Dissertationen zum Thema „Online diagnostický systém“

This master thesis deals with data processing from an online diagnostic system. The theoretical part of the work is divided into three chapters, the first deals with technical diagnostics as such, there are basic terminology and concepts associated with technical diagnostics, such as diagnostic methods, tools, models and signals. The second part is focused on diagnostic systems and their division. This chapter also provides some examples of online diagnostic monitoring systems from three different companies. The last chapter of the theoretical part is devoted to the description of the possibilities of data collection from diagnostic systems, ie. description of communication buses and protocols. The practical part of this work focuses on creating an application for data collection and visualization.

This master’s thesis deals with online vibration diagnostics of the milling machine spindle. It is aimed at the implementation of the complex online vibration diagnostics system to chosen milling machine, the suggestion of the vibration measurement system and the suggestion of the evaluation of changes of the technical condition of this milling machine spindle. The description of the vibration diagnostics, the description of the milling machine spindle construction and function and the suggestions for the practical application of vibration diagnostics are also parts of this thesis.

This thesis deals with vibration diagnostics and on-line monitoring systems of several producers. Furthermore, this thesis contains a description and evaluation of measuring the vibrations of the machine tool spindle when machining test samples with different cutting conditions and evaluating the surface quality of test samples.

This master’s thesis deals with online diagnostics of the machine tool and marginal analysis of surface texture. The first part of the thesis deals with problems theoretically, there are described the machine tools, maintenance, methods of technical diagnostics, specifically online vibrodiagnostic systems and an overview of profile and area parameters. In the practical part, two test samples created under ideal and degraded conditions. In the practical part, two signal processing processes under ideal and degraded conditions are compared using signal analysis from an online vibrodiagnostic system and surface texture analysis. Part of the work is the evaluation of the performed analyzes and recommendations for practice.

Cette thèse est consacrée à l'étude de diagnostic de pannes pour les systèmes pile à combustible de type PEMFC. Le but est d'améliorer la fiabilité et la durabilité de la membrane électrolyte polymère afin de promouvoir la commercialisation de la technologie des piles à combustible. Les approches explorées dans cette thèse sont celles du diagnostic guidé par les données. Les techniques basées sur la reconnaissance de forme sont les plus utilisées. Dans ce travail, les variables considérées sont les tensions des cellules. Les résultats établis dans le cadre de la thèse peuvent être regroupés en trois contributions principales.La première contribution est constituée d'une étude comparative. Plus précisément, plusieurs méthodes sont explorées puis comparées en vue de déterminer une stratégie précise et offrant un coût de calcul optimal.La deuxième contribution concerne le diagnostic online sans connaissance complète des défauts au préalable. Il s'agit d'une technique adaptative qui permet d'appréhender l'apparition de nouveaux types de défauts. Cette technique est fondée sur la méthodologie SSM-SVM et les règles de détection et de localisation ont été améliorées pour répondre au problème du diagnostic en temps réel.La troisième contribution est obtenue à partir méthodologie fondée sur l'utilisation partielle de modèles dynamiques. Le principe de détection et localisation de défauts est fondé sur des techniques d'identification et sur la génération de résidus directement à partir des données d'exploitation.Toutes les stratégies proposées dans le cadre de la thèse ont été testées à travers des données expérimentales et validées sur un système embarqué
Aiming at improving the reliability and durability of Polymer Electrolyte Membrane Fuel Cell (PEMFC) systems and promote the commercialization of fuel cell technologies, this thesis work is dedicated to the fault diagnosis study for PEMFC systems. Data-driven fault diagnosis is the main focus in this thesis. As a main branch of data-driven fault diagnosis, the methods based on pattern classification techniques are firstly studied. Taking individual fuel cell voltages as original diagnosis variables, several representative methodologies are investigated and compared from the perspective of online implementation.Specific to the defects of conventional classification based diagnosis methods, a novel diagnosis strategy is proposed. A new classifier named Sphere-Shaped Multi-class Support Vector Machine (SSM-SVM) and modified diagnostic rules are utilized to realize the novel fault recognition. While an incremental learning method is extended to achieve the online adaptation.Apart from the classification based diagnosis approach, a so-called partial model-based data-driven approach is introduced to handle PEMFC diagnosis in dynamic processes. With the aid of a subspace identification method (SIM), the model-based residual generation is designed directly from the normal and dynamic operating data. Then, fault detection and isolation are further realized by evaluating the generated residuals.The proposed diagnosis strategies have been verified using the experimental data which cover a set of representative faults and different PEMFC stacks. The preliminary online implementation results with an embedded system are also supplied

The thesis deals with the use of tribodiagnostics in predictive maintenance in corporate practice. It is generally dealt with maintenance, then tribodiagnostics in the company Škoda Auto a.s. Used offline and online diagnostic tools are described. Emphasis is placed on verifying the functionality of the newly purchased online filter unit. The experimental part is focused on detailed analysis of the above, including comparison of measurement results offline and online diagnostics. There is also an economic evaluation of savings obtained by operation the online filter unit. In conclusion, the tribodiagnostics recommendations are given for Škoda Auto a.s.

