Abstract: This study proposes a hybrid minimal repair policy
which combines periodic maintenance policy with age-based maintenance policy for a serial production system. Parameters of such policy are defined as and which indicate as hybrid minimal
repair time and planned preventive maintenance time
respectively . Under this hybrid policy, the system is
repaired minimally if it fails during , . A perfect repair is
conducted on the first failure after at any machines. At the same time, we take opportunity to advance the preventive maintenance of
other machines simultaneously. If the system is still operating
properly up to , then the preventive maintenance is carried out as its
predetermined schedule. For a given , we obtain the optimal value which minimizes the expected cost per time unit. Numerical
example is presented to illustrate the properties of the optimal solution.
Abstract: Recently, the issue of machine condition monitoring
and fault diagnosis as a part of maintenance system became global
due to the potential advantages to be gained from reduced
maintenance costs, improved productivity and increased machine
availability. The aim of this work is to investigate the effectiveness
of a new fault diagnosis method based on power spectral density
(PSD) of vibration signals in combination with decision trees and
fuzzy inference system (FIS). To this end, a series of studies was
conducted on an external gear hydraulic pump. After a test under
normal condition, a number of different machine defect conditions
were introduced for three working levels of pump speed (1000, 1500,
and 2000 rpm), corresponding to (i) Journal-bearing with inner face
wear (BIFW), (ii) Gear with tooth face wear (GTFW), and (iii)
Journal-bearing with inner face wear plus Gear with tooth face wear
(B&GW). The features of PSD values of vibration signal were
extracted using descriptive statistical parameters. J48 algorithm is
used as a feature selection procedure to select pertinent features from
data set. The output of J48 algorithm was employed to produce the
crisp if-then rule and membership function sets. The structure of FIS
classifier was then defined based on the crisp sets. In order to
evaluate the proposed PSD-J48-FIS model, the data sets obtained
from vibration signals of the pump were used. Results showed that
the total classification accuracy for 1000, 1500, and 2000 rpm
conditions were 96.42%, 100%, and 96.42% respectively. The results
indicate that the combined PSD-J48-FIS model has the potential for
fault diagnosis of hydraulic pumps.
Abstract: This paper presents the optimal design and development
of an axial flux motor for blood pump application. With the design
objective of maximizing the motor efficiency and torque, different
topologies of AFPM machine has been examined. Selection of
optimal magnet fraction, Halbach arrangement of rotor magnets and
the use of Soft Magnetic Composite (SMC) material for the stator
core results in a novel motor with improved efficiency and torque
profile. The results of the 3D Finite element analysis for the novel
motor have been shown.
Abstract: This paper presents a method for determining the
uniaxial tensile properties such as Young-s modulus, yield strength
and the flow behaviour of a material in a virtually non-destructive
manner. To achieve this, a new dumb-bell shaped miniature
specimen has been designed. This helps in avoiding the removal of
large size material samples from the in-service component for the
evaluation of current material properties. The proposed miniature
specimen has an advantage in finite element modelling with respect
to computational time and memory space. Test fixtures have been
developed to enable the tension tests on the miniature specimen in a
testing machine. The studies have been conducted in a chromium
(H11) steel and an aluminum alloy (AR66). The output from the
miniature test viz. load-elongation diagram is obtained and the finite
element simulation of the test is carried out using a 2D plane stress
analysis. The results are compared with the experimental results. It is
observed that the results from the finite element simulation
corroborate well with the miniature test results. The approach seems
to have potential to predict the mechanical properties of the
materials, which could be used in remaining life estimation of the
various in-service structures.
Abstract: Support vector machines (SVMs) have shown
superior performance compared to other machine learning techniques,
especially in classification problems. Yet one limitation of SVMs is
the lack of an explanation capability which is crucial in some
applications, e.g. in the medical and security domains. In this paper, a
novel approach for eclectic rule-extraction from support vector
machines is presented. This approach utilizes the knowledge acquired
by the SVM and represented in its support vectors as well as the
parameters associated with them. The approach includes three stages;
training, propositional rule-extraction and rule quality evaluation.
Results from four different experiments have demonstrated the value
of the approach for extracting comprehensible rules of high accuracy
and fidelity.
Abstract: In this paper we examine the use of global texture analysis based approaches for the purpose of Persian font recognition in machine-printed document images. Most existing methods for font recognition make use of local typographical features and connected component analysis. However derivation of such features is not an easy task. Gabor filters are appropriate tools for texture analysis and are motivated by human visual system. Here we consider document images as textures and use Gabor filter responses for identifying the fonts. The method is content independent and involves no local feature analysis. Two different classifiers Weighted Euclidean Distance and SVM are used for the purpose of classification. Experiments on seven different type faces and four font styles show average accuracy of 85% with WED and 82% with SVM classifier over typefaces
Abstract: Feature and model selection are in the center of
attention of many researches because of their impact on classifiers-
performance. Both selections are usually performed separately but
recent developments suggest using a combined GA-SVM approach to
perform them simultaneously. This approach improves the
performance of the classifier identifying the best subset of variables
and the optimal parameters- values. Although GA-SVM is an
effective method it is computationally expensive, thus a rough
method can be considered. The paper investigates a joined approach
of Genetic Algorithm and kernel matrix criteria to perform
simultaneously feature and model selection for SVM classification
problem. The purpose of this research is to improve the classification
performance of SVM through an efficient approach, the Kernel
Matrix Genetic Algorithm method (KMGA).
