Abstract: The Rolling Element Bearing (REB) vibration diagnosis is worth of special interest by the variety of REB and the wide necessity of those elements in industrial applications. The presence of a localized fault in a REB gives rise to a vibrational response, characterized by the modulation of a carrier signal. Frequency content of carrier signal (Spectral Frequency –f) is mainly related to resonance frequencies of the REB. This carrier signal is modulated by another signal, governed by the periodicity of the fault impact (Cyclic Frequency –α). In this sense, REB fault vibration response gives rise to a second-order cyclostationary signal. Second order cyclostationary signals could be represented in a bi-spectral map, where Spectral Coherence –SCoh are plotted against f and α. The Improved Envelope Spectrum –IES, is a useful approach to execute REB fault diagnosis. IES could be applied by the integration of SCoh over a predefined bandwidth on the f axis. Approaches to select f-bandwidth have been recently exposed by the definition of a metric which intends to evaluate the magnitude of the IES at the fault characteristics frequencies. This metric is represented in a 1/3-binary tree as a function of the frequency bandwidth and centre. Based on this binary tree the optimal frequency band is selected. However, some advantages have been seen if the metric is changed, which in fact tends to dictate different optimal f-bandwidth and so improve the IES representation. This paper evaluates the behaviour of the IES from a different metric optimization. This metric is based on the sample correlation coefficient, detecting high peaks in the selected frequencies while penalizing high peaks in the neighbours of the selected frequencies. Prior results indicate an improvement on the signal-noise ratio (SNR) on around 86% of samples analysed, which belong to IMS database.
Abstract: Building system is highly vulnerable to different kinds
of faults and human misbehaviors. Energy efficiency and user comfort
are directly targeted due to abnormalities in building operation. The
available fault diagnosis tools and methodologies particularly rely on
rules or pure model-based approaches. It is assumed that model or
rule-based test could be applied to any situation without taking into
account actual testing contexts. Contextual tests with validity domain
could reduce a lot of the design of detection tests. The main objective of this paper is to consider fault validity when
validate the test model considering the non-modeled events such
as occupancy, weather conditions, door and window openings and
the integration of the knowledge of the expert on the state of the
system. The concept of heterogeneous tests is combined with test
validity to generate fault diagnoses. A combination of rules, range
and model-based tests known as heterogeneous tests are proposed
to reduce the modeling complexity. Calculation of logical diagnoses
coming from artificial intelligence provides a global explanation
consistent with the test result. An application example shows the efficiency of the proposed
technique: an office setting at Grenoble Institute of Technology.
Abstract: Fault diagnosis of composite asynchronous sequential
machines with parallel composition is addressed in this paper. An
adversarial input can infiltrate one of two submachines comprising
the composite asynchronous machine, causing an unauthorized state
transition. The objective is to characterize the condition under
which the controller can diagnose any fault occurrence. Two control
configurations, state feedback and output feedback, are considered in
this paper. In the case of output feedback, the exact estimation of
the state is impossible since the current state is inaccessible and the
output feedback is given as the form of burst. A simple example is
provided to demonstrate the proposed methodology.
Abstract: Centrifugal pumps are complex machines which can experience different types of fault. Condition monitoring can be used in centrifugal pump fault detection through vibration analysis for mechanical and hydraulic forces. Vibration analysis methods have the potential to be combined with artificial intelligence systems where an automatic diagnostic method can be approached. An automatic fault diagnosis approach could be a good option to minimize human error and to provide a precise machine fault classification. This work aims to introduce an approach to centrifugal pump fault diagnosis based on artificial intelligence and genetic algorithm systems. An overview of the future works, research methodology and proposed experimental setup is presented and discussed. The expected results and outcomes based on the experimental work are illustrated.