Cette thèse contribue au développement des recherches dans le domaine du Pronostic et Health Management : gestion de l’état de santé des systèmes complexes. Dans un contexte de management opérationnel et de sûreté de fonctionnement des systèmes, nous proposons d’étudier comment la modélisation par un Modèle Graphique Probabiliste Dynamique (MGPD) permet le diagnostic de l’état de santé courant d’un système, le pronostic de cet état et de l’évolution des dégradations, ainsi que l’estimation de sa durée de vie résiduelle en fonction de ses conditions opérationnelles. La dégradation des composants est en général inconnue et nécessite un arrêt du système pour être observée. Cependant, cela est difficile, voire impossible, durant l’exploitation du système. Néanmoins, un ensemble de grandeurs observables sur le système ou le composant peut caractériser le niveau de dégradation et faciliter l’estimation de la durée de vie résiduelle du composant et du système. Les MGPD offrent une approche adaptée à la modélisation de l’évolution de l’état de santé des systèmes et des composants. Nous étendons la modélisation classique des modèles de la famille des HMM vers les IOHMM pour permettre une propagation temporelle de l’incertitude afin de résoudre le problème de pronostic de l’état de santé et de l’estimation de la durée de vie résiduelle. Cette recherche comprend l’extension des algorithmes d’apprentissage et d’inférence appliqués aussi bien dans le cas d’un composant que pour un système structuré. Cette thèse a pour but de contribuer à lever les verrous scientifiques suivants : - Considérer l'état de santé du système par un modèle stochastique et apprendre les paramètres du modèle à partir des mesures disponibles sur le système. - Établir un diagnostic de l’état de santé du système et le pronostic de son évolution en intégrant plusieurs conditions opérationnelles. - Estimer la durée de vie résiduelle des composants et des systèmes structurés (série, parallèle) à partir de ses composants. L’enjeu est majeur, car le pronostic de la dégradation des composants du système permet de définir des stratégies soit de pilotage soit de maintenance par rapport à la durée de vie résiduelle du système. Cela permet la réduction de la probabilité d’occurrence d’un arrêt pour cause de dysfonctionnement du système, soit en ajustant la vitesse de dégradation pour s’accorder à un plan de maintenance préventif, soit en planifiant les interventions de maintenance de manière proactive
This thesis contributes to prognosis and health management for assessing health condition of complex systems. In the context of operational management and operational safety of systems, we propose to investigate how Dynamic Probabilistic Graphical Modelling (DPGM) can be used to diagnose the current health state of systems, prognostic the future health state, and the evolution of degradation, as well as estimate its remaining useful life based on its operating conditions. System degradation is generally unknown and requires shutting down the system to be observed. However, this is difficult or even impossible during system operation. Though, a set of observable quantities on a system or component can characterise the level of degradation and help to estimate the remaining useful life of components and systems. The DPGM provides an approach suitable for modelling the evolution of the health state of systems and components. The aim of this thesis is to transpose and capitalize on the experience of these previous works in a prognostic context on the basis of a more efficient DPGM taking into account the available knowledge on the system. We extend the classical HMM family models to the IOHMM to allow the time propagation of uncertainty to address prognostic problems. This research includes the extension of learning and inference algorithms. Variants of the HMM model are proposed to incorporate the operating environment into the prognosis. The aim of this thesis is to contribute to solving the following scientific locks: - Considering the state of health whatever the complexity of the system by a stochastic model and learning the model parameters from the available measurements on the system. - Establish a diagnosis of the state of health of the system and the prognosis of its evolution by integrating several operational conditions. - Estimate the remaining useful life of components and structured systems with series and parallel components. This is a major challenge because the prognosis of the degradation of system components makes it possible to define strategies for either control or maintenance in relation to the residual life of the system. This allows the reduction of the probability of occurrence of a shutdown due to a system malfunction either by adjusting the degradation speed to fit in with a preventive maintenance plan or by proactively planning maintenance interventions

