical epoxy polymer. By measuring the sensitivity change of the FBG sensor, changes of the thermal-mechanical properties of the polymer with temperature and pressure can be measured. The experimental results show that this technique is capable of providing continuous in-line monitoring such properties with high sensitivity during transformation between the glassy state and the rubbery state of a polymer within the temperature and pressure range of 20 to 180 C and 0 to 15 MPa. (c) 2007 Society of Photo-Optical Instrumentation Engineers.