This paper presents the condition monitoring and fault diagnosis of rolling element bearings using Support Vector Machines (SVM). The vibration response of healthy bearings and bearings with various component defects ...
详细信息
ISBN:
(纸本)9780791848982
This paper presents the condition monitoring and fault diagnosis of rolling element bearings using Support Vector Machines (SVM). The vibration response of healthy bearings and bearings with various component defects such as outer race, inner race, balls and their combination have been analyzed. From the obtained vibration spectrum, it is clearly seen that a discrete peak of excitation appeared for the specific defect of bearings. In this paper, various faults of the bearings has been simulated and classified. The process includes, data acquisition, feature extraction from time response and a knowledge based system to classify faults. Features defining feature vectors are formed using statistical techniques and are fed as input to the support vector machine (SVM) classifiers. Knowledge based system developed for classification can be used for automatic recognition of machinery faults based on feature vector.
The proceedings contain 48 papers. The topics discussed include: programming pervasive spaces;the operating system for the computer of the 21st century;extracting social and community intelligence from digital footpri...
ISBN:
(纸本)3642163548
The proceedings contain 48 papers. The topics discussed include: programming pervasive spaces;the operating system for the computer of the 21st century;extracting social and community intelligence from digital footprints: an emerging research area;smart itinerary recommendation based on user-generated GPS trajectories;inferring user search intention based on situation analysis of the physical world;ontology-enabled activity learning and model evolution in smart homes;support vector machines for inhabitant identification in smart houses;towards non-intrusive sleep patternrecognition in elder assistive environment;the making of a dataset for smart spaces;introduction to the business processes with ambient media - challenges for ubiquitous and pervasive systems;alerting accidents with ambiguity: a tangible tabletop application for safe and independent chemistry experiments;and dependency relation based detection of lexicalized user goals.
It is well known that Approximated Maximum Likelihood(AML) estimator has the best performance for short time sampling wideband source bearing estimation. But for a long time, the heavy computational load of maximizing...
详细信息
In this paper we introduce a new approach to automatic attribute and granularity selection for building optimum regression trees. The method is based on the minimum description length principle (MDL) and aspects of gr...
详细信息
Robot skills can be taught and recognized by a Programming-by-Demonstration technique where first a human operator demonstrates a set of reference skills. The operator's motions are then recorded by a data-capturi...
详细信息
ISBN:
(纸本)9781424469208
Robot skills can be taught and recognized by a Programming-by-Demonstration technique where first a human operator demonstrates a set of reference skills. The operator's motions are then recorded by a data-capturing system and modeled via fuzzy clustering and a Takagi-Sugeno modeling technique. The resulting skill models use the time as input and the operator's actions as outputs. During the recognition phase, the robot recognizes which skill has been used by the operator in a novel demonstration. This is done by comparison between the time clusters of the test skill and those of the reference skills. Finally, the robot executes the recognized skill by using the corresponding reference skill model. Drastic differences between learned and real world conditions which occur during the execution of skills by the robot are eliminated by using the Broyden update formula for Jacobians. This method was extended for fuzzy models especially for time cluster models. After the online training of a skill model the updated model is used for further executions of the same skill by the robot.
In this paper we are highlighting the signals that are not Fourier transformable and give its Fourier transform using PCA (Principle Component Analysis), LDA (Linear Discriminant Analysis).Such signals are step signal...
详细信息
Typical testors are a useful tool for both feature selection and for determining feature relevance in supervised classication problems. Nowadays, generating all typical testors of a training matrix is computationally ...
详细信息
ISBN:
(纸本)9783642167720
Typical testors are a useful tool for both feature selection and for determining feature relevance in supervised classication problems. Nowadays, generating all typical testors of a training matrix is computationally expensive;all reported algorithms have exponential complexity, depending mainly on the number of columns in the training matrix. For this reason, different approaches such as sequential and parallel algorithms, genetic algorithms and hardware implementations techniques have been developed. In this paper, we introduce a fast implementation of the algorithm CT_EXT (which is one of the fastest algorithms reported) based on an accumulative binary tuple, developed for generating all typical testors of a training matrix. The accumulative binary tuple implemented in the CT_EXT algorithm, is a useful way to simplifies the search of feature combinations which fulfill the testor property, because its implementation decreases the number of operations involved in the process of generating all typical testors. In addition, experimental results using the proposed fast implementation of the CT_EXT algorithm and the comparison with other state of the art algorithms that generated typical testors are presented.
Event matching technique against lots of multidimensional range predicates is important in the many applications. Existing techniques are restricted to the domain that all the predicates are specified by the user. How...
详细信息
ISBN:
(纸本)9781424472956
Event matching technique against lots of multidimensional range predicates is important in the many applications. Existing techniques are restricted to the domain that all the predicates are specified by the user. However, generally some predicates can be omitted by the application requirement. If some predicates are omitted, it is impossible to insert predicates to existing indexing scheme. One of alternatives can be naïve sequential matching that all the events are compared with all the predicates one by one and more advanced alternative is intersection of multi 1-d index. To solve this problem, we propose a novel index scheme for multidimensional range predicates that can be omitted by the application requirement.
Recently, manufacturing companies tried to increase competitiveness in their business collaboration with cooperative companies rather than within their own company. In Korea, more than 600 manufacturing companies are ...
详细信息
ISBN:
(纸本)9781424472956
Recently, manufacturing companies tried to increase competitiveness in their business collaboration with cooperative companies rather than within their own company. In Korea, more than 600 manufacturing companies are using web-based collaboration systems developed by the government-led project, referred to as i-Manufacturing. The project is being conducted to facilitate on-line collaboration among manufacturing companies since 2004. There are lots of functions and services supported by each collaboration system. In order to re-apply a developed system to other companies, however, we have to modify, upgrade, or newly develop some parts of the system, which is called as customization. During customization processes, we reconfigure functions or services of the system to satisfy user requirements. To facilitate reconfiguration of collaboration systems, therefore, we first define user patterns. Then we introduce an analysis technique to investigate and analyze patterns by using data mining. The analysis technique verifies normal or abnormal patterns(e.g., drastic increase in using a specific function or service), and automatically make the system recognize an abnormal pattern as a new normal pattern when the abnormal pattern continue for a long time. Including the analysis technique, we also suggest a reconfiguration guideline for reorganizing function modules into services in a specific collaboration system.
The proceedings contain 38 papers. The topics discussed include: context aware computing and its utilization in event-based systems;logic-based representation, reasoning and machine learning for event recognition;an e...
ISBN:
(纸本)9781605589275
The proceedings contain 38 papers. The topics discussed include: context aware computing and its utilization in event-based systems;logic-based representation, reasoning and machine learning for event recognition;an event view model and DSL for engineering an event-based SOA monitoring infrastructure;an RFID architecture based on an event-oriented component model;content-based rendezvous with upgraph combination;event semantics in event dissemination architectures for massive multiuser virtual environments;complex event processing synergies with predictive analytics;event processing for large-scale distributed games;reliable fault-tolerant sensors for distributed systems;business-oriented development methodology for complex event processing: demonstration of an integrated approach for process monitoring;placement of replicated tasks for distributed stream processing systems;and an approach for iterative event pattern recommendation.
暂无评论