Abstract: The excessive consumption of fossil energies (electrical energy) during summer caused by the technological development involves more and more climate warming.
In order to reduce the worst impact of gas emissions produced from classical air conditioning, heat driven solar absorption chiller is pretty promising; it consists on using solar as motive energy which is clean and environmentally friendly to provide cold.
Solar absorption machine is composed by four components using Lithium Bromide /water as a refrigerating couple. LiBr- water is the most promising in chiller applications due to high safety, high volatility ratio, high affinity, high stability and its high latent heat. The lithium bromide solution is constitute by the salt lithium bromide which absorbs water under certain conditions of pressure and temperature however if the concentration of the solution is high in the absorption chillers; which exceed 70%, the solution will crystallize.
The main aim of this article is to study the phenomena of the crystallization and to evaluate how the dependence between the electric conductivity and the concentration which should be controlled.
Abstract: Purpose of this work is to develop an automatic classification system that could be useful for radiologists in the breast cancer investigation. The software has been designed in the framework of the MAGIC-5 collaboration. In an automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features based generally on morphological lesion differences. A study in the space features representation is made and some classifiers are tested to distinguish the pathological regions from the healthy ones. The results provided in terms of sensitivity and specificity will be presented through the ROC (Receiver Operating Characteristic) curves. In particular the best performances are obtained with the Neural Networks in comparison with the K-Nearest Neighbours and the Support Vector Machine: The Radial Basis Function supply the best results with 0.89 ± 0.01 of area under ROC curve but similar results are obtained with the Probabilistic Neural Network and a Multi Layer Perceptron.
Abstract: An electronic portal image device (EPID) has become
a method of patient-specific IMRT dose verification for radiotherapy.
Research studies have focused on pre and post-treatment verification,
however, there are currently no interventional procedures using EPID
dosimetry that measure the dose in real time as a mechanism to
ensure that overdoses do not occur and underdoses are detected as
soon as is practically possible. As a result, an EPID-based real time
dose verification system for dynamic IMRT was developed and was
implemented with MATLAB/Simulink. The EPID image acquisition
was set to continuous acquisition mode at 1.4 images per second. The
system defined the time constraint gap, or execution gap at the image
acquisition time, so that every calculation must be completed before
the next image capture is completed. In addition, the
Abstract: Text categorization techniques are widely used to many Information Retrieval (IR) applications. In this paper, we proposed a simple but efficient method that can automatically find the relationship between any pair of terms and documents, also an indexing matrix is established for text categorization. We call this method Indexing Matrix Categorization Machine (IMCM). Several experiments are conducted to show the efficiency and robust of our algorithm.
Abstract: This paper details few mechanical modeling and
design issues of RF MEMS switches. We concentrate on an
electrostatically actuated broad side series switch; surface
micromachined with a crab leg membrane. The same results are
extended to any complex structure. With available experimental data
and fabrication results, we present the variation in dynamic
performance and compliance of the switch with reference to few
design issues, which we find are critical in deciding the dynamic
behavior of the switch, without compromise on the RF
characteristics. The optimization of pull in voltage, transient time and
resonant frequency with regard to these critical design parameters are
also presented.
Abstract: Artificial Intelligence (AI) methods are increasingly being used for problem solving. This paper concerns using AI-type learning machines for power quality problem, which is a problem of general interest to power system to provide quality power to all appliances. Electrical power of good quality is essential for proper operation of electronic equipments such as computers and PLCs. Malfunction of such equipment may lead to loss of production or disruption of critical services resulting in huge financial and other losses. It is therefore necessary that critical loads be supplied with electricity of acceptable quality. Recognition of the presence of any disturbance and classifying any existing disturbance into a particular type is the first step in combating the problem. In this work two classes of AI methods for Power quality data mining are studied: Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs). We show that SVMs are superior to ANNs in two critical respects: SVMs train and run an order of magnitude faster; and SVMs give higher classification accuracy.
Abstract: The operational behavior of a six-phase squirrel cage
induction machine with faulted stator terminals is presented in this
paper. The study is carried out using the derived mathematical model
of the machine in the arbitrary reference frame. Tests are conducted
on a 1 kW experimental machine.