Abstract: In the present paper, five electric current based methods to analyze electric faults on the stator of induction motors (IM) are used and compared. The analysis tries to extend the application of the multiple reference frames diagnosis technique. An eccentricity indicator is presented to improve the application of the Park’s Vector Approach technique. Most of the fault indicators are validated and some others revised, agree with the technical literatures and published results. A tri-phase 3hp squirrel cage IM, especially modified to establish different fault levels, is used for validation purposes.
Abstract: A strategy of fault diagnosis and tolerance for
asynchronous sequential machines is discussed in this paper. With
no synchronizing clock, it is difficult to diagnose an occurrence
of permanent or stuck-in faults in the operation of asynchronous
machines. In this paper, we present a fault detector comprised of
a timer and a set of static functions to determine the occurrence
of faults. In order to realize immediate fault tolerance, corrective
control theory is applied to designing a dynamic feedback controller.
Existence conditions for an appropriate controller and its construction
algorithm are presented in terms of reachability of the machine and
the feature of fault occurrences.
Abstract: The operational life of rotating machines has to be extended using a predictive condition maintenance tool. Among various condition monitoring techniques, vibration analysis is most widely used technique in industry. Signals are extracted for evaluating the condition of machine; further diagnostics is carried out with detected signals to extend the life of machine. With help of detected signals, further interpretations are done to predict the occurrence of defects. To study the problem of defects, a test rig with various possibilities of defects is constructed and experiments are performed considering the unbalanced condition. Further, this paper presents an approach for fault diagnosis of unbalance condition using Elman neural network and frequency-domain vibration analysis. Amplitudes with variation in acceleration are fed to Elman neural network to classify fault or no-fault condition. The Elman network is trained, validated and tested with experimental readings. Results illustrate the effectiveness of Elman network in rotor-bearing system.
Abstract: Connected vehicles are one of the promising technologies for future Intelligent Transportation Systems (ITS). A connected vehicle system is essentially a set of vehicles communicating through a network to exchange their information with each other and the infrastructure. Although this interconnection of the vehicles can be potentially beneficial in creating an efficient, sustainable, and green transportation system, a set of safety and reliability challenges come out with this technology. The first challenge arises from the information loss due to unreliable communication network which affects the control/management system of the individual vehicles and the overall system. Such scenario may lead to degraded or even unsafe operation which could be potentially catastrophic. Secondly, faulty sensors and actuators can affect the individual vehicle’s safe operation and in turn will create a potentially unsafe node in the vehicular network. Further, sending that faulty sensor information to other vehicles and failure in actuators may significantly affect the safe operation of the overall vehicular network. Therefore, it is of utmost importance to take these issues into consideration while designing the control/management algorithms of the individual vehicles as a part of connected vehicle system. In this paper, we consider a connected vehicle system under Co-operative Adaptive Cruise Control (CACC) and propose a fault diagnosis scheme that deals with these aforementioned challenges. Specifically, the conventional CACC algorithm is modified by adding a Kalman filter-based estimation algorithm to suppress the effect of lost information under unreliable network. Further, a sliding mode observer-based algorithm is used to improve the sensor reliability under faults. The effectiveness of the overall diagnostic scheme is verified via simulation studies.
Abstract: Fault diagnosis of Linear Parameter-Varying (LPV)
system using an adaptive Kalman filter is proposed. The LPV model
is comprised of scheduling parameters, and the emulator parameters.
The scheduling parameters are chosen such that they are capable of
tracking variations in the system model as a result of changes in the
operating regimes. The emulator parameters, on the other hand,
simulate variations in the subsystems during the identification phase
and have negligible effect during the operational phase. The nominal
model and the influence vectors, which are the gradient of the feature
vector respect to the emulator parameters, are identified off-line from
a number of emulator parameter perturbed experiments. A Kalman
filter is designed using the identified nominal model. As the system
varies, the Kalman filter model is adapted using the scheduling
variables. The residual is employed for fault diagnosis. The
proposed scheme is successfully evaluated on simulated system as
well as on a physical process control system.