碩士
國立臺灣師範大學
工業教育學系
104
In the Internet era IT and internet technology use in teaching and learning change the learning process. There are many systems on the market but most of the systems only focus on examines and teachers receive the result. In order to assist teachers’ teaching and enforcement students’ learning this system establish a platform allows teachers and students communication in the system. The system divided into three jurisdiction patterns the students carrier, the teacher carries and the management carries also included wired evaluates, the question bank edit, the examination paper edit, the test question edition, the test question diagnostic analysis, result processing, the result diagnostic analysis, the prompt back coupling, the class and grade account number management, the system administration as well as the teaching discussion area function, then in enhancement study efficiency. The findings discovered that, users after on-line evaluation diagnosis system practice analysis, the student’s learning result shows obvious progression, also by master learning, repeat practicing and discussion via this system students gain more knowledge to enforcement their learning result.

碩士
臺中技術學院
資訊科技與應用研究所
97
Today, humans are engrossed with the quality of life. Aesthetics beauty and color are gradually becoming an integral part of life. Anything beauty requires color presentation and design; and it is human nature to be preoccupied with outward beauty. However, there are individuals who do not have the abilities to match the right color scheme to suit one’s personal quality and have to rely on beauty consultation centers. The need to continuously seek consultations on suitable cosmetic colors and dressing are expensive and time consuming. To the average worker it is not simply expensive but also inconvenient. The gene color (G-Color) is a concept based on genetic characteristics which was used by the modern beauticians to advise clients on the proper color scheme for cosmetic applications and dressings. The advice was made based on the individual client’s personal traits and facial features. In this research, a G-Color MSN robot was developed to make online color gene diagnostics. The system is based on the MSN instant messaging. Members signed on to the G-Color MSN robot to input personal traits and facial features which was in “question-and-answer” mode where a diagnostic was made on the suitable color scheme for the individual. The G-Color MSN robot is easy to use and users may use the robot anytime-anywhere to get diagnostics without having to pay high consultant fees. Experimental testing is conducted in two parts. The first part compared diagnostic results from an expert and G-Color MSN robot. The comparisons showed accuracy to be 80.95%. In the second part, comparisons between SVM and the G-Color MSN robot showed an accuracy of 80%. Experimental results showed that the G-Color MSN robot can be used for G-Color consultation. It also shortens the time required for consultation and is free. Furthermore, the G-Color MSN robot may be used in beauty salons by beauticians to recommend the suitable color schemes for an individual.

碩士
逢甲大學
自動控制工程所
100
During the last several decades, there has been a dramatic increase in the number of publications on computer-assisted diagnostic system of traditional Chinese medicine (TCM). The relevant equipments and diagnostic systems in recent years had increased noticeably. However, the data could not communicate between different systems, due to the difference in the data format. Therefore, the aim of this research is to develop an interactive online diagnostic platform of traditional Chinese medicine with a function of multimedia data mining system. This model provided the fundamentals of data format, database, online platform, interaction, and data mining functions. We have developed the web-based interactive online diagnostic platform for TCM, and the platform provides the user can share the data in a standard format. To conclude, through this model of interactive and data exchange for multimedia data of TCM, the system can be helpful to provide the fundamental service of data management, data exchange, and data mining.

碩士
國立高雄第一科技大學
電子工程研究所
100
In this thesis, a smart integrated circuit (IC) and system is proposed for renewable energy harvester designs. In this design, an 8-bit microcontroller chip is used to sense and transform the voltages and currents of a solar panel, a wind turbine, and a battery into digital output values. These output values are calculated in the microcontroller chip to build the maximum power point tracking and charging algorithm. A smart charger IC is then designed to follow the microcontroller’s output command to control the solar panel or the wind turbine at the maximum power point, and also to charge the battery at an effective mode with the proposed power converter. The proposed IC is then combined with the microcontroller chip to implement the renewable energy harvester. The operating information of the solar panel, wind turbine, energy harvester, and the battery will be transferred to a notebook computer through the RS232 interface of the microcontroller chip, so that the proposed design and management tool can judge the system''s operating conditions and make appropriate maintenance decisions. Two chips are finished in this thesis. All of the chips had been taped out through the design flow of the Chip Implementation Center (CIC). Then, the chips are fabricated with TSMC 0.35μm 2P4M 3.3V/5V mixed-signal process. These chips are smart charge pulse-width modulator (SCPWM) IC, P-V charge pulse-width modulator (PCPWM) IC, respectively. The chip area of SCPWM is about 0.98*0.98 mm2. The chip area of PCPWM is about 0.98*0.98 mm2.