Steady-state and dynamic performance are analyzed for the
machine unloaded and loaded conditions. The results shows that with
one of the stator phases experiencing either an open- circuit or short
circuit fault the machine still produces starting torque, albeit the
running performance is significantly derated.
Abstract: Understanding road features such as lanes, the color
of lanes, and sidewalks in a live video captured from a moving
vehicle is essential to build video-based navigation systems. In this
paper, we present a novel idea to understand the road features using
support vector machines. Various feature vectors including color
components of road markings and the difference between two
regions, i.e., chosen AOIs, and so on are fed into SVM, deciding
colors of lanes and sidewalks robustly. Experimental results are
provided to show the robustness of the proposed idea.
Abstract: Helical milling operations are used to generate or
enlarge boreholes by means of a milling tool. The bore diameter can be
adjusted through the diameter of the helical path. The kinematics of
helical milling on a three axis machine tool is analysed firstly. The
relationships between processing parameters, cutting tool geometry
characters with machined hole feature are formulated. The feed motion
of the cutting tool has been decomposed to plane circular feed and
axial linear motion. In this paper, the time varying cutting forces acted
on the side cutting edges and end cutting edges of the flat end cylinder
miller is analysed using a discrete method separately. These two
components then are combined to produce the cutting force model
considering the complicated interaction between the cutters and
workpiece. The time varying cutting force model describes the
instantaneous cutting force during processing. This model could be
used to predict cutting force, calculate statics deflection of cutter and
workpiece, and also could be the foundation of dynamics model and
predicting chatter limitation of the helical milling operations.
Abstract: This paper presents a novel methodology for Maximum Power Point Tracking (MPPT) of a grid-connected 20 kW Photovoltaic (PV) system using neuro-fuzzy network. The proposed method predicts the reference PV voltage guarantying optimal power transfer between the PV generator and the main utility grid. The neuro-fuzzy network is composed of a fuzzy rule-based classifier and three Radial Basis Function Neural Networks (RBFNN). Inputs of the network (irradiance and temperature) are classified before they are fed into the appropriated RBFNN for either training or estimation process while the output is the reference voltage. The main advantage of the proposed methodology, comparing to a conventional single neural network-based approach, is the distinct generalization ability regarding to the nonlinear and dynamic behavior of a PV generator. In fact, the neuro-fuzzy network is a neural network based multi-model machine learning that defines a set of local models emulating the complex and non-linear behavior of a PV generator under a wide range of operating conditions. Simulation results under several rapid irradiance variations proved that the proposed MPPT method fulfilled the highest efficiency comparing to a conventional single neural network.
Abstract: A new generation of manufacturing machines
so-called MIMCA (modular and integrated machine control
architecture) capable of handling much increased complexity in
manufacturing control-systems is presented. Requirement for more
flexible and effective control systems for manufacturing machine
systems is investigated and dimensioned-which highlights a need for
improved means of coordinating and monitoring production
machinery and equipment used to- transport material. The MIMCA
supports simulation based on machine modeling, was conceived by
the authors to address the issues. Essentially MIMCA comprises an
organized unification of selected architectural frameworks and
modeling methods, which include: NISTRCS, UMC and Colored
Timed Petri nets (CTPN). The unification has been achieved; to
support the design and construction of hierarchical and distributed
machine control which realized the concurrent operation of reusable
and distributed machine control components; ability to handle
growing complexity; and support requirements for real- time control
systems. Thus MIMCA enables mapping between 'what a machine
should do' and 'how the machine does it' in a well-defined but
flexible way designed to facilitate reconfiguration of machine
systems.
Abstract: The contribution is dealing with the influence of high speed parameters on the quality of machined surface. In general the principle of high speed cutting lies in achieving faster machine times with concurrent increase in accuracy and quality of the machined areas in largely irregular, mathematically hard to define shapes. High speed machining is a highly effective method of machining with the following goals: increasing of machining productivity, increasing of quality of the machined surface, improving of machining economy, improving of ecological aspects of machining. This article is based on an experiment performed by the Department of Machining and Assembly of the Faculty of Mechanical Engineering of VŠBTechnical University of Ostrava.
Abstract: The paper addresses a problem of optimal staffing in
open shop environment. The problem is to determine the optimal
number of operators serving a given number of machines to fulfill the
number of independent operations while minimizing staff idle. Using
a Gantt chart presentation of the problem it is modeled as twodimensional
cutting stock problem. A mixed-integer programming
model is used to get minimal job processing time (makespan) for
fixed number of machines' operators. An algorithm for optimal openshop
staffing is developed based on iterative solving of the
formulated optimization task. The execution of the developed
algorithm provides optimal number of machines' operators in the
sense of minimum staff idle and optimal makespan for that number of
operators. The proposed algorithm is tested numerically for a real life
staffing problem. The testing results show the practical applicability
for similar open shop staffing problems.