Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Anomaly detection is an important problem that has been researched within diverse research areas and application domains. Many anomaly detection techniques have been specifically developed for certain application domains, while others are more generic. This survey tries to provide a structured and comprehensive overview of the research on anomaly detection. We have grouped existing techniques into different categories based on the underlying approach adopted by each technique. For each category we have identified key assumptions, which are used by the techniques to differentiate between normal and anomalous behavior. When applying a given technique to a particular domain, these assumptions can be used as guidelines to assess the effectiveness of the technique in that domain. For each category, we provide a basic anomaly detection technique, and then show how the different existing techniques in that category are variants of the basic technique. This template provides an easier and more succinct understanding of the techniques belonging to each category. Further, for each category, we identify the advantages and disadvantages of the techniques in that category. We also provide a discussion on the computational complexity of the techniques since it is an important issue in real application domains. We hope that this survey will provide a better understanding of the different directions in which research has been done on this topic, and how techniques developed in one area can be applied in domains for which they were not intended to begin with.

/pdf/anomaly-detection-a-survey-1x33g9m0a3.pdf

Anomaly detection: A survey

物件導向軟體之架構(Object-Oriented Software Construction)探討

A review on machinery diagnostics and prognostics implementing condition-based maintenance

This report contains a review of the technical literature concerning the detection, location, and characterization of structural damage via techniques that examine changes in measured structural vibration response. The report first categorizes the methods according to required measured data and analysis technique. The analysis categories include changes in modal frequencies, changes in measured mode shapes (and their derivatives), and changes in measured flexibility coefficients. Methods that use property (stiffness, mass, damping) matrix updating, detection of nonlinear response, and damage detection via neural networks are also summarized. The applications of the various methods to different types of engineering problems are categorized by type of structure and are summarized. The types of structures include beams, trusses, plates, shells, bridges, offshore platforms, other large civil structures, aerospace structures, and composite structures. The report describes the development of the damage-identification methods and applications and summarizes the current state-of-the-art of the technology. The critical issues for future research in the area of damage identification are also discussed.

/pdf/damage-identification-and-health-monitoring-of-structural-3s8adkph2w.pdf

Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: A literature review

In-service cables such as stay cables and suspenders of cable-stayed bridges and suspension bridges, are subjected to dynamic loads (e.g., the vehicle loads and wind excitation). Performing vibration measurements and subsequently identifying the dynamic properties and establishing a dynamic model of cable vibration are essential for their dynamic analysis, condition assessment, and performance prediction. For example, based on the taut-string theory, the cable tension, as a critical indicator of cable performance and health state, can be computed using its frequency that can be identified from the measured cable vibration responses. Traditional contact-type wired or wireless sensors, such as accelerometers and strain gauge sensors, require physically attaching to the structure for vibration measurements, which could induce the mass effect. In addition, installing these sensors on structures is costly, time-consuming, and allows instrumentations at a limited number of places. On the other hand, digital video cameras have emerged as a cost effective and agile non-contact vibration measurement method, offering high-resolution, simultaneous, measurements. Recently, digital video camera measurements processed by advanced computer vision and machine learning algorithms have been successfully used for experimental and operational full-field vibration measurement and modal analysis. This study develops a video measurement and processing based technique that can autonomously and blindly extract the full-field dynamic parameters of cable vibration from the video measurements. In addition, by exploiting the taut string theory, full-order (as many modes as possible) dynamic parameters are also extracted. Therefore, a full-field, full-order dynamic (modal) model of cable vibration is established. Laboratory experiments are conducted to validate the developed approach.

Establishment of Full-Field, Full-Order Dynamic Model of Cable Vibration by Video Motion Manipulations

This paper presents recent developments in an extremely compact, wireless impedance sensor node for use in structural health monitoring (SHM) The sensor node uses a low-cost integrated circuit that can measure and record the electric impedance of a piezoelectric active-sensor The sensor node also integrates several components, including a microcontroller for local computing, telemetry for wireless data transmission, multiplexers for managing up to seven piezoelectric transducers per node, energy storage mediums, and several triggering options including a wireless triggering circuit into one package to truly realize a comprehensive, self-contained wireless active-sensor node for SHM applications It is estimated that this sensor node requires less than 75 mW of total power to operate measurement, computation and data transmission In addition, the sensor node can also be used for the active-sensor self-diagnostic process that can monitor the operational condition of piezoelectric transducers used in SHM applications The performance of this miniaturized and portable device is compared to our previous results and its broader capabilities are demonstratedCopyright © 2008 by ASME

Recent Advances in Impedance-Based Wireless Sensor Nodes

Damage‐Sensitive Features

The concept of wireless sensor nodes and sensor networks has been widely investigated within the field of structural health monitoring (SHM). However, the ability to power sensors, onboard processing, and telemetry components is a significant challenge in many SHM applications. Several energy harvesting techniques, based on piezoelectric and thermoelectric materials, have been proposed and studied to solve such problems. This article summarizes recent advances and research issues in energy harvesting relevant to SHM applications. We also present a feasibility study for using radio frequency (RF) signals to wirelessly deliver energy to power sensing systems. The goal of this study is to provide a method for supplying power to a sensor node periodically so that the SHM system does not rely on an onboard power source.


Keywords:

structural health monitoring (SHM);
energy harvesting;
wireless energy transmission;
sensor networks;
sensor nodes

Energy Harvesting and Wireless Energy Transmission for SHM Sensor Nodes

The RAPTOR telescope systems are astronomical observatories that operate in remote locations in New Mexico searching for astrophysical transients called gamma-ray bursts. Their operating condition should remain at good levels in order to have accurate observations. Currently, the first component of the RAPTOR telescopes to fail is a capstan driving mechanism that operates in a run-to failure mode. The capstans wear relatively frequently because of their manufacturing material and can cause damage to other more expensive components, such as the drive wheels and the telescope optics. Monitoring the condition of these systems seems a reasonable solution since the unpredictable rate at which the capstans experience wear, in combination with the remote locations and high duty cycles of these telescope systems, make it unprofitable to choose a strategy of replacing the capstans at chosen intervals. Experimental tests of the telescope systems reported here recorded vibration signals during clockwise and counterclockwise rotations, similar to a motion known as "homing-sequence". The Empirical Mode Decomposition (EMD) method in combination with the Hilbert Transform (HT) and a new alternative method for the estimation of the instantaneous features of a signal that applies an energy tracking operator, called Teager-Kaiser Energy operator, and an energy separation algorithm to the data being analysed, are the time-frequency analysis methods used for analysis here.

Charles R. Farrar

Papers

Establishment of Full-Field, Full-Order Dynamic Model of Cable Vibration by Video Motion Manipulations

Recent Advances in Impedance-Based Wireless Sensor Nodes

Damage‐Sensitive Features

Energy Harvesting and Wireless Energy Transmission for SHM Sensor Nodes

Damage Detection in RAPTOR Telescope Systems Using Time-Frequency Analysis Methods