碩士
臺中健康暨管理學院
資訊工程學系碩士班
93
The study of math emphasizes the understanding of all the concepts in their entirety and their cross-application. In teaching math, it’s also perceived that math concepts are strongly connected: if a student has any misconception regarding any of the concepts, it will affect his math learning in the future; when lots of misconceptions are formed, it would be difficult to identify his learning problems, which may result in severe learning barriers. This study focuses on the curriculum taught in a middle school, using the solution to a system of linear equations as an example. It develops a computerized adaptive test based on students’ knowledge structure, which takes less time than the traditional ones in helping a student discover his weakness in math learning by identifying the concepts that he is confused with. With the prevalence of computers, integrating computers into classroom teaching becomes an important topic in education. With the online-diagnosing system, students can take the test whenever they like. The real-time result allows the student to immediately grasp his own learning situation, and the instant feedback allows the teacher to understand the student’s situation more easily too. This can be taken as a reference to improve a teacher’s teaching skills by the adjustment of his teaching pace and style in order to provide the remedial education.

碩士
國立臺灣師範大學
工業教育學系
103
This thesis is using Knowledge Map as the fundamental to construct an On-Line examination system of the Basic Electricity for the students who study Electrical Engineering in the vocational high schools. The Aptana Studio 3 was taken as the web integration development environment to adopt HTML、JavaScript、JQuery、MySQL、PHP and CSS etc. programs to develop this On-Line examination system. Examination and assessment both are the im-portant steps of learning for the students. In the past, their learning effect was assessed by the scores from examination only. However the teachers could not give any suggestion to improve their learning effect based on those scores. Therefore, this system will provide Knowledge Map of the Basic Electricity as the learning reference for the students. Through it, the students will understand the basic knowledge of Electrical Engineering much easily. Also they will understand what kind of prior knowledge they should equip before learning a new technology. The feature of this system is not only it can calculate the students’ scores immediately after examination, but also it will diagnose, assess and analogize the learning effect from students’ answers to suggest them to improve the prior knowledge they are lacking to increase their learning ability efficiently.

碩士
亞洲大學
資訊工程學系碩士班
95
This research used 7 indicators of mathematical field of grade 4 and 5 in elementary school as an example to study if Bayesian networks combining knowledge structure can effectively diagnose the students’ mistaken types. It also developed a set of on-line diagnostic test based on mathematical indicator of Grade 1-9 Curriculum and remedial instruction system to allow the students to immediately find their test results through on-line diagnostic test and have on-line remedial instruction according to diagnostic report. The research expected to reach the effects of evaluation, diagnosis and remedial instruction. The research findings are below: 1. Bayesian networks reveal good results when diagnosing fraction-related indicators and they can lead to satisfying identification rate; Bayesian networks adding knowledge structure have better identification rate than the model of Bayesian networks. 2. The predicted scores of computerized adaptive diagnostic test are higher than those of complete answers. However, the former reveals certain predication rate which can save not only the tests, but also the test time. The test can also immediately diagnose the students’ mistaken types and sub-skills. 3. After computerize adaptive remedial instruction the students’ performance improves significantly which shows the significant effect of computerize adaptive remedial instruction.

碩士
國立臺中教育大學
教育測驗統計研究所
100
The constructed-response items can collect the students' problem solving, including all actions while problem solving, so it is suitable to analyze the error types. In the " exponential laws and scientific notations " unit of math subject, students have many misconceptions. If the questions are "multiple-choice types", students can solve by guessing, teachers can not know what error types the students have. This research is base on constructed-response items to build a " exponential laws and scientific notations " diagnostic test system. With practicing, record the problem solving processes to analyze the students' error types. The study result gives: a. The diagnostic test system of this research can record the problem solving processes, analyze the students' error types, and score automatically. b. The reliability of added constructed-response items is better than the multiple-choice types (0.890 < 0.894). Therefore, in the case of the same test length, if the questions can be added a part of constructed-response items, the reliability will be higher. c. We can obtain more students' answering responses from the constructed-response items than the multiple choice questions. d. The development- “automatic analyzing constructed-response items model” of this thesis can identify the error type attaining to 100% accuracy, and the performance is excellent. On the identification of sub-skills, and error types, rhe Bayesian network which is combined the multiple choice questions with the solving responses of the constructed-response items is better than the traditional one.