Abstract: Energy consumption data, in particular those involving
public buildings, are impacted by many factors: the building structure,
climate/environmental parameters, construction, system operating
condition, and user behavior patterns. Traditional methods for data
analysis are insufficient. This paper delves into the data mining
technology to determine its application in the analysis of building
energy consumption data including energy consumption prediction,
fault diagnosis, and optimal operation. Recent literature are reviewed
and summarized, the problems faced by data mining technology in the
area of energy consumption data analysis are enumerated, and research
points for future studies are given.
Abstract: In this paper, a robust fault detection and isolation
(FDI) scheme is developed to monitor a multivariable nonlinear
chemical process called the Chylla-Haase polymerization reactor,
when it is under the cascade PI control. The scheme employs a radial
basis function neural network (RBFNN) in an independent mode to
model the process dynamics, and using the weighted sum-squared
prediction error as the residual. The Recursive Orthogonal Least
Squares algorithm (ROLS) is employed to train the model to
overcome the training difficulty of the independent mode of the
network. Then, another RBFNN is used as a fault classifier to isolate
faults from different features involved in the residual vector. Several
actuator and sensor faults are simulated in a nonlinear simulation of
the reactor in Simulink. The scheme is used to detect and isolate the
faults on-line. The simulation results show the effectiveness of the
scheme even the process is subjected to disturbances and
uncertainties including significant changes in the monomer feed rate,
fouling factor, impurity factor, ambient temperature, and
measurement noise. The simulation results are presented to illustrate
the effectiveness and robustness of the proposed method.
Abstract: Lots of motors have been being used in industry.
Therefore many researchers have studied about the failure diagnosis of
motors. In this paper, the effect of measuring environment for
diagnosis of gear fault connected to a motor shaft is studied. The fault
diagnosis is executed through the comparison of normal gear and
abnormal gear. The measured FFT data are compared with the normal
data and analyzed for q-axis current, noise and vibration. For bad and
good environment, the diagnosis results are compared. From these, it
is shown that the bad measuring environment may not be able to detect
exactly the motor gear fault. Therefore it is emphasized that the
measuring environment should be carefully prepared.
Abstract: This paper presents a novel integrated hybrid
approach for fault diagnosis (FD) of nonlinear systems. Unlike most
FD techniques, the proposed solution simultaneously accomplishes
fault detection, isolation, and identification (FDII) within a unified
diagnostic module. At the core of this solution is a bank of adaptive
neural parameter estimators (NPE) associated with a set of singleparameter
fault models. The NPEs continuously estimate unknown
fault parameters (FP) that are indicators of faults in the system. Two
NPE structures including series-parallel and parallel are developed
with their exclusive set of desirable attributes. The parallel scheme is
extremely robust to measurement noise and possesses a simpler, yet
more solid, fault isolation logic. On the contrary, the series-parallel
scheme displays short FD delays and is robust to closed-loop system
transients due to changes in control commands. Finally, a fault
tolerant observer (FTO) is designed to extend the capability of the
NPEs to systems with partial-state measurement.
Abstract: Operations, maintenance and reliability of wind
turbines have received much attention over the years due to the rapid
expansion of wind farms. This paper explores early fault diagnosis
technique for a 5MW wind turbine system subjected to multiple
faults, where genetic optimization algorithm is employed to make the
residual sensitive to the faults, but robust against disturbances. The
proposed technique has a potential to reduce the downtime mostly
caused by the breakdown of components and exploit the productivity
consistency by providing timely fault alarms. Simulation results show
the effectiveness of the robust fault detection methods used under
Matlab/Simulink/Gatool environment.
Abstract: ESPRIT-TLS method appears a good choice for high
resolution fault detection in induction machines. It has a very high
effectiveness in the frequency and amplitude identification.
Contrariwise, it presents a high computation complexity which
affects its implementation in real time fault diagnosis. To avoid this
problem, a Fast-ESPRIT algorithm that combined the IIR band-pass
filtering technique, the decimation technique and the original
ESPRIT-TLS method was employed to enhance extracting accurately
frequencies and their magnitudes from the wind stator current with
less computation cost. The proposed algorithm has been applied to
verify the wind turbine machine need in the implementation of an online,
fast, and proactive condition monitoring. This type of remote
and periodic maintenance provides an acceptable machine lifetime,
minimize its downtimes and maximize its productivity. The
developed technique has evaluated by computer simulations under
many fault scenarios. Study results prove the performance of Fast-
ESPRIT offering rapid and high resolution harmonics recognizing
with minimum computation time and less memory cost.
Abstract: In this work, the main problem considered is the
detection and the isolation of the actuator fault. A new formulation of
the linear system is generated to obtain the conditions of the actuator
fault diagnosis. The proposed method is based on the representation
of the actuator as a subsystem connected with the process system in
cascade manner. The designed formulation is generated to obtain the
conditions of the actuator fault detection and isolation. Detectability
conditions are expressed in terms of the invertibility notions. An
example and a comparative analysis with the classic formulation
illustrate the performances of such approach for simple actuator fault
diagnosis by using the linear model of nuclear reactor.
Abstract: In this paper, an automatic system of diagnosis was
developed to detect and locate in real time the defects of the wound
rotor asynchronous machine associated to electronic converter. For
this purpose, we have treated the signals of the measured parameters
(current and speed) to use them firstly, as indicating variables of the
machine defects under study and, secondly, as inputs to the Artificial
Neuron Network (ANN) for their classification in order to detect the
defect type in progress. Once a defect is detected, the interpretation
system of information will give the type of the defect and its place of
appearance.
Abstract: Occurrence of a multiple-points fault in machine operations could result in exhibiting complex fault signatures, which could result in lowering fault diagnosis accuracy. In this study, a multiple-points defect model (MPDM) is proposed which can simulate fault signature-s dynamics for n-points bearing faults. Furthermore, this study identifies that in case of multiple-points fault in the rotary machine, the location of the dominant component of defect frequency shifts depending upon the relative location of the fault points which could mislead the fault diagnostic model to inaccurate detections. Analytical and experimental results are presented to characterize and validate the variation in the dominant component of defect frequency. Based on envelop detection analysis, a modification is recommended in the existing fault diagnostic models to consider the multiples of defect frequency rather than only considering the frequency spectrum at the defect frequency in order to incorporate the impact of multiple points fault.
Abstract: This paper proposes method of diagnosing ball screw
preload loss through the Hilbert-Huang Transform (HHT) and
Multiscale entropy (MSE) process. The proposed method can
diagnose ball screw preload loss through vibration signals when the
machine tool is in operation. Maximum dynamic preload of 2 %, 4 %,
and 6 % ball screws were predesigned, manufactured, and tested
experimentally. Signal patterns are discussed and revealed using
Empirical Mode Decomposition(EMD)with the Hilbert Spectrum.
Different preload features are extracted and discriminated using HHT.
The irregularity development of a ball screw with preload loss is
determined and abstracted using MSE based on complexity
perception. Experiment results show that the proposed method can
predict the status of ball screw preload loss. Smart sensing for the
health of the ball screw is also possible based on a comparative
evaluation of MSE by the signal processing and pattern matching of
EMD/HHT. This diagnosis method realizes the purposes of prognostic
effectiveness on knowing the preload loss and utilizing convenience.
Abstract: The electric power industry is currently undergoing an unprecedented reform. One of the most exciting and potentially profitable recent developments is increasing usage of artificial intelligence techniques. The intention of this paper is to give an overview of using neural network (NN) techniques in power systems. According to the growth rate of NNs application in some power system subjects, this paper introduce a brief overview in fault diagnosis, security assessment, load forecasting, economic dispatch and harmonic analyzing. Advantages and disadvantages of using NNs in above mentioned subjects and the main challenges in these fields have been explained